Program Book ICWT 2024

10th ICWT 2024, Batam, Indonesia
Program Book
ICWT 2024
10th
International Conference on
Wireless and Telematics
https://icwt-seei.org/2024/
July 04-05, 2024

Nagoya Hill
Batam, Indonesia
July 04-05, 2024, Nagoya Hill, Hill Batam i

Copyright @ 2024 by
The Institute of Electrical and Electronics Engineers, Inc.
All right reserved
10th International Conference Wireless and Telematics (ICWT) 2024
Copyright and Reprint Permission: Abstracting is permitted with credit to

the source. Libraries are permitted to photocopy beyond the limit of U.S.
copyright law for private use of patrons those articles in this volume that
carry a code at the bottom of the first page, provided the per-copy fee
indicated in the code is paid though Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923. For reprint or republication
permission, email to IEEE Copyrights Manager at pubs-
permissions@ieee.org.
All right reserved. Copyright ©2024 by IEEE.
Conference Record Number 62080
July 04-05, 2024, Nagoya Hill, Hill Batam ii

Table of Contents
Table of Contents ................................................................ iii

Message from the General Chair .......................................... 1
Organizers and Organizing Commitee ................................. 3
Technical Program ............................................................... 6
Keynote Speaker ................................................................ 29
Telematics, Networking, & Telematics Applications ........ 33
Wireless Technology, Mobile Communications, Radar
System ................................................................................ 57
Electrical, Computer, & Biomedical Engineering.............. 92
Poster Session ................................................................... 114
July 04-05, 2024, Nagoya Hill, Hill Batam iii

This page is intentionally left blank.
July 04-05, 2024, Nagoya Hill, Hill Batam iv

Message from the General Chair

It gives me great pleasure to
welcome you to the 10th
International Conference on
Wireless and Telematics
(ICWT) 2024. ICWT is one
of the conferences which are
organized by
Telecommunications
Engineering Research
Group, School of Electrical
Engineering and Informatics,
Institut Teknologi Bandung
(STEI–ITB). This year, the 10th ICWT 2024 is held in Batam,
Indonesia.
ICWT has been expected to become one of the important

conferences in Indonesia in the area of information and
communications technology. This conference provides a forum for
universities, industries, government, and public sectors to expose
and exchange their innovative ideas and methods.
Since 2015, this conference has brought together a tremendous and

rich diversity of authors and speakers to share ideas and new
perspectives on a wide range of research on information and
communication technologies topics. This year, we accepted 77
papers, therefore acceptance rate approaching 60%. Submissions
whose authors span not only from domestic universities but also
from our overseas colleagues, such as: Malaysia, India, Japan, and
China.
The 10th ICWT 2024 will not happen without the hard work of the
organizers behind the scenes. We had an excellent team that has
worked very hard to organize ICWT 2024. I would like to thank
Universitas Islam Negeri Sunan Gunung Djati as our co-host, the
steering committee; International advisory committee; Our
July 04-05, 2024, Nagoya Hill, Hill Batam 1

Sponsors; and particularly I want to thank all members of the

Technical Program Committee for their hard work in providing
thorough and insightful reviews on time. Special thanks also go to
all authors since ICWT 2024 would not be possible without the
contributions of the authors.
Finally, I wish all participants a successful and fruitful conference.

I hope you will find this program interesting, useful, and
stimulating.
Prof. Dr. Ir. Hendrawan, M.Sc.

General Chair

Organizers and Organizing Commitee
Organized by
Supported by
Committee
General Chair
 Hendrawan (ITB, Indonesia)
 Eki Ahmad Zaki Hamidi (UIN SGD, Indonesia)
Steering Committee
 Widyawardana Adiprawita (ITB, Indonesia)
 Adit Kurniawan (ITB, Indonesia)
 Muhammad Ali Ramdhani (UIN SGD Bandung)
 Ulfiah (UIN SGD Bandung)
 Hasniah Aliah (UIN SGD Bandung)
 Husnul Qodim (UIN SGD Bandung)
 Deni Miharja (UIN SGD Bandung)
International Advisory Committee

 Tutun Juhana (CAS, IEEE Indonesia Section)
 Noriaki Kamiyama (Ritsumeikan University, Japan)
 Andrew, Chen-Yeon CHU (Feng Chia University, Taiwan)
 Elhadj Dogheche (Univ. Polytechnique Hauts-de-France,
France)
 Alyani Ismail (Universiti Putra Malaysia, Malaysia)
 Andriyan Bayu Suksmono (ITB, Indonesia)
 Rifqy Hakimi (Essex University, UK)

 Muhammad Ammar Wibisono (Universiteit Twente, the

Netherlands)
 Harashta Tatimma Larasati (Pusan National University, South
Korea)
 Nur’ain Izzati Shuhaimi (UiTM, Malaysia)
 Tommi Hariyadi (KAIST, South Korea)
 Yasuo Musashi (Kumamoto University, Japan)
Technical Program Committee

Chair
 Iskandar (ITB, Indonesia)
Members
 Wervyan Shalannanda (CAS, IEEE Indonesia Section)
 Nanang Ismail (UIN SGD Bandung)
 Edi Mulyana (UIN SGD, Indonesia)
 Adam Faroqi (UIN SGD, Indonesia)
 Mufid Ridlo Effendi (UIN SGD, Indonesia)
 Nana Rachmana Syambas (ITB, Indonesia)
 Achmad Munir (ITB, Indonesia)
 Ian Josef Matheus Edward (ITB, Indonesia)
 Mohammad Ridwan Effendi (ITB, Indonesia)
 Agus Subekti (BRIN, Indonesia)
Treasure
 Irma Zakia (ITB, Indonesia)
Organizing Committee
 Rina Mardiati (UIN SGD, Indonesia)
 Adam Faroqi (UIN SGD, Indonesia)
 Mufid Ridlo Effendi (UIN SGD, Indonesia)
 Lia Kamelia (UIN SGD, Indonesia)
 Asep Puady (ITB, Indonesia)
 Ibni Inggrianti (ITB, Indonesia)
 Zenal Aripin (ITB, Indonesia)
 Fadjrianah (ITB, Indonesia)
 Galih Nugraha Nurkahfi (ITB, Indonesia)
 Daniel Wiyogo Dwiputro (ITB, Indonesia)


Technical Program


10th ICWT 2024

Conference Program – at a Glance
Thursday, 4 July 2024 (Day 1)

08.00 - 08.30 REGISTRATION
OPENING CEREMONY
Venue: Nagoya Hill, Batam
Prof. Dr. Ir. Hendrawan, M.Sc.
General Chair 10th ICWT 2024
WELCOME REMARKS
08.30 – 09.00
1. Dr. Tutun Juhana, ST. MT
Dean of School of Electrical Engineering and Informatics,
Bandung Institute of Technology
2. Dr. Hj. Hasniah Aliah, M.Si
Dean of Faculty of Science and Technology of UIN
KEYNOTE SPEAKER
09.00 – 09.45
Prof. Iyad Dayoub, BSC, MASC, PHD, P.
ENG.
Professor, Univ. Polytechnique Hauts-de-France, France
09.45 – 10.00 COFFEE BREAK
KEYNOTE SPEAKER
10.00 – 10.45 Dr. Thi-Thu-Huong Le
Pusan National University, South Korea
10.45 – 11.00 Photo Session & Prepare for Presentation
Parallel Sessions 1
Offline Room1, Offline Room2
11.0 0 - 12.00
Online Room-1, Online Room-2, Online Room-3, Online
Room-4, Poster
12.0 0 - 13.00 LUNCH BREAK
Parallel Sessions 2
Offline Room1, Offline Room2
13.0 0 - 15.00
Online Room-1, Online Room-2, Online Room-3, Online
Room-4
15.00 - 15.20 COFFEE BREAK
Parallel Sessions 3
15.20 - 16.40 Online Room-1, Online Room-2, Online Room-3, Online
Room-4
Friday, 5 July 2024 (Day 2)
08.00 - 08.20 REGISTRATION
08.20 – 11.00 ICWT Committee Meeting
END OF CONFERENCE


ICWT 2024
Parallel Sessions 1
Poster
Offline Room-1 Offline Room-2 (Paper ID
XXX)
Time
Track: Track:
(WIB)
Telematics, Electrical, Computer,
Networking, & & Biomedical
Telematics Application Engineering
Chair Ridha Muldina Negara Rohmat Tulloh
11.00–11.20 Paper ID 012 Paper ID 048
11.20–11.40 Paper ID 023 Paper ID 060
11.40–12.00 Paper ID 024 Paper ID 063
ICWT 2024 Parallel Sessions 1
Online Room-1 Online Room-2
Track: Track:
Time Electrical, Computer, & Telematics,
(WIB) 001
Biomedical Engineering Networking, &
002
Telematics
008
Application
018
Chair Jupriyadi Adi Sucipto
058
11.00–11.20 Paper ID 055 Paper ID 072 059
Paper ID 056 Paper ID 075 087
11.20–11.40 107
11.40–12.00 Paper ID 061 Paper ID 076
Online Room-3 Online Room-4
Track: Track:
Time Electrical, Computer, & Wireless Technology,
(WIB) Biomedical Engineering Mobile
Communication, &
Radar System
Chair Ratna Mayasari Sri Astuti
11.00–11.20 Paper ID 083 Paper ID 090
11.20–11.40 Paper ID 084 Paper ID 093
11.40–12.00 Paper ID 085 Paper ID 095
12.00-13.00 LUNCH


ICWT 2024
Parallel Sessions 2
Poster
Offline Room-1 Offline Room-2 (Paper ID
XXX)
Track: Track:
Time (WIB)
Telematics, Wireless
Networking, & Technology, Mobile
Telematics Communication, &
Application Radar System
Galih Nugraha
Chair Rina Mardiati
Nurkahfi
13.00–13.20 Paper ID 070 Paper ID 078
13.20–13.40 Paper ID 073 Paper ID 080
13.40–14.00 Paper ID 074 Paper ID 104
14.00–14.20 Paper ID 032 Paper ID 065
14.20-14.40 Paper ID 033 Paper ID 067 001
14.40-15.00 Paper ID 039 Paper ID 068 002
Thursday, 4 July 2024 (Day 1) 008
018
ICWT 2024 Parallel Sessions 2 058
Online Room-1 Online Room-2 059
Track: Track: 087
Electrical, Wireless Technology, 107
Time (WIB)
Computer, & Mobile
Biomedical Communication, &
Engineering Radar System
Rin Rin Eki Ahmad Zaki
Chair
Nurmalasari Hamidi
13.00–13.20 Paper ID 006 Paper ID 011
13.20–13.40 Paper ID 007 Paper ID 015
13.40–14.00 Paper ID 009 Paper ID 016
14.00–14.20 Paper ID 010 Paper ID 017
14.20-14.40 Paper ID 014 Paper ID 019
14.40-15.00 Paper ID 013 Paper ID 020


Poster
Online Room-3 Online Room-4 (Paper ID
XXX)
Track: Track:
Time (WIB)
Telematics, Wireless Technology,
Networking, & Mobile Communication, 001
Telematics & Radar System 002
Application 008
Chair Nike Sartika Lia Kamelia 018
13.00–13.20 Paper ID 021 Paper ID 043 058
13.20–13.40 Paper ID 025 Paper ID 045 059
087
13.40–14.00 Paper ID 027 Paper ID 046 107
14.00–14.20 Paper ID 031 Paper ID 049
14.20-14.40 Paper ID 034 Paper ID 052
14.40-15.00 Paper ID 042 Paper ID 054


Poster
Online Room-1 Online Room-2 (Paper ID
XXX)
Track: Track:
Time (WIB)
Electrical, Telematics,
Computer, & Networking, &
Biomedical Telematics Application
Engineering
Azwar
Chair Mudzakir Edi Mulyana
Ridwan
15.20–15.40 Paper ID 062 Paper ID 079
15.40–16.00 Paper ID 064 Paper ID 081
16.00–16.20 Paper ID 071 Paper ID 082 001
Paper ID 106 Paper ID 035 002
16.20–16.40 008
Thursday, 4 July 2024 (Day 1) 018
ICWT 2024 Parallel Sessions 3 058
059
Online Room-3 Online Room-4 087
Track: 107
Track:
Wireless
Time (WIB) Electrical,
Technology,
Computer, &
Mobile
Biomedical
Communication, &
Engineering
Radar System
Chair Saepul Uyun Nanang Ismail
15.20–15.40 Paper ID 086 Paper ID 096
15.40–16.00 Paper ID 088 Paper ID 103
16.00–16.20 Paper ID 089 Paper ID 105
16.20–16.40 Paper ID 053
END OF CONFERENCE

PARALLEL SESSIONS 1
Thursday, 4 July 2024, Offline Room-1: 11.00 – 12.00
Track: Telematics, Networking, & Telematics Application
Session Chair: Ridha Muldina Negara, MT.
Time Title / Author
Object Detection for the Visually Impaired: A
11.00–11.20 Systematic Literature Review
Paper ID 012 Ghatan Fauzi Nugraha1, Aradea1, Rianto1, Rina
Mardiati2
Semantic Segmentation of Firearms Detection
Using YOLOv8 Model
11.20 – 11.40
Fadli Alfa Rizki, Mohamad Irfan, Dian Sa’adillah
Paper ID 023
Maylawati, Gitarja Sandi, Muhammad Insan Al-Amin,
Muhammad Ali Ramdhani
Microstrip Array Antenna Using Proximity
Coupled and Dolph Chebyshev Distribution for
11.40 – 12.00 Wi-Fi 6E
Paper ID 024 Nurul Fahmi Arief Hakim1, Irfan Maulana1, Agus Heri
Setya Budi1, Azwar Mudzakkir Ridwan2, Yuyun
Yuningsih3, Devi Eryanti3
PARALLEL SESSIONS 1
Track: Electrical, Computer, & Biomedical Engineering
Session Chair: Rohmat Tulloh, MT.
Time Title / Author
Fisher-Yates Approach for Daily Qur'an Chatbot
11.00–11.20
Syahrul Safarudin Hasan1, Wildan Budiawan Zulfikar1,
Paper ID 048
Nur Lukman3, Agung Wahana2, Aldy Rialdy Atmadja1
Systematic Review on Interpretability in Computer
11.20 – 11.40 Aided Alcohol Use Disorder Diagnosis
Paper ID 060 Nur Zahrati Janah, Adhistya Erna Permanasari, Noor
Akhmad Setiawan
Application of Convolutional Neural Network
Algorithm for Analyzing Sentiments on the
11.40 – 12.00 Kampus Merdeka Policy
Paper ID 063 Mohamad Irfan1, Theo Vectra Riyadi1, Aldy Rialdy
Atmadja1, Rifqi Syamsul Fuadi1, Abdul Muin2

PARALLEL SESSIONS 1
Thursday, 4 July 2024, Online Room-1: 11.00 – 12.00
Session Chair: Jupriyadi, MT
Time Title / Author
Design and Build a Prediction System State of
Charge (SoC) Battery Lithium-ion for Battery
11.00–11.20
Based Management Internet of Things (IoT)
Paper ID 055
Muhammad Alwi Sya'ban1, Adam Faroqi1, Lia
Kamelia1, Devi Eryanti2
Design and Build of IoT-Based Dragon Fruit Plant
Maintenance System Using the Nodemcu ESP8266
11.20 – 11.40
Microcontroller
Paper ID 056
Abdul Haris1, Adam Faroqi1, Edi Mulyana1, Siti
Sanah2
Behavior in Financial Technology: Study of
11.40 – 12.00 Intention to Use of Mobile Payment
Paper ID 061 Leni Susanti1, Siti Hajar Mohamad2, Doni Purnama
Alamsyah1
PARALLEL SESSIONS 1
Session Chair: Adi Sucipto, MT.
Time Title / Author
ANTIS: Automatic and Anthropometric
Measurement and Weight IoT-Monitoring for
11.00–11.20 Enhanced Infant Nutrition Assessment Using Dual
Paper ID 072 Sensor and Fuzzy
Ahmad Hafidz Fajrian, Rin Rin Nurmalasari, Lia
Kamelia
U-Net Approach to Mandibular Image
11.20 – 11.40 Segmentation From Panoramic X-rays
Paper ID 075 Nada Aisy Fadhila1, Khoerun Nisa Syaja’ah1, Yudha
Satya Perkasa1, Rin Rin Nurmalasari2
Cervical Cancer Classification From Pap Smear
Using XCeption Model
11.40 – 12.00
Ima Siti Maesaroh1, Khoerun Nisa Syaja'ah1, Yudha
Paper ID 076
Satya Perkasa1, Sri Hadianti3, Dwiza Riana4, Rin Rin
Nurmalasari2

PARALLEL SESSIONS 1
Session Chair: Ratna Mayasari, MT.
Time Title / Author
Implementation of Smart Electric Meter for Solar
11.00–11.20 System Based on Internet of Things
Paper ID 083 Rizky Mahesa Ramadhan, Lia Kamelia, Rin Rin
Nurmalasari, Rangga Julfian Hakim, Rizky Nurhadhi
Indonesian Sign Language Translation System
Using ResNet-50 Architecture-Based
11.20 – 11.40
Convolutional Neural Network
Paper ID 084
Zalfa Humaira Salsabila, Rin Rin Nurmalasari, Lia
Kamelia
A Comparative Analysis of Machine Learning
11.40 – 12.00 Algorithms: K-Nearest Neighbor (KNN) and
Paper ID 085 Random Forest for Predicting Sports Poses
Nur Syifa Maulida, Lia Kamelia, Rin Rin Nurmalasari
PARALLEL SESSIONS 1
Track: Wireless Technology, Mobile Communication, &
Radar System
Session Chair: Sri Astuti, MT.
Time Title / Author
Design of 13.56 MHz Transmitter Coil using
Miniaturization Technique with Planar Inverted F
11.00–11.20 Structure and Defected Ground Structure
Paper ID 090 Hafidz Fadilah1, Azwar Mudzakkir Ridwan1, Aan Eko
Setiawan1, Ahmad Saepul Milah1, Zenal Aripin3, M.
Idham Kurniawan2
Monitoring Harvest Power System Using
Piezoelectric Disk Electric-Based IoT
11.20 – 11.40
Elisa Apriliyani1, Enjang A Juanda2, Azwar Mudzakkir
Paper ID 093
Ridwan1, Aan Eko Setiawan1, Nurul Fahmi Arief
Hakim2, Raihan Nurhakim2
Tomato Harvesting Robot Prototype: Fuzzy-
Controlled Arm with Vision-Based Tomato
11.40 – 12.00
Detection
Paper ID 095
Hardiansyah Firdaus, Rina Mardiati, Aan Eko
Setiawan

PARALLEL SESSIONS 2
Session Chair: Galih Nugraha Nurkahfi, MT.
Time Title / Author
SEMU5G: an Emulator for Improving Model
Accuracy and Performance Evaluation of 5G
Network Core Architectures based on SDN
13.00 – 13.20
Ridha Muldina Negara1,2, Arga Valen Lazuardi2,
Paper ID 070
Muhammad Litfan Rahmansyah2, Salman Alfarizi
Novel Bajri2, Rohmat Tulloh3, Syaiful Ahdan6, Eki
Ahmad Zaki Hamidi4, Jupriyadi5,6
NDN Collaborative Caching Replacement and
Placement Policy Performance Evaluation
13.20 – 13.40 Ridha Muldina Negara1,2, Khibran Muhammad
Paper ID 073 Akbar2, Eki Ahmad Zaki Hamidi3, Ivah Puspita
Payangan2, Mutiara Dewi Puspitaningsih2, Adzka
Zahratu Putri As'ari2, Jupriyadi4,5
Mini Prototype of the Futuristic Bin with an
Automatic Waste Sortation System for
Managing the Garbage Problems in Society
13.40 – 14.00
Muhammad Taufik Dwi Putra1, Anugrah
Paper ID 074
Adiwilaga1, Jezzy Putra Munggaran1, Muhammad
Azka Adhitama1, Raihan Anwar As'ad1, Ahmad Ali
Alhafidz1, Rin Rin Nurmalasari3, Agus Juhana2
Automatic Water Cooling System to Increase
The Performance of PV
14.00 – 14.20
Paper ID 032 Riyani Prima Dewi1, Nike Sartika2, Novita Asma
Ilahi1, Saepul Rahmat1, Afrizal Abdi Musyafiq1,
Erna Alimudin1
The Impact of Collusive Interest Flooding
14.20 – 14.40 Attack on NDN Network Performance
Paper ID 033 Jupriyadi1,3, Ridha Muldina Negara1,2, Eueung
Mulyana1, Nana Rachmana Syambas1
Implementation of Wi-Fi 5 Ghz
Communication Using UDP Unicast Protocol
14.40 – 15.00
On Wheeled Soccer Robots
Paper ID 039
Arif Sumardiono1, Erna Alimudin1, Hendi Purnata1,
Riyani Prima Dewi1, Saepul Rahmat1, Nike Sartika2

PARALLEL SESSIONS 2
Track: Wireless Technology, Mobile Communication, & Radar
System
Session Chair: Dr. Rina Mardiati, MT.
Time Title / Author
Classification of Cassia Cinnamon and Ceylon
13.00–13.20 Cinnamon using Deep Learning
Paper ID 078 Agus Pratondo1, Nanang Ismail2, Brian Nayaka
Prasetya1, Muhammad Fatah Al Khadami1
PIT Performance Measurement using NFD
Pipeline Parameters in Named Data Networking
13.20 – 13.40
(NDN)
Paper ID 080
Adi Sucipto1,2, Syaiful Ahdan1,2, Iskandar1, Nana
Rachmana Syambas1
LoRa Based Underwater Wireless Network
13.40 – 14.00
System
Paper ID 104
Naufal Zaidan Nabhan, Joko Suryana
Convolutional Neural Network with VGG-16
Architecture for Object Image Classification in
14.00 – 14.20 Arabic Language
Paper ID 065 Iqbal Putra Ramadhan, Dian Sa'adillah Maylawati,
Diena Rauda Ramdania, Agung Wahana, Cepy
Slamet, Rifqi Syamsul Fuadi
Microstrip Antenna with Mushroom-like EBG
14.20 – 14.40 for Gain Enhancement at 2.3 GHz Frequency
Paper ID 067 Tubagus Arya Wirangga1, Ahmad Saepul Milah1,
Nanang Ismail1, Harfan Hian Ryanu2
The Effect of Uniplanar-Type EBG on
Microstrip Antenna Gain at the 2.3 GHz
14.40 – 15.00
Frequency
Paper ID 068
Reza Anggi Rizkia1, Ahmad Saepul Milah1, Nanang
Ismail1, Harfan Hian Ryanu2

PARALLEL SESSIONS 2

Session Chair: Rin Rin Nurmalasari, MT.
Time Title / Author
Dynamic Adaptive Video Streaming: A Comparative
Examination of NDN and IP Architectures in Real-
world Scenarios
13.00 – 13.20
Ega Helmi Mubarok, Komang Maesa Dithanugraha,
Paper ID 006 Hasan Nur Arifin, Muhammad Alief Fauzan Bariadi,
Aditya Arya Putranda, Nana Rahmana Syambas, Intan
Putri Maharani Prasetyaningrum, Fikry Adam
Performance Analysis of Forwarding Strategy in
NDN-Based Vehicle Networks
13.20 – 13.40 Zakariyya Gambetta Muhammad K1, Bagus Isa Ahmad1,
Paper ID 007 Felisha Dita Zafirah1, Nindya Dwi Larasati1, Naura Tiara
Riesani1, Fildza Haniifa Butsaina1, Syaiful Ahdan2, Eki
Ahmad Zaki Hamidi2, Nana Rachmana Syambas2
Leveraging The Existing Sensor Signals on
Komatsu D31P Bulldozer Through IoT-
13.40 – 14.00 Enabled Embedded System for Heavy
Paper ID 009 Equipment Unit
Aditya Kurniawan1, Kholilatul Wardani2, Eki
Ahmad Zaki Hamidi3
Deep Imitation Learning Behavior Fusion
Algorithm for Visual Based Autonomous
14.00 – 14.20
Driving Scenario
Paper ID 010
Handoko Supeno, Bambang Riyanto Trilaksono,
Widyawardana Adiprawita, Nur Ulfa Maulidevi
Exploring Deep Learning Models for Air
14.20 – 14.40 Quality Forecasting: A Comparative Review
Paper ID 014 Aang Aliyudin1, Aradea1, Rianto1, Rina
Mardiati2
Geographic Information System-Based
Recommendations For Analyzing The
14.40 – 15.00 Distribution Of Critical Land In The Citanduy
Paper ID 013 River Basin Using The Modified SWAT Method
Teguh Nurhadi Suharsono1, Rama Rosdiana2, Rina
Mardiati3

PARALLEL SESSIONS 2
System
Session Chair: Eki Ahmad Zaki Hamidi, MT
Time Title / Author
Analysis of Uplink and Downlink Interference
between NGSO and GSO Satellites at Ku Band
13.00–13.20
Frequency: Case Study on Starlink and
Paper ID 011
Telkom-3S
A.Susanto, Iskandar
Local Oscillator Phase Shifter for IQ Mixer in A
Synthetic Aperture Radar Application Based on
13.20 – 13.40
Quadrature Hybrid Branch-line Coupler
Paper ID 015
Abdurrasyid Ruhiyat1, Farohaji Kurniawan2, Catur
Apriono1
Robust PI Controller Design of Electro-
hydraulic Power Steering System for
13.40 – 14.00 Commercial Vehicles
Paper ID 016 Rina Ristiana1, Sunarto Kaleg1, Rina Mardiati2,
Alexander Christantho Budiman1, Abdul Hapid1,
Muhammad Arjuna Putra Perdana1
Estimated The Assist Map of The Electro-
Hydraulic Power Steering System for An
14.00 – 14.20 Electric Vehicle Microbus
Paper ID 017 Muhammad Arjuna Putra Perdana1, Rina Ristiana1,
Rina Mardiati2, Sunarto Kaleg1, Aam Muharam1,
Amin1
Ray Tracing-Based Channel Modeling for
Wireless Transmission in Indoor Environments
14.20 – 14.40
Nabil Taufiq Nur Izzat1, Nurul Fahmi Arief
Paper ID 019
Hakim1, Iwan Kustiawan1, Hana Arisesa2, Mariya
Al Qibtiya1, Azwar Mudzakkir Ridwan3
Automatic License Plate Recognition:
Automated Tax Verification for Registered
Vehicles via SAMBARA Server
14.40 – 15.00
Aif Umar Nawawi1, Arya Muhammad Dendyana1,
Paper ID 020
Galuh Yudha Prastyo1, Silmi Ath Thahirah Al
Azhima1, Nurul Fahmi Arief Hakim1, Azwar
Mudzakkir Ridwan2

PARALLEL SESSIONS 2
Thursday, 4 July 2024
Online Room-3: 13.00 – 15.00
Session Chair: Nike Sartika, MT.
Time Title / Author
Minimum Error Design for Transceiver Models
Using Software-Defined-Radio and GNU Radio
13.00 – 13.20
Muhammad Kamil Ridwan1, Nurul Fahmi Arief
Paper ID 021
Hakim1, Iwan Kustiawan1, Desi Fatimah1, Azwar
Mudzakkir Ridwan2
Performance Evaluation of Coverage Prediction
in 4G Networks Using Machine Learning
13.20 – 13.40 Classification Algorithm
Paper ID 025 Hajiar Yuliana1,2, Iskandar3, Hendrawan3, Azwar
Mudzakkir Ridwan4, Atik Charisma1, Nivika
Tiffany Somantri1
Enhancing the Accuracy of Conductivity
Parameters from Real-Time Rainwater Quality
13.40 – 14.00 Measurements based on Internet of Things
Paper ID 027 Utilizing Machine Learning
Ardiansyah Ramadhan1, Israel Mendonça3,
Masayoshi Aritsugi3, Indra Chandra2
Measurement Of Optimizer Performance On
The EfficientNet Architecture In Convolutional
14.00 – 14.20 Neural Network For Classification Of Matoa
Paper ID 031 Maturity Levels
Elliana Gautama1,2, Titik Khawa Abdul Rahman2,
Lia Kamelia3
Design of an Infantometer and Baby Scale in a
Digitalized Anthropometry Kit for Stunting
14.20 – 14.40 Prevention
Paper ID 034 Noer Fajrin1, Hendrawan1, Pingkan Aditiawati2, Ian
Josef Matheus Edward1, Kamarisima2, Rahma
Widya Ningrum2
Low-Cost Stereo Vision System Using ESP32-
CAM and YOLO
14.40 – 15.00
Danish Muhammad Hafidz∗ , Daven Darmawan
Paper ID 042
Sendjaya† , Muhammad Ogin Hasanuddin‡ ,
Pranoto Hidaya Rusmin§

PARALLEL SESSIONS 2

System
Session Chair: Lia Kamelia, Ph.D.
Time Title / Author
Using ESCH256 Algorithm of the SPARKLE
Permutation Family to Enhance the Security of
13.00–13.20 ESP-NOW Communication for Smart Lock
Paper ID 043 Systems
Arkan Dzaky Raihan Noor, Muhammad Ogin
Hasanuddin
Comparative Analysis of 5G New Radio (NR)
13.20 – 13.40 Network Planning on Low-, Mid-, and High-
Paper ID 045 Band in the Bandung City Area
David A. Immanuel1 , Iskandar2
Wrapper-Based Feature Selection to Improve
The Accuracy of Intrusion Detection System
13.40 – 14.00 (IDS)
Paper ID 046 Jupriyadi1,4, Arief Budiman4,5, Eki Ahmad Zaki
Hamidi1,2, Syaiful Ahdan1,4, Ridha Muldina
Negara1,3
An SAD architecture Verilog based for pattern
matching
14.00 – 14.20
Muhammad Adli Rizqulloh1, Erik Haritman1, Resa
Paper ID 049
Pramudita1, Roer Eka Pawinanto1, Nike Sartika2,
Agus Ramelan3
Cascade-Controller System of Hydroponic
14.20 – 14.40 Nutrient Solution
Paper ID 052 Putra Wisnu Agung Sucipto1, Nurhadi1, Annisa
Firasanti1, Eki Ahmad Zaki Hamidi2
Design of Feed Automation System and Water
Quality Monitoring in Louhan Fish Aquarium
14.40 – 15.00
Based on the Internet of Things
Paper ID 054
M. Alfarezi Putra1, Adam Faroqi1, Aan Eko
Setiawan1, Ida Nuraida2

PARALLEL SESSIONS 3

Session Chair: Azwar Mudzakir Ridwan, MT.
Time Title / Author
Enhancing Electricity Usage Monitoring and

Prediction with IoT-based Multiple Linear
15.20 – 15.40 Regression Algorithm
Paper ID 062 Fahni Amsyari1, Dian Sa’adillah1 Maylawati1,
Agung Wahana1, Wisnu Uriawan1, Nur Lukman1,
Hidayatul Fikra2
Performance Comparison of Vedic and
Karatsuba Algorithm for 64-Bit Barrett Modular
15.40 – 16.00 Multiplication
Paper ID 064 Handy Jonarta, Rafael Aditya Cahyo W., Muhammad
Ogin Hasanuddin, Infall Syafalni, Nana Sutisna,
Wildan Trusaji, Trio Adiono
Implementation Of Convolutional Neural
Network (CNN) in The Android-Based
16.00 – 16.20
Application for Detecting Coffee Bean Maturity
Paper ID 071
Eka Imamul Muhlisin1, Rin Rin Nurmalasari1, Lia
Kamelia1, Ramdani Wahyu Sururie2
Comprehensive Risk Evaluation Model for Data

16.20 – 16.40
Center Security Risk Assessment
Paper ID 106
Sayed Shoaib Hussaini, Budi Raharjo

PARALLEL SESSIONS 3

Session Chair: Edi Mulyana, MT.
Time Title / Author
Overfitting Defeat with Dropout for Image
15.20 – 15.40
Classification in Convolutional Neural Networks
Paper ID 079
Kukuh Nugroho, Hendrawan, Iskandar
Implementation of the Hybrid Base ABR

Algorithm in Named Data Networking Based
15.40 – 16.00
Video Streaming System
Paper ID 081
Hasan Nur Arifin, Nana Rachmana Syambas, Tutun
Juhana
Multiple Faces Recognition Model for Automatic

Attendance Using Multi-task Convolutional
16.00 – 16.20
Neural Network and Support Vector Machine
Paper ID 082
Rizky Nurhadhi, Lia Kamelia, Rin Rin
Nurmalasari, Shabrina Katresnawati
Implementation of Ultrasonic Sensor Using

LoRaWAN Protocol for Monitoring Water
16.20 – 16.40 Level in Aquaponic Pond
Paper ID 035 Ketut Agung Enriko1, Fikri Nizar Gustiyana2, FA
Brian Ganda Pratama2, Ahmad Luthfi2, Sri
Kuntadi2, Nila Indah Febriyanti2

PARALLEL SESSIONS 3

System
Session Chair: Saepul Uyun, MT.
Time Title / Author
Implementation of 13.56 MHz Crystals in
Oscillators for Wireless Power Charging
15.20 – 15.40 Applications
Paper ID 086 Fahmi Fathurrahiman1, Azwar Mudzakkir Ridwan1,
Aan Eko Setiawan1, Zenal Aripin2, Nurul Fahmi
Arief Hakim3
Semantic Segmentation of Iris from Pterygium

for Biometric Process using U-Net
15.40 – 16.00
Ambhara Putri Rie Kikhaawa1, Khoerun Nisa
Paper ID 088
Syaja’ah1, Yudha Satya Perkasa1, Rin Rin
Nurmalasari2, Wei-Hua Yang3
Text Classification Using NLP by Comparing

LSTM and Machine Learning Method
16.00 – 16.20
Zaidan Muhammad Mahdi1, Retno Fauziah
Paper ID 089
Istiqomah2, Alfarelzi1, Sri Astuti1, Ibnu Asror2,
Ratna Mayasari1
Design and Implementation of BIST logic for

16.20 – 16.40 High Speed and Energy Efficient CSLA
Paper ID 053 Garima Solanki1 , Sourav Agarwal2 , Tushar
Mishra3, Vansh Khandelwal4

PARALLEL SESSIONS 3

Session Chair: Nanang Ismail, MT.
Time Title / Author
Qibla-Finding Robot based on Vector and

Fuzzy Logic Methods for Accurate Directional
15.20 – 15.40
Alignment
Paper ID 096
Najin Khoerus Syakirin, Rina Mardiati, Aan Eko
Setiawan
Intelligence Malware Detection System with

15.40 – 16.00
LinearSVC
Paper ID 103
Zahrina Maryam, Hendrawan, Setia Juli Irzal Ismail
Synchronization Technique for Orthogonal

16.00 – 16.20
Time Frequency Space (OTFS)
Paper ID 105
Bella Wahmilyana Asril, E.Y. Hamid

POSTER SESSION
Paper ID Title / Author
001 User Clustering Using Modified K-means in
PD-NOMA
Rizki Nugraha1, Iskandar2, Hurianti
Vidyaningtyas3,4
002 Development and Deployment of Web
Application For User Clustering Using
Modified K-Means On PD-NOMA
Putra Rafii Pradana Santoso2, Iskandar1, Hurianti
Vidyaningtyas2,3
008
Load Balancing Algorithm Application on
Named Data Networking
Syah Muafa Al Kautsar1, Albert1, Gatra Akhira1,
Muhammad Arrifqi1, Achmad Kabir Rifai1, Geo
Perdana Saragih1, Ade Nurhayati1,2, Galih Nugraha
Nurkahfi1, Nana Rachmana Syambas1
018 Comparative Evaluation of Forwarding
Strategies in NDN-Simulated SEA-ME-WE
Network Topology
Patricia Adventia Pratama1, Bevan Julio Krisna1,
Nana Rachmana Syambas1, Samuel Tommy
Setiadjie1, Rima Saefiana1, Ratna Mayasari2,
Dennys Yudhistira Hutapea1, Daru Kristian Aji1,
Ridha Muldina Negara2
058 Layer 2 Communication Performance
Comparison of Ad-hoc Mesh and Software-
Defined Vehicular Network OpenFlow
Ribka Tiarma1, Nana Rachmana Syambas1, Eueung
Mulyana1, Galih Nugraha Nurkahfi2
059 Optimizing Customer Experience Through Low
Latency Search: A Delay Analysis of Heuristic
and linear search Methods for E-Commerce
Applications
Agil Fuad Gumelar, Tutun Juhana, Usman Zakaria

POSTER SESSION
Paper ID Title / Author
087 Selection of a Simple 5G Handover Algorithm:

Preparation for Named Data Networking
(NDN)
Ade Nurhayati , Syaiful Ahdan1,3, Tutun Juhana1,
1,2
Nana Rachmana Syambas1

107 A Systematic Review Network Slicing: Revenue
Optimization for Slice Admission Control
Objectives
Innel Lindra1, Hendrawan1, Agus Subekti2


Keynote Speaker

INSA Hauts-de-
France & Univ.
Polytechnique Hauts-de-
France
Prof. Iyad Dayoub, BSC, MASC, PHD, P. ENG.
Biography— Iyad Dayoub is Professor of Communications

Engineering, He has been awarded BEng from University of
Tishreen/Latakia in 1993 and completed his Post Graduate
Diploma from Damascus University in 1995. He received the MSc
degree from Nancy University/National Polytechnic Institute of
Lorraine (INPL)/ France in 1997. He awarded his PhD in 2001
from Valenciennes University/ Institute of Electronics,
Microelectronics and Nanotechnology (IEMN)/ in collaboration
with Alcatel Business Systems/ Colombes-Paris.
Professor Dayoub published over 200 papers in international

Journals and referred conferences. He is IEEE Senior member,
member of several Journal Editorial Boards such as Electonics &
Signal Processing Journal, and member of several international
conference advisory committees, technical program committees
and organization committees.
His current Research activities are 5G & beyond such as mmWave

communications, cognitive radios, NOMA, full-duplex
communications, IoT with IA, new waveforms (both Radio and
Radio over Fiber). He has Sound background in Intelligent
Transportation Systems (Railways & Vehicular Communications)
where He is seconded to the IRT-Railenium for 2 months a year
since 2017.

Pusan National University,

South Korea
Dr. Thi-Thu-Huong Le
Biography—Thi-Thu-Huong Le received a bachelor’s degree from

the Hung Yen University of Technology and Education (HYUTE),
Vietnam, in 2007, a master’s degree from the Hanoi University of
Science and Technology (HUST), in 2013, and a Ph.D. degree from
Pusan National University (PNU), South Korea, in 2020.
She has three years of experience as a Postdoctoral Researcher at
PNU, starting in 2020. She had seven years of experience as a
Lecturer at HYUTE. She has participated in machine learning
projects like NILM, IDS, AI Industry 4.0, AI security, and deep
learning-based CFD. She is a research professor at the Blockchain
Platform Research Center, PNU. Her research interests include
machine learning, deep learning, AI security, data analysis,
explainable AI, and signal processing.


Telematics, Networking, & Telematics

Applications

Paper ID 012
Object Detection for the Visually Impaired: A Systematic Literature

Review
Ghatan Fauzi Nugraha1, Aradea1, Rianto1, Rina Mardiati2

1
Departement of Informatics
Siliwangi University,Tasikmalaya, Indonesia
2
Department of Electrical Engineering
UIN Sunan Gunung Djati, Bandung, Indonesia
fauzighatan@gmail.com, aradea@unsil.ac.id, rianto@unsil.ac.id,
r_mardiati@uinsgd.ac.id
Abstract— The development of object detection devices has become a

significant investigative focus to enhance autonomy and quality of life for the
visually impaired. In this context, technologies of computer vision are the
predominant basis for evaluating and comparing the performance of several
current object detection models that have the potential to be employed as the
main component in assistive devices for the visually impaired. This inquiry
presented experimental results and performance evaluations demonstrating
the existence of approaches offering a good balance between detection
accuracy and inference speed, a pivotal criterion in the development of object
detection devices. However, significant challenges remain to be resolved,
notably regarding the dependence of model performance on the utilized
methodological approaches and the quality of the available datasets.
Therefore, further steps are needed to overcome it. In this paper, future
studies are proposed to focus on developing more complex and specialized
model architectures to increase object detection accuracy and reduce
inference time. Additionally, the diversification of datasets is an imperative
concern to ensure broader coverage of possible usage situations. The
integration of additional features (e.g. speech recognition and object distance
estimation) is also proposed to improve the usability of object detection aids
for the visually impaired. By taking these steps, it is expected to produce a
more sophisticated, reliable, and bettersuited object detection device based
on the user's needs. Apart from direct benefits for the visually impaired
groups, the contribution of this study is also expected to enrich the literature
in the computer vision domain, particularly in the context of applying
technology to improve human welfare in general.
Keywords—Computer Vision, Model Performance Evaluation, Object

Detection, Visually Impaired

Paper ID 021
Minimum Error Design for Transceiver Models Using Software-

Defined-Radio and GNU Radio
Muhammad Kamil Ridwan1, Nurul Fahmi Arief Hakim1, Iwan

Kustiawan1, Desi Fatimah1, Azwar Mudzakkir Ridwan2
1
Dept. of Electrical Eng. Education, Universitas Pendidikan Indonesia
2
Dept. of Electrical Engineering UIN Sunan Gunung Djati
Bandung, Indonesia
kamilridwan29@gmail.com, nurulfahmi@upi.edu, iwan
kustiawan@upi.edu, desifatimah00@upi.edu
azwarmudzakkiridwan@uinsgd.ac.id
Abstract—Software Defined Radio (SDR) refers to a radio system in

which software defines a subset or the entirety of its components.
SDR technology offers numerous benefits, such as its adaptable
nature and the capability to offer more cost-effective radio systems
through the substitution of hardware functions with software
configurations. In light of the numerous benefits associated with SDR
technology, considerable ongoing research is devoted to this subject
matter. Nevertheless, a significant research deficit persists in the
advancement of SDR technology with respect to the errors that are
present in the SDR model design process. The purpose of this article
is to examine how SDR transceivers are designed and simulated
utilizing the GNU Radio software. The simulation consists of three
schemes, the first of which conducts 2870 trials and the second and
third of which employ ADALM-PLUTO as the transceiver. The
outcomes derived from this investigation provide empirical evidence
that the transceiver configuration utilizing SDR and GNU Radio is
capable of executing 16-QAM modulation. By utilizing the block
diagram integrated within GNU Radio, one can achieve improved
reception quality with reduced noise levels. Moreover, the
adaptability of SDR facilitates the configuration of transceivers and
allows for their utilization as required.
Index Terms—GNU Radio, open-source, transceiver, Software-defined

radio.

Paper ID 023
Semantic Segmentation of Firearms Detection Using YOLOv8

Model
Fadli Alfa Rizki, Mohamad Irfan, Dian Sa’adillah Maylawati, Gitarja

Sandi, Muhammad Insan Al-Amin, Muhammad Ali Ramdhani
Department of Informatics
UIN Sunan Gunung Djati Bandung
Bandung, Indonesia
fadlialfarizki99@gmail.com, irfan.bahaf@uinsgd.ac.id
https://orcid.org/0000-0003-4963-287X, diansm@uinsgd.ac.id
https://orcid.org/0000-0002-1193-3370, sandi@uinsgd.ac.id
https://orcid.org/0000-0002-5091-7509,
muhammad.insanalamin@uinsgd.ac.id https://orcid.org/0000-0003-
2320-5431, m_ali_ramdhani@uinsgd.ac.id https://orcid.org/0000-0002-
6492-067X
Abstract—Firearm misuse by civilians remains a pressing concern,

with alarming statistics indicating high cases of firearm-related
suicides and firearm-related murders. Addressing this issue requires
effective monitoring systems to identify high-risk areas prone to
shootings. Recent technological advancements offer promising
solutions, notably the YOLOv8 model, which enables rapid object
recognition in digital images, including firearms. This research aims
to develop a firearm detection model using the YOLOv8 from images
and real live video. This research used 2,492 data images with
semantic segmentation, and nine iterations of model development
were conducted. The variation model experiment is based on data
division ratios of training and testing data (70:10, 80:20, 90:10) and
epochs (30, 40, 50). These experiments revealed that the model
utilizing a 90:10 data split and 50 epochs achieved the best
performance, with precision, recall, mAP50, and mAP50-95 values
for medium image sizes reaching 92%, 95%, 95%, and 82%,
respectively.
Keywords— convolutional neural network, detection, firearms,

semantic segmentation, YOLOv8

Paper ID 024
Microstrip Array Antenna Using Proximity Coupled and Dolph

Chebyshev Distribution for Wi-Fi 6E
Nurul Fahmi Arief Hakim1, Irfan Maulana1, Agus Heri Setya Budi1,
Azwar Mudzakkir Ridwan2, Yuyun Yuningsih3, Devi Eryanti3
1
2
Dept. of Electrical Engineering
3
Faculty of Da’wah and Communication
UIN Sunan Gunung Djati
Bandung, Indonesia
nurulfahmi@upi.edu, irfanmaulana17@upi.edu, agusheri@upi.edu
azwarmudzakkiridwan@uinsgd.ac.id, yuyunyuningsih@uinsgd.ac.id,
devieryanti@uinsgd.ac.id
Abstract—This article investigates the power distribution of square

patch antennas as described by Dolph Chebyshev. Experimental
research methodologies, including fabrication, measurement, and
optimization of antenna parameters, were utilized in this study. By
applying the Dolph Chebyshev distribution, the sidelobe caused by
an antenna modification involving multiple elements can be pressed.
Three elements of square-shaped configuration comprise the
arrangement antenna, which functions at a frequency of 6.175 GHz.
In order to achieve an expanded frequency range, every patch
element is coupled via proximity coupling to a 50 Ω SMA connector.
The configuration antenna is constructed utilizing two strata of FR-4
dielectric substrate, each measuring 0.8 mm in thickness and
measuring 68.7 mm by 50.8 mm. The simulation yielded a bandwidth
of 252 MHz, whereas the measurements indicated a bandwidth of 159
MHz. The simulation demonstrated a gain exceeding 8 dB, whereas
the measurements indicated a gain of 11 dB, and the sidelobe was
−14.8 dB.
Index Terms—antenna, array, proximity coupled, wifi 6E.

Paper ID 025
Performance Evaluation of Coverage Prediction in 4G Networks

Using Machine Learning Classification Algorithm
Hajiar Yuliana1,2, Iskandar3, Hendrawan3, Azwar Mudzakkir Ridwan4,

Atik Charisma1, Nivika Tiffany Somantri1
1
Faculty of Engineering Universitas Jenderal Achmad Yani
2
Doctoral Program of School of Elecrtical Engineering and Informatics
3
School of Electrical Engineering and Informatics (STEI)
4
Department of Electrical Engineering UIN Sunan Gunung Djati
Bandung, Indonesia
33222316@mahasiswa.itb.ac.id, iskandar@itb.ac.id, hend@itb.ac.id
azwarmudzakkirridwan@uinsgd.ac.id,
atik.charisma@lecture.unjani.ac.id, nivika.tiffany@lecture.unjani.ac.id
Abstract—Network coverage prediction is an important aspect of 4G

network planning and optimization. In this study, we conducted a
comprehensive analysis of the performance of various machine
learning algorithms in predicting 4G network coverage. The data
collected includes information on location, signal intensity,
topography, and other important variables. Data pre-processing is
performed to clean, normalize, and select relevant features. Next, we
compared the performance of several machine learning algorithms,
including Logistic Regression, KNN, Naïve Bayes, Decision Tree,
SVM, XGBoost, AdaBoost, and Bayesian Network Classifier. The
model training process is performed using subsets of training,
validation, and testing data, with evaluation performed using various
evaluation metrics such as accuracy, precision, recall, F1-score,
RMSE, MAE, and R-squared. The experimental results show that
Random Forest stands out with excellent performance, while Naïve
Bayes tends to produce lower performance. The conclusion of this
study provides valuable insights into the potential applications of
machine learning algorithms in predicting 4G network coverage with
accuracy and reliability.
Keywords— 4G Network Coverage, Coverage Prediction, Machine

Learning Algorithm, Performance Analysis

Paper ID 027
Enhancing the Accuracy of Conductivity Parameters from Real-

Time Rainwater Quality Measurements based on Internet of Things
Utilizing Machine Learning
Ardiansyah Ramadhan1, Israel Mendonça3, Masayoshi Aritsugi3, Indra

Chandra2
Graduate School of Electrical Engineering
1
Engineering Physics - School of Electrical Engineering, Center of
2
Excellence of Sustainable Energy and Climate Change
Telkom University, Bandung, Indonesia
3
Faculty of Advanced Science and Technology, Kumamoto University
Kumamoto, Japan
airamadhan@student.telkomuniversity. ac.id, israel@cs.kumamoto-
u.ac.jp, aritsugi@cs.kumamoto-u.ac.jp,
indrachandra@telkomuniversity.ac.id
Abstract—Rainwater, a valuable natural resource, can act as a carrier for various

pollutants and ions. Conductivity is one of the leading indicators of water quality,
reflecting the presence of dissolved substances acquired during its passage
through the atmosphere. Currently, real-time measurement of rainwater quality
based on the Internet of Things has been carried out at one measuring station,
namely on the rooftop of the Gedung Kuliah Umum (GKU) Telkom University,
Indonesia. However, real-time conductivity data has a relatively low level of
accuracy when compared with daily laboratory test data. Conductivity data from
real-time measurements during rain has two different conditions: when the water
temperature is uncontrolled and controlled. Therefore, this research aims to
increase accuracy and obtain the best general model in various conditions using
machine learning. The methods used are Artificial Neural Network (ANN),
Support Vector Regression (SVR), and Extreme Gradient Boosting (XGBoots).
Next, each model result is evaluated for Root Mean Square Error (RMSE), Mean
Absolute Error (MAE), Mean Relative Error (MRE), and Rsquared (R2 ). Based
on the model evaluation results, the XGBoost model performs best in increasing
accuracy. XGBoost Grid Search with RMSE results close to zero, a very high R2
of 0.99, and a relatively low MAE value of 3.5, which shows very accurate
prediction capabilities. They were followed by SVR RBF Random Search, which
also shows good performance with low RMSE and R2 of 0.23 and 0.78 and MAE
value of less than 0.16. Meanwhile, ANN epoch 100 has comprehensive
performance but still needs to improve prediction accuracy compared to the other
two models. Therefore, XGBoost with Grid Search can be the primary choice for
improving accuracy based on the evaluation results given.
Keywords—Artificial Neural Network, Conductivity, Extreme Gradient Boosting,

Improving Accuracy, Support Vector Machine

Paper ID 031
Measurement Of Optimizer Performance On The EfficientNet

Architecture In Convolutional Neural Network For Classification
Of Matoa Maturity Levels
Elliana Gautama1,2, Titik Khawa Abdul Rahman2, Lia Kamelia3

1
Department of Informatics Engineering, Perbanas Institute
Jakarta, Indonesia
2
Department of School and Graduates Studies, Asia e University
Kuala Lumpur, Malaysia
3
Department of Electrical Engineering, UIN Sunan Gunung Djati
Bandung, Indonesia
elliana@perbanas.id,c70105170005@aeu.edu.my, ,titik.khawa@aeu.edu
.my,lia.kamelia@uinsgd.ac.id
Abstract—In the rapidly developing field of deep learning, the efficacy of

neural network architectures and their components greatly impacts
performance in practical applications. This research evaluates the impact of
various optimization algorithms on the EfficientNet architecture in a
Convolutional Neural Network (CNN) framework for the classification of
Matoa fruit maturity levels. Matoa (Pometia Pinnata) is a tropical fruit
whose timely harvest is essential to ensure its quality. The EfficientNet
architecture, known for its balance between accuracy and computational
efficiency, was designed for the task. This research uses several leading
optimizers, including SGD, Adam, and AdaDelta, to examine the
convergence behavior, training efficiency, and classification accuracy. This
data set consists of 9466 annotated images of Matoa fruit at various stages of
ripeness, collected and preprocessed to suit research needs. The results
provide a comparative analysis of the performance of each optimizer,
highlighting the trade-off between speed and accuracy. 10-fold cross-
validation is used to evaluate model performance. The accuracy results from
the comparison of several optimizers used was achieved by the Adam
optimizer with an accuracy value of 93%, so that the model developed using
the CNN method, EfficientNet architecture, Adam optimizer, and 10-fold
crossvalidation can classify the maturity level of matoa optimally. The
findings of this research can help agricultural technologists and computer
scientists develop AI tools optimized for precision agriculture.
Keywords— EfficientNet Architecture, Convolutional Neural Network (CNN),

Optimizer Performance, Matoa Fruit Maturity Classification, Precision
Agriculture.

Paper ID 032
Automatic Water Cooling System to Increase The Performance of

PV
Riyani Prima Dewi1, Nike Sartika2, Novita Asma Ilahi1, Saepul

Rahmat1, Afrizal Abdi Musyafiq1, Erna Alimudin1
1
Electrical and Mechatronics Engineering, Cilacap State Polytechnic
Cilacap, Indonesia
2
Departement of Electrical Engineering UIN Sunan Gunung Djati
Bandung, Indonesia
riyanipd@pnc.ac.id, nikesartika@uinsgd.ac.id, nasmailahi@pnc.ac.id
saepulrahmat@pnc.ac.id, afrizal.abdi.m@pnc.ac.id,
ernaalimudin@pnc.ac.id
Abstract—This study experimentally examined the impact of cooling

the front surface of the photovoltaic (PV) system on its electrical
power generation and efficiency. The proposed cooling technology is
not only characterized by its simplicity and affordability, but it also
offers an effective method for preheating the supply water in a
reverse osmosis desalination plant. The cooling system is designed to
work automatically to flow water on the pv surface using RTC and
microcontroller for its automatic operating system. A pair of
identical photovoltaic (PV) modules were subjected to simultaneous
testing, one with cooling and one without. An automatic water cooling
system Implementation of the proposed cooling system resulted in a
10.30% improvement in the power generation of the PV. During
testing, the PV module with a water cooling system achieved an
electrical efficiency of 19.8%, but the module without cooling only
reached 17.4%.
Keywords—: Solar energy; PV; Power generation; Water cooling

Paper ID 033
The Impact of Collusive Interest Flooding Attack on NDN Network

Performance
Jupriyadi1,3, Ridha Muldina Negara1,2, Eueung Mulyana1, Nana

Rachmana Syambas1
1
School of Electrical Engineering and Informatics
2
School of Electrical Engineering Telkom University
Bandung, Indonesia
3
Faculty of Engineering and Computer Science
Universitas Teknokrat Indonesia
Bandarlampung, Indonesia
jupriyadi@teknokrat.ac.id, ridhanegara@telkomuniversity.ac.id,
eueung@stei.itb.ac.id, nana@stei.ce.itb.ac.id
Abstract—Collusive Interest Flooding Attack (CIFA) is a denial of

service (DoS) attack in Named Data Networking (NDN) networks.
Using a fake server, this attack develops the Interest Flooding Attack
(IFA) attack. The attacker sends a large number of requests to the
fake server. Sending requests to fake servers is intended to avoid
existing flood attack detection systems. In this paper, an analysis of
the impact of CIFA attacks is carried out by carrying out attacks and
varying network topology configurations. This is done to get a more
comprehensive idea of the possible impact of the attack. The
performance metric used is the interest satisfaction ratio (ISR) value,
which describes the ratio of the number of data packets consumers
receive to the total interest packet requests sent. The results of the
simulations show that CIFA attacks can reduce the performance of
the NDN network. This can be seen based on the ISR value, which
decreases when a CIFA attack occurs. By increasing the size of the
content store (CS) on the router, you can increase the ISR value so
that the impact of attacks can be minimized.
Keywords—NDN, collusive interest flooding attack, Named Data

Networking, CIFA

Paper ID 034
Design of an Infantometer and Baby Scale in a Digitalized

Anthropometry Kit for Stunting Prevention
Noer Fajrin1, Hendrawan1, Pingkan Aditiawati2, Ian Josef Matheus

Edward1, Kamarisima2, Rahma Widya Ningrum2
1
2
School of Life Sciences and Technology
Bandung, Indonesia
frinrin23@gmail.com, hendrawan@stei.itb.ac.id, pingkan@sith.itb.ac.id
ian@stei.itb.ac.id, kamarisima@sith.ac.id,
rahma.kartasasmita@gmail.com
Abstract—This research is part of the development of a digitalization

system for anthropometry kits aimed at analyzing child growth and
development to prevent stunting. The focus lies on crafting a
prototype capable of simultaneously measuring weight and body
length, utilizing ESP32, load cells, and ultrasonics, and employing
Bluetooth for data transmission to enable integration with mobile
apps. However, this study solely concentrates on the prototype of the
measuring device. Testing took place at the Cibeunying Sukaluyu
Integrated Health Service Post in Bandung City. The results revealed
that the placement of the same load at different positions can yield
varying load readings, attributed to two factors. Firstly, the precision
factor of the load cell itself, estimated at approximately +0.05% of its
maximum value. With four load cells used, each having a maximum
load of 50 kg, the resulting potential variation amounts to 0.1 kg.
Secondly, discrepancies arise due to different object positions,
leading to uneven pressure on the four load cells and subsequently
increasing differences. Additionally, the ultrasonic sensor accuracy
reaches 1%, resulting in increased sensor discrepancy as the
measured length extends.
Keywords—Stunting, Anthropometry, Body Weight, Body Height

Paper ID 035
Implementation of Ultrasonic Sensor Using LoRaWAN Protocol

for Monitoring Water Level in Aquaponic Pond
Ketut Agung Enriko1, Fikri Nizar Gustiyana2, FA Brian Ganda

Pratama2, Ahmad Luthfi2, Sri Kuntadi2, Nila Indah Febriyanti2
1
Dept. Telecommunication and Electrical Engineering
Institut Teknologi Telkom Purwokerto
Purwokerto, Indonesia
2
Research and Innovation Management Telkom Indonesia
Bandung, Indonesia
enriko@ittelkom-pwt.ac.id, fikrinizargustiana7899@gmail.com,
brian.pratama@telkom.co.id, ahmad.luthfi@itdri.id,
srikuntadi@telkom.co.id, nilaindahfebriyanti19@gmail.com
Abstract—This research evaluates the effectiveness of implementing

ultrasonic sensors utilizing the LoRaWAN protocol for monitoring
water levels in aquaponic ponds. The study demonstrates that these
sensors offer a remarkably high accuracy in water level monitoring.
Accuracy testing reveals that each measurement yields an average of
87% and a maximum of 100% accuracy for each sample, with
negligible variance between sensor measurements and manual
readings. Furthermore, battery life assessment indicates a significant
extension in battery longevity facilitated by the LoRaWAN protocol.
The protocol's wireless communication efficiency enables periodic
data transmission with minimal power consumption. Calculations
suggest that with a delivery interval of 1 minute, the estimated battery
lifespan is 1 year, while with 60-minute intervals, it may extend to
approximately 66 years. This technology holds immense promise for
enhancing the efficiency and dependability of monitoring systems in
aquaponic farming, thereby contributing to sustainable agricultural
practices.
Keywords—LoRaWAN, Aquaponic, Sensor, IoT

Paper ID 039
Implementation of Wi-Fi 5 Ghz Communication Using UDP

Unicast Protocol On Wheeled Soccer Robots
Arif Sumardiono1, Erna Alimudin1, Hendi Purnata1, Riyani Prima

Dewi1, Saepul Rahmat1, Nike Sartika2
1
Electronics Engineering, Cilacap State of Polytechnic
Cilacap, Indonesia
2
Electrical Engineering UIN Sunan Gunung Djati Bandung
Bandung, Indonesia
arifsumardiono@pnc.ac.id, ernaalimudin@pnc.ac.id,
hendipurnata@pnc.ac.id, riyanipd@pnc.ac.id, saepulrahmat@pnc.ac.id,
nikesartika@uinsgd.ac.id
Abstract—The Indonesian Wheeled Football Robot Contest (KRSBI-

B) is a competition in the field of robotics that uses the capabilities of
ball detection, robot navigation, ball kicking, and communication
between robots as well as communication from the Referee Box.
Communication on KRSBI-B is a must-have skill for each team so
that the robot can receive orders from the referee through the
Referee Box. Communication at the national level KRSBI-B 2023 is
required to use a 5 GHz Wi-Fi network with a maximum bandwidth
of 10.8 Mbps, where previously wheeled soccer robots could only be
connected to a 2.4 GHz WiFi network. The aim of this project is to
communicate using laptop hardware so that it can be connected to 5
GHz Wi-Fi which is applied to a wheeled soccer robot. The
communication protocol used is UDP with a Unicast data
transmission procedure in order to provide fast communication, and
small bandwidth so that it complies with KRSBI-B communication
regulations. There is an interface on the laptop communication
software that is made to facilitate monitoring of communication data.
By implementing this communication, the result is that
communication can run smoothly with a success rate of 100%, can
communicate with an access point distance of up to 15 meters, the
bandwidth used is 10.333 Mbps, and the robot can move to the
position ordered from the Referee Box, as well as the interface
ommunication can display messages received or sent.
Keywords— wheeled football robot, robot communication, UDP

Unicast

Paper ID 042
Low-Cost Stereo Vision System Using ESP32-CAM and YOLO
Danish Muhammad Hafidz∗ , Daven Darmawan Sendjaya† ,

Muhammad Ogin Hasanuddin‡ , Pranoto Hidaya Rusmin§
Bandung, Indonesia
Email: { ∗13221080, †13221062}@mahasiswa.itb.ac.id, { ‡moginh,
§pranoto.hr}@itb.ac.id
Abstract—Today, distance estimation algorithm (DEA) is very

important to enable machines to map the world and interact with it,
independent of human controls. These algorithms are beneficial for
various usages, such as the ADAS or Humanoid Robots, to control
the movements of each devices. In order to implement DEA, we can
integrate the concept of stereo camera with ESP32-CAM connectivity
capability to implement distance estimation algorithm in three-
dimensional space, by simulating how human eyes work. The
objectives of this experiment are to build a Stereo Camera with
ESP32-CAM module, design a Distance Estimation Algorithm (DEA),
implement DEA to estimate the distance of objects in three-
dimensional space, and measure the error and processing delay
generated by DEA. The stereo camera implementation stage involves
designing the device with ESP32-CAM, uploading the camera web
server program, and calculating camera parameters. Distance
Estimation Algorithm (DEA) algorithm is executed using Python and
YOLO to analyze images from both camera eyes. DEA provides
optimal performance for objects with distances below 7 meters after
calibration, the calculation of object distances above 7 meters
produces significant errors. The DEA implementation also shows
limitations in real-time, taking 0.1 - 2 seconds from ESP32-CAM
video recording to object calculation results due to delays from the
internet network and computer processing limitations.
Keywords—Distance Estimation Algorithm, Stereo Camera, ESP32-

CAM, YOLO

Paper ID 072
ANTIS: Automatic and Anthropometric Measurement and Weight

IoT-Monitoring for Enhanced Infant Nutrition Assessment Using
Dual Sensor and Fuzzy
Ahmad Hafidz Fajrian, Rin Rin Nurmalasari, Lia Kamelia

Bandung, Indonesia
1207070007@student.uinsgd.ac.id, id rinrinnm@uinsgd.ac.id,
lia.kamelia@uinsgd.ac.id
Abstract—Stunting, a condition of impaired growth and development

due to malnutrition, remains a serious problem in Indonesia despite
a yearly decrease in its prevalence. The stunting rate still has not met
the standard target set by WHO. Therefore, additional efforts are
needed to reduce stunting rates further. The development of ANTIS
was driven by the need for a tool to detect early stunting in infants
more adaptively and sensitively. ANTIS employs a fuzzy logic
approach to effectively handle data variations that may not conform
to standards, allowing for early preventive measures against
nutritional problems in infants. The components used in ANTIS
include the HC-SR04 load cell and ultrasonic sensors for accurately
measuring the baby's weight and height. Data on the baby's age and
gender are entered manually via a push button. ANTIS features a
20x4 I2C LCD screen and a thermal printer. A key innovation of
ANTIS is its ability to provide an adaptive assessment of the baby's
nutritional condition and integrate technology for real-time
monitoring. Thus, ANTIS is expected to become a reliable tool for the
early detection of stunting in infants, making a significant
contribution to efforts to address stunting in Indonesia.
Keywords—ANTIS, LoadCell, Ultrasonic, Fuzzy Logic, malnutrition

Paper ID 075
U-Net Approach to Mandibular Image Segmentation From

Panoramic X-rays
Nada Aisy Fadhila1, Khoerun Nisa Syaja’ah1, Yudha Satya Perkasa1,

Rin Rin Nurmalasari2
1
Department of Physics
2
Bandung, Indonesia
nadaaisyf@gmail.com, nisasya@uinsgd.ac.id, yudha@uinsgd.ac.id,
rinrinnm@uinsgd.ac.id
Abstract—Medical image segmentation is essential for disease

diagnosis and identification. U-Net, developed by Olaf Ronneberger,
Philipp Fischer, and Thomas Brox in 2015, is a well-known model for
effectively handling biomedical image segmentation tasks. Previously,
Amir Hossein attempted mandibular image segmentation with 93%
accuracy with MATLAB performance verification. This study
applies the UNet model for mandibular image segmentation with data
from Amir Hossein and ground truth from Label Studio. The best
performance was found at 500 epochs, with a Dice score of 0.9731.
This research uses data augmentation, ReLU, Sigmoid activation,
and hyperparameter variations, such as batch size, learning rate, and
number of epochs, to improve performance. The result is a prediction
mask for mandibular image segmentation with the best performance
at 500 epochs, a Dice score of 0.9731.
Keywords—Segmentation, U-Net, Deep Learning, Mandibular.

Paper ID 076
Cervical Cancer Classification From Pap Smear Using XCeption

Model
Ima Siti Maesaroh1, Khoerun Nisa Syaja'ah1, Yudha Satya Perkasa1, Sri
Hadianti3, Dwiza Riana4, Rin Rin Nurmalasari2
1
2
Bandung, Indonesia
3
4
Department of Computer Science
Universitas Nusa Mandiri
Jakarta, Indonesia
1197030017@student.uinsgd.ac.id, nisasya@uinsgd.ac.id, yudha@uinsgd.ac.id,
sri.shv@nusamandiri.ac.id, dwiza@nusamandiri.ac.id, rinrinnm@uinsgd.ac.id
Abstract—Cervical cancer is a disease that affects the reproductive

organs of women and contributes to a high mortality rate among
them. The rise in mortality can be attributed to delays in cancer
screening, underscoring the importance of early prevention measures.
While the Pap smear test is an effective method for early detection
and diagnosis of cervical cancer, manual inspection of test results still
has limitations in terms of observation and time-consuming
laboratory processes. This study aims to implement the Convolutional
Neural Network (CNN) model using transfer learning with the Xception
architecture to classify cervical cancer based on Pap smear images,
categorizing them into 2 or 4 classes. The dataset used in this study consists
of 400 cervical cell images, including normal cells, koilocytes, L-Sil, and H-
Sil, sourced from RepoMedUNM. The study involves several steps: data
splitting, data resizing and augmentation, hyperparameter tuning, CNN
model development using the Xception architecture, training, testing, and
model evaluation. The research results indicate that the Convolutional
Neural Network (CNN) with transfer learning using the Xception method can
accurately classify cervical cancer based on Pap smear images, achieving an
accuracy of 100% in the 2-class classification and 85% in the 4-class
classification.
Keywords—cervical cancer, pap smear, convolutional neural network,

xception.

Paper ID 070
SEMU5G: an Emulator for Improving Model Accuracy and

Performance Evaluation of 5G Network Core Architectures based
on SDN
Ridha Muldina Negara1,2, Arga Valen Lazuardi2, Muhammad Litfan

Rahmansyah2, Salman Alfarizi Novel Bajri2, Rohmat Tulloh3, Syaiful
Ahdan6, Eki Ahmad Zaki Hamidi4, Jupriyadi5,6
1The
University Center of Excellence Intelligent Sensing-IoT
2School of Electrical Engineering
3Telecommunication Technology School of Applied Science
Telkom University
4Electrical Engineering Department UIN Sunan Gunung Djati Bandung
5Electrical Engineering and Informatics, Institut Teknologi Bandung
6Faculty of Engineering & Computer Science
Universitas Teknokrat Indonesia,Lampung, Indonesia

ridhanegara@telkomuniversity.ac.id,argavalen@student.telkomuniversity.ac .id,
mlitfanrahmansyah@student.telkomuniversity.ac.id,salmanmelon@student.telko
muniversit y.ac.id,rohmatth@telkomuniversity.ac.id
syaifulahdan@teknokrat.ac.id,ekiahmadzaki@uinsgd.ac.id,jupriyadi@teknokrat.
ac.id
Abstract—The rapid advancement of 5G Radio Access Network (RAN)

architecture is facilitating the construction of 5G networks, marking a
significant milestone in telecommunications evolution. Given the complexity
of the 5G core architecture, traditional simulation methods are insufficient,
necessitating novel approaches. Emulator systems are crucial for creating
dynamic, controlled environments that enable exploration of real-world
scenarios without physical constraints. SEMU5G, a 5G core emulator, SDN
controller, and RAN simulator, utilizes Open5GS implemented in Docker
Container for test system flexibility and isolation. Additionally, an SDN
controller is integrated to monitor data flows in the User Plane Function
(UPF) and gNB simulated by UERANSIM in Mininet-WiFi. This
comprehensive integration facilitates effective and flexible real-world
exploration, providing a dynamic and controlled test environment for 5G
core research. Scenario testing comprises two stages: firstly, a fixed network
topology is employed to compare resource usage and confirm successful
SEMU5G integration. Secondly, a mobile network topology is utilized to
implement a mobile device scenario and compare the Quality of Service
(QoS) of SEMU5G with other available emulators. These stages ensure
thorough evaluation of SEMU5G's performance and its comparative
advantage over existing solutions.
Keywords— Software Defined Networking (SDN), Mininet, Radio Access
Network (RAN), UERANSIM, Mininet-WiFi.

Paper ID 073
NDN Collaborative Caching Replacement and Placement Policy

Performance Evaluation
Ridha Muldina Negara1,2, Khibran Muhammad Akbar2, Eki Ahmad Zaki

Hamidi3, Ivah Puspita Payangan2, Mutiara Dewi Puspitaningsih2, Adzka
Zahratu Putri As'ari2, Jupriyadi4,5
1
The University Center of Excellence Intelligent Sensing-IoT
2
School of Electrical Engineering
Telkom University
3
Electrical Engineering Department UIN Sunan Gunung Djati Bandung
4
Bandung, Indonesia
5
ridhanegara@telkomuniversity.ac.id,
khibranakbar@student.telkomuniversit y.ac.id,
ekiahmadzaki@uinsgd.ac.id, ivahpuspita@student.telkomuniversity.
ac.id, mutiaradewi@student.telkomuniversity .ac.id,
adzkazahratu@student.telkomuniversit y.ac.id,
jupriyadi@teknokrat.ac.id
Abstract—Named Data Networking (NDN) is an emerging technology that

aims to provide rapid and efficient content distribution and retrieval in the
network. This paper focuses on the impact of selecting the pair between cache
replacement and placement within NDN, an aspect often overlooked in
previous research. Four cache placement strategies will be implemented in
the initial simulation: Leave Copy Down (LCD), Leave Copy Everywhere
(LCE), Prob Cache, and Cache Less for More (CL4M). Each cache
placement will be combined with two replacements: least frequently used
(LFU) and least recently used (LRU). The effects of these combinations on
critical metrics such as cache hit ratio, latency, link load, and path stretch
are investigated. The results reveal that LCE combined with LFU
demonstrates remarkable enhancements in all measurements over LCE
coupled with LRU. These findings underscore the importance of meticulous
selection and optimization of cache placement and replacement strategies in
NDN environments to augment network efficiency and performance.
Keywords: Named Data Networking, Cache Placement, Cache Replacement,
Cache Strategy

Paper ID 074
Mini Prototype of the Futuristic Bin with an Automatic Waste

Sortation System for Managing the Garbage Problems in Society
Muhammad Taufik Dwi Putra1, Anugrah Adiwilaga1, Jezzy Putra

Munggaran1, Muhammad Azka Adhitama1, Raihan Anwar As'ad1,
Ahmad Ali Alhafidz1, Rin Rin Nurmalasari3, Agus Juhana2
1
Computer Engineering Study Program
2
Multimedia Education Study Program
Universitas Pendidikan Indonesia
3
Bandung, Indonesia
tdputra@upi.edu, anugrah.adiwilaga@upi.edu, jezzy.putra.munggaran@upi.edu,
muhammad17azkaa@upi.edu, raianwr@upi.edu, ahmadali@upi.edu,
rinrinnm@uinsgd.ac.id, agus.juhana@upi.edu
Abstract—This article explores the development and implementation

of futuristic bins with automated waste sortation systems to address
improper and mixed waste disposal challenges. Despite the existence
of categorized waste bins, individuals often dispose of waste
improperly, hindering the sorting and decomposition processes.
Integrating advanced technologies, such as Artificial Intelligence (AI),
futuristic bins offer innovative solutions. Futuristic bins with
automated sorting systems facilitate accurate waste sorting, reduce
mixing, and optimize waste disposal. These bins increase efficiency,
reduce costs, and optimize collection routes while promoting
environmental cleanliness and awareness of waste management. The
article describes the research methodology, including preliminary
study, specification requirements analysis, prototype development,
and testing. The results demonstrate the implemented system's
accuracy and real-time object detection capabilities. The study also
acknowledges limitations and suggests future research directions,
including improving communication between object detection
algorithms and microcontrollers. Overall, the development of
futuristic bins with automated waste sortation systems represents a
significant advancement in waste management, with the potential to
revolutionize waste management systems and create sustainable cities
and communities.
Keywords—Futuristic bin, Object detection, Smart Trash, Waste

Management

Paper ID 079
Overfitting Defeat with Dropout for Image Classification in

Convolutional Neural Networks

Bandung, Indonesia
33222024@mahasiswa.itb.ac.id, hend@itb.ac.id, iskandar@itb.ac.id
Abstract—Data is a crucial part of machine learning, used to create

models whose patterns are defined with specific algorithms. The
primary purpose of the machine learning process is to generate the
best prediction models. However, in general, models that trained well
in the training phase differ when used in the testing phase. This
phenomenon is called overfitting, a common problem in machine
learning, mainly when developing image classification models using
Convolutional Neural Networks (CNNs). The main task of the output
layer in CNNs is to classify the truth labels from the data input images.
One machine learning performance is defined by how well the models
can predict the truth labels. Overfitting is one of the problems that
can influence the model's performance. Dropout is one of the
regularization techniques intended to solve this problem. Our
purpose in this research is to know the relationship between the use
of dropout and CNN's performance in image classification from the
point of view of overfitting problems. The experimental results stated
that this technique can alleviate overfitting by 35% with a learning
rate value 0.001.
Keywords—Overfitting, Convolutional Neural Networks, Dropout,

Learning rate, Image Classification

Paper ID 081
Implementation of the Hybrid Base ABR Algorithm in Named Data

Networking Based Video Streaming System
Hasan Nur Arifin, Nana Rachmana Syambas, Tutun Juhana

Bandung, Indonesia
33222024@mahasiswa.itb.ac.id, nana@itb.ac.id, tutun@itb.ac.id
Abstract—This study aims to implement the Hybrid Base algorithm

in video streaming based on Named Data Networking (NDN) with a
focus on improving performance and user experience (Quality of
Experience). The Hybrid Base algorithm combines the advantages of
Buffer Base and Throughput Base algorithms to provide optimal and
adaptive video playback quality, striving to minimize delay and
maximize bandwidth utilization. This approach allows the algorithm
to adjust video bitrate based on user buffer conditions and current
network throughput. Through an experimental approach, this study
compares the performance of the Hybrid Base algorithm with
existing Adaptive Bit Rate (ABR) algorithms, showing superiority in
terms of video quality, network performance, and playback
performance for the Hybrid Base algorithm.
Keywords— Hybrid Base Algorithm, Video Streaming, Named Data

Networking

Paper ID 082
Multiple Faces Recognition Model for Automatic Attendance Using

Multi-task Convolutional Neural Network and Support Vector
Machine
Rizky Nurhadhi, Lia Kamelia, Rin Rin Nurmalasari, Shabrina

Katresnawati
Bandung, Indonesia
1207070113@student.uinsgd.ac.id, lia.kamelia@uinsgd.ac.id,
rinrinnm@uinsgd.ac.id, 1207070136@student.uinsgd.ac.id
Abstract—Attendance is important in various areas, especially in

educational environments such as schools and universities.
Traditional attendance methods using signatures are considered
inefficient because they take time, forgery, imitation, and
miscalculation. Biometric attendance can simplify the attendance
process, such as face recognition, which is designed to automatically
recognise multiple faces. The face recognition model for the
automatic attendance system is built with the MTCNN method for
detecting and extracting facial features in images or frames from
camera recordings and SVM method for classifying facial images that
have been extracted and trained. The experiments involved
variations in the distance of the face from the camera, the number of
individuals in the frame, k-fold cross-validation, and SVM kernel
type. Results show that face distance, number of individuals in the
frame, and SVM kernel type can affect face recognition accuracy.
The linear kernel produces the best accuracy, especially when only
one individual is present in the frame, with the highest accuracy
reaching 99.155%.
Keywords—MTCNN, SVM, Multiple faces, attendance


Wireless Technology, Mobile Communications,

Radar System

Paper ID 011
Analysis of Uplink and Downlink Interference between NGSO and

GSO Satellites at Ku Band Frequency: Case Study on Starlink and
Telkom-3S
A.Susanto, Iskandar
alan Ganesha no. 10 Bandung 40132, INDONESIA
23220348@std.stei.itb.ac.id
Abstract—We continue to investigate the aggregate interference in

uplink and downlink caused by NonGeostationary Satellite Orbit
(NGSO) constellations to GeoStationary Satellite Orbit (GSO)
systems. We use Starlink and Telkom-3S satellites as a case study. We
use the SGP4 to propagate their orbit based on their TLEs, a (near-)
real-time orbital element to simulate the uplink and downlink
interference scenario. We found that in the latitude range 0 to 15◦N,
the higher the GSO terminal latitude, the less the number of
interference events. We also observed that the Equivalent Power Flux
Density (EPFD) would exceed the interference limits recommended
by ITU article 22 without a separation angle. To prevent EPFD from
exceeding ITU limits the separation angle must be greater than 5◦.
Keywords— co-frequency, constellations, equivalent power flux

density, interference analysis, non-geostationary satellite orbit,
separation angle.

Paper ID 015
Local Oscillator Phase Shifter for IQ Mixer in A Synthetic Aperture

Radar Application Based on Quadrature Hybrid Branch-line
Coupler
Abdurrasyid Ruhiyat1, Farohaji Kurniawan2, Catur Apriono1

1
Universitas Indonesia
Depok, Indonesia
2
Research Center for Aeronautics
Technology National Research and Innovation Agency
Bogor, Indonesia
abdurrasyid.ruhiyat@ui.ac.id, faro001@brin.go.id
catur@eng.ui.ac.id
Abstract—Synthetic Aperture Radar (SAR) is a powerful instrument

in microwave remote sensing applications because of its capability for
all-weather time operation and producing high-resolution images.
The IQ mixer is a vital component of the RF (radio frequency) SAR
subsystem, which increases the bandwidth of the transmitted signal
and improves the resolution of the SAR system. The IQ mixer is
composed of a quadrature LO phase shifter, two matched mixers,
and a power combiner-splitter. This paper proposes a design for a
quadrature LO phase shifter based on a microstrip quadrature
hybrid branch-line coupler. The coupler provides different good
phase outputs around 90◦. The output and input transmission lines
have been placed on the outer side of the coupler to offer a flat gain
over a range bandwidth (150 MHz), positively influencing the output
response. The fabricated coupler satisfies all the specification
requirements and can be used as an LO phase shifter.
Index Terms—coupler, IQ mixer, phase shifter, synthetic aperture

radar, local oscillator

Paper ID 016
Robust PI Controller Design of Electro-hydraulic Power Steering

System for Commercial Vehicles
Rina Ristiana1, Sunarto Kaleg1, Rina Mardiati2, Alexander Christantho

Budiman1, Abdul Hapid1, Muhammad Arjuna Putra Perdana1
1
Research Center of Transportation Technology
National Research and Innovation agency
2
West Java, Indonesia
rina008@brin.go.id 0000-0002-7842-1126, suna024@brin.go.id 0000-
0002-5800-8713, r_mardiati@uinsgd.ac.id 0000-0002-3949-7830,
alex003@brin.go.id 0000-0002-3034-4352, abdu019@brin.go.id 0000-
0002-0599-4206, marj001@brin.go.id
Abstract— A controller design has been developed that is applied to electro-

hydraulic power steering (EHPS) systems for commercial vehicles. The
EHPS system is built in three subsystems, namely mechanical, hydraulic, and
electric motor subsystems. An EHPS model was built based on these three
subsystems to evaluate, validate and determine the EHPS control design,
especially for EHPS motor speed control. The model takes into consideration
modifications in the input of the driver steering torque given to the steering
wheel angle and changes in the supply of the hydraulic pump pressure in the
piston cylinder, which produces a shift of the sector gear distance. The EHPS
model is separated into three parts: linearization mechanical, nonlinear
hydraulic, and linear electric motor. Based on the EHPS model, a PI
(Proportional-Integral) controller is designed to control the speed of the
EHPS motor by assuming the part 1 and 2 models are in numerical form, so
that the control design is only based on the EHPS part 3 model, resulting in
a simpler controller design that still takes into account important variables
in the EHPS system. The PWM (pulse width modulation) technique is used
to control the speed of the EHPS motor by specifically regulating the motor's
input voltage. The goal of this PI Controller design is to obtain optimal EHPS
characteristics from changes in steering wheel angle, pressure, and motor
speed due to differences in steering wheel torque input, which results in the
displacement of the ball nut, which explicitly provides movement to the
vehicle wheels, which is known as turning maneuver, and to ensure that the
steering feels comfortable at high speeds while also providing good steering
at low speeds.
Keywords: PI Controller, EHPS modelling, commercial vehicles, hydraulic

pressure, mechanical steering, BLDC motor

Paper ID 017
Estimated The Assist Map of The Electro-Hydraulic Power Steering

System for An Electric Vehicle Microbus
Muhammad Arjuna Putra Perdana1, Rina Ristiana1, Rina Mardiati2,

Sunarto Kaleg1, Aam Muharam1, Amin1
1
2
Department of Tarbiyah and Teacher
Bandung, Indonesia
marj001@brin.go.id, rina008@brin.go.id 0000-0002-7842-1126,
r_mardiati@uinsgd.ac.id 0000-0002-3949-7830, suna024@brin.go.id
0000-0002-5800-8713, aam.muharam@brin.go.id 0000-0001-8803-
7999, amin007@brin.go.id 0000-0001-8600-9054
Abstract—The assist map of the electro-hydraulic power steering

(EHPS) system depicts the power and portability rules of the steering
in the shape of a curve. The assist map characteristic determines the
assist torque value depending on the vehicle’s longitudinal speed,
vehicle lateral acceleration, EHPS motor speed, and steering wheel
torque. The assist map is required to ensure portability and
performance, as well as to recognize disturbances in the steering
system caused by variations in steering drive torque and vehicle
lateral acceleration. To obtain precise steering performance, an assist
map estimate is required, which utilize the extended Kalman Filter
approach. The assist map curve is produced while the vehicle is
stationary or at rest; however, variations in steering drive torque and
vehicle lateral acceleration may be observed and approximated to
determine the magnitude of the steering wheel angle, hydraulic
pressure, and EHPS motor speed. According to the assist map
concept, the assist torque rules may be determined based on changes
in vehicle lateral acceleration and EHPS motor speed, which are then
supplied to the EHPS system to support the driver with navigation,
particularly when the vehicle turns.
Keywords: Extended Kalman Filter, Assist Map Curve, EHPS system,

Assist Torque, Light Vehicle, Steering Navigating, Vehicle Turn
Maneuvers.

Paper ID 019
Ray Tracing-Based Channel Modeling for Wireless Transmission in

Indoor Environments
Nabil Taufiq Nur Izzat1, Nurul Fahmi Arief Hakim1, Iwan Kustiawan1,
Hana Arisesa2, Mariya Al Qibtiya1, Azwar Mudzakkir Ridwan3
1
Dept. of Electrical Eng. Education
2
Research Center for Telecommunication National
Research and Innovation Agency
3
Bandung, Indonesia
nabiltni@upi.edu, nurulfahmi@upi.edu, iwan
kustiawan@upi.edu,hana002@brin.go.id, mariyalqibtiya@upi.edu,
azwarmudzakkiridwan@uinsgd.ac.id
Abstract—The application of ray tracing techniques to channel

modeling for MIMO-OFDM wireless communication systems in
complex indoor environments will be the subject of this article. Signal
propagation characteristics are analyzed using ray tracing methods,
specifically the Shooting Bouncing Ray (SBR) method, which takes
into account position variations between the transmitter (Tx) and
receiver (Rx) and the impact of building materials on the channel.
The research incorporates several parameters, namely received
signal intensity, path loss, signal delay, and the number of
interactions with the environment. A significant effect of Tx and Rx
position changes on channel quality is demonstrated by the research
findings. When the Rx is positioned at a greater distance from the Tx,
path loss and signal delay increase. These results offer crucial insights
for designers of wireless communications systems who are tasked
with optimizing network performance while considering
environmental factors that have an impact on signal propagation.
Index Terms—OFDM, Ray tracing, receiver, transmitter

Paper ID 020
Automatic License Plate Recognition: Automated Tax Verification

for Registered Vehicles via SAMBARA Server
Aif Umar Nawawi1, Arya Muhammad Dendyana1, Galuh Yudha

Prastyo1, Silmi Ath Thahirah Al Azhima1, Nurul Fahmi Arief Hakim1,
Azwar Mudzakkir Ridwan2
1
2
Bandung, Indonesia
aifumar29@upi.edu, aryamuhammadd@upi.edu,
galuhyudhaprastyo@upi.edu, silmithahirah@upi.edu
nurulfahmi@upi.edu, azwarmudzakkiridwan@uinsgd.ac.id
Abstract—The technology known as Automatic License Plate

Recognition (ALPR) originated from the application of computer
vision, a subfield of artificial intelligence systems. The purpose of
developing this ALPR system was to verify the tax status and
comprehensive vehicle information in the West Java region of
Indonesia. The objective of this study was to instill in each motorist a
sense of traffic order and to assist Indonesia in the implementation of
an Intelligent Traffic System (ITS). In order to operate efficiently, the
system incorporated in this design is constructed using a variety of
Python programming language modules. YOLOv8 (You Only Look
Once version 8), EasyOCR, OpenCV, and Selenium are the primary
Python modules utilized; additional modules include RegEx (Regular
Expression), Time, Numpy, and BytesIO. In this application, number
plates are searched for and detected using the dilation method so that
EasyOCR can extract every character with minimal error.
Additionally, machine learning serves as the foundation for each of
the aforementioned procedures and processes. The fundamental
computation of machine learning from bespoke data sets is performed in
Google Colab. The outcome of this investigation is a system capable of
furnishing information pertaining to motor vehicle systems to an individual
who has failed to remit vehicle tax. Aside from that, the developed system can
detect counterfeit license plates, making it a valuable tool for educating
drivers about the importance of remaining vigilant.
Index Terms—ALPR, Artificial Intelligence (AI), Computer Vision

(CV), image processing, unsupervised learning.

Paper ID 043
Using ESCH256 Algorithm of the SPARKLE Permutation Family

to Enhance the Security of ESP-NOW Communication for Smart
Lock Systems
Arkan Dzaky Raihan Noor, Muhammad Ogin Hasanuddin School of

Electrical Engineering and Informatics Institut Teknologi Bandung
Bandung, Indonesia Email: 13221048@mahasiswa.itb.ac.id,
moginh@itb.ac.id
Abstract—The smart lock system enables users to lock and unlock

doors using electronic devices. However, ensuring security in data
transmission between electronic devices is a major concern in
developing this system. The ESCH256 cryptographic algorithm of the
SPARKLE permutation family is utilized to enhance the security of
data transmission in the smart lock system through ESP-NOW
communication protocol. This research conducts tests on the software
level of the smart lock system that employs the ESP-NOW protocol
with ESCH256 algorithm, especially regarding its data transmission.
From the research conducted, it was obtained that the average time
spent to run the entire system was 1.556 seconds. The results of this
study demonstrate that the utilization of the ESCH256 algorithm can
reliably enhance the security of data transmission in the smart lock
system through ESP-NOW communication implemented on
constrained devices.
Index Terms—Smart lock, IoT, ESP32, ESP-NOW, ESCH256,

SPARKLE, Lightweight cryptography. I. INTRODUCTIO

Paper ID 045
Comparative Analysis of 5G New Radio (NR) Network Planning

on Low-, Mid-, and High-Band in the Bandung City Area
David A. Immanuel1 , Iskandar2

Telecommunications Engineering
Bandung, Indonesia
1
davidimanuel842@gmail.com, 2iskandar815@gmail.com
Abstract—In this research, 5G New Radio (NR) network planning is

conducted using the coverage planning method in a variety of
scenarios with varying frequency bands, including the low-band (700
MHz), mid-band (2.3 GHz), and high-band (26 GHz) with channel
bandwidths of 10 MHz, 100 MHz, and 800 MHz, respectively. This
5G NR network planning was conducted by simulating the Atoll 3.4
software in the Bandung City area with an area of 167.31 square
kilometers. The process carried out in the planning coverage in the
form of a Link Budget to determine the Maximum Allowable Path
Loss (MAPL) and identify the number of gNodeB sites required for
each scenario. The propagation model employed in this 5G NR
planning is UMa (Urban Macro) with the limitation of each scenario
carried out only in Downlink Outdoor-to-Outdoor (O2O) conditions.
The results of the parameters analyzed in this study are presented in
terms of Coverage including SS-RSRP, and SS-SINR, and in terms of
Capacity including Data Rate (Cell Capacity). The observation
results of SS-RSRP mean from highest to lowest, are as follows: at
low-band is -62.98 dBm, at high-band is -65.47 dBm, and at mid-band
is -69.24 dBm. The observation results of SS-SINR mean from lowest
to highest, are as follows: at low-band is 22.04 dB, at mid-band is 24.3
dB, and at high-band is 31.06 dB. The observation results of Cell
Capacity mean from lowest to highest, are as follows: at lowband is
71.645 Mbps, at mid-band is 178.712 Mbps, and at highband is
1139.837 Mbps.
Keywords—5G NR, Frequency Bands, Coverage Area, Link Budget,

SS-RSRP, SS-SINR, Data Rate (Cell Capacity)

Paper ID 046
Wrapper-Based Feature Selection to Improve The Accuracy of

Intrusion Detection System (IDS)
Jupriyadi1,4, Arief Budiman4,5, Eki Ahmad Zaki Hamidi1,2, Syaiful

Ahdan1,4, Ridha Muldina Negara1,3
1
2
3
Bandung, Indonesia
4
5
Department Computer Science and Electronics
Universitas Gadjah Mada
Yogyakarta, Indonesia
jupriyadi@teknokrat.ac.id, arief.budiman10@teknokrat.ac.id,
ekiahmadzaki@uinsgd.ac.id, ridhanegara@telkomuniversity.ac.id
Abstract—An Intrusion Detection System (IDS) is a system that can detect

attacks on a network. IDS systems can be built using machine learning
approaches. However, significant network traffic and many features cause
machine learning algorithms to work slowly. Not all available features are
essential for use as a reference in an intrusion detection system. Feature
selection is important because it can reduce the dataset and speed up
computing time in building models and detection systems. In this paper,
feature selection using the wrapper method is proposed. Its performance is
compared with filter- based feature selection, such as chi-square, information
gain, gain ratio, and correlation-based feature selection (CFS). Based on the
exploration, the results showed that the feature selection algorithm using the
wrapper method was superior to other algorithms. The number of critical
features discovered and the detection system's precision serve as proof of the
wrapper approach. With only three features as a reference, the detection
system produces an accuracy of 96.22%. It increases 5.84% from 90.38% to
96.22%. Computation time also increases because the number of features
used is tiny compared to other algorithms and uses all features.
Keywords—NSL-KDD, intrusion detection system, IDS, feature selection,

machinelearning

Paper ID 049
An SAD architecture Verilog based for pattern matching
Muhammad Adli Rizqulloh1, Erik Haritman1, Resa Pramudita1, Roer

Eka Pawinanto1, Nike Sartika2, Agus Ramelan3
1
Industrial Automation and Robotic Engineering Education
2
Bandung, Indonesia
3
Department of Electrical Engineering Universitas Sebelas Maret
Surakarta, Indonesia
muhammad.adli.riz@upi.edu, erikharitman@upi.edu, resa.pd@upi.edu,
roer_eka@upi.edu, nikesartika@uinsgd.ac.id,
agusramelan@staff.uns.ac.id
Abstract—With the increasing importance of data security,

encryption methods like advanced encryption standard (AES) have
become pivotal in safeguarding sensitive information. However, the
vulnerability of AES keys to side channel attacks (SCA) poses a
significant threat. This paper proposes a novel approach using sum-
of-absolute-difference (SAD) architecture for pattern matching,
implemented on field-Programmable gate arrays (FPGAs) using
Verilog. The SAD architecture, integrated with absolute difference
(AD) architecture, offers an efficient solution for detecting the onset
of AES computational processes without relying on victim signaling,
crucial for realworld scenarios. The architecture's functionality and
performance were validated through simulations using Vivado
software, demonstrating its efficacy in detecting data stream
similarities, with potential applications in embedded systems.
Keywords—pattern-matching; sum-absolute-difference; FPGA; SCA;

Paper ID 052
Cascade-Controller System of Hydroponic Nutrient Solution
Putra Wisnu Agung Sucipto1, Nurhadi1, Annisa Firasanti1, Eki Ahmad

Zaki Hamidi2
1
Electrical Engineering Department Universitas Islam “45”
Bekasi, Indonesia
2
Bandung, Indonesia
wisnu@unismabekasi.ac.id, nurhadi19@uinismabekasi.ac.id,
annisa_firasanti@unismabekasi.ac.id, ekiahmadzaki@uinsgd.ac.id
Abstract—Generally, the process of making hydroponic nutrients

with AB mix basic ingredients is done manually. Starting from
defining the ratio of solutions "A" and "B", water supply, and the
stirring process until the solution is evenly mixed. Nowadays, our
technology is allowing the process to be done automatically. The
multilevel control system applied to this process has 2 control
variables and 1 support system, namely stirrer. The first variable;
water volume; is controlled by the PID method, with servo output as
the setting of the water tap opening. The second variable; nutrient
concentration; is also controlled by the PID method, where the output
is a DC pump that injects solutions "A" and "B". In maximizing the
yield of an even solution, stirrer is also applied which is controlled by
the multilevel control method. This means that the stirrer is regulated
by PID control of water volume and PID of nutrient concentration.
The test results of the variable water volume system with Kp = 1.2,
Ki = 0.0001, and Kd = 0.0008 obtained a rise time of 76 seconds and
a settling time of 74 seconds for a volume input value of 2.5 liters. The
variation of nutrient concentration with Kp = 2, Ki = 0.0001, and Kd
= 0.001 obtained a rise time of 30 seconds and a settling time of 55
seconds for the difference in concentration between the initial and
final values of 50 ppm.
Keywords— Hydroponic, AB mix, Stirer, Cascade Controller, PID

Paper ID 053
Design and Implementation of BIST logic for High Speed and

Energy Efficient CSLA
Garima Solanki1 , Sourav Agarwal2 , Tushar Mishra3, Vansh

Khandelwal4
Dept. of Electrical Engg.1,2,3,4
Delhi Technological University
Delhi, India
garimasolanki@dtu.ac.in1, souravagarwalcr7skr@gmail.com2,
tusharmishra263@gmail.com3, vansh.khandelwal06@gmail.com4
Abstract—This paper investigates the implementation of BuiltIn Self-

Testing (BIST) techniques tailored specifically for highspeed and
energy-efficient Carry Select Adders (CSLA). The study proposes a
novel CSLA design dominated by carry generation logic, aiming to
optimize performance while minimizing power consumption and
chip area. By integrating BIST circuitry into the CSLA architecture,
the research contributes to enhancing fault detection and ensuring
system integrity in high-speed computing applications. The proposed
CSLA design employs carry generation logic to improve
computational efficiency while mitigating power consumption and
propagation delay. Through simulation using Xilinx Vivado, the
study evaluates the effectiveness of the integrated BIST circuit in
detecting and diagnosing faults within the CSLA. The results
demonstrate the practical viability of the BISTenhanced CSLA
design in maintaining arithmetic computation integrity and
reliability in high-speed computing environments. This research
underscores the importance of integrating BIST techniques into
CSLA designs to enhance fault detection and diagnosis capabilities,
thereby improving overall system reliability and performance. The
findings offer valuable insights into the implementation of BIST
methodologies in high-speed and energyefficient computing
architectures, with implications for enhancing fault tolerance and
system robustness in real-world applications.
Index Terms—Carry Select Adder(CSLA), Built-In Self Test(BIST),

Multiple Input Signature Register(MISR), Linear Feedback Shift
Register(LFSR), Circuit Under Test(CUT)

Paper ID 054
Design of Feed Automation System and Water Quality Monitoring

in Louhan Fish Aquarium Based on the Internet of Things
M. Alfarezi Putra1, Adam Faroqi1, Aan Eko Setiawan1, Ida Nuraida2

1
2
Department of Mathematics Education
Bandung, Indonesia
malfareziputra28@gmail.com, adam.faroqi@uinsgd.ac.id,
aaneko37@gmail.com, idanuraida@uinsgd.ac.id
Abstrak -The development of ornamental fish farming in Indonesia continues

to increase, especially in freshwater ornamental fish. One of the freshwater
ornamental fish that is in great demand by the public is the louhan fish. In
its development, louhan fish maintenance activities often encounter obstacles
and require perseverance in maintaining them so they can grow properly.
Routine feeding of louhan fish is important because it determines the
nutrients for its growth and development. Common problems that are often
encountered in louhan keepers are lack of time, having other activities, and
being in a remote position that makes it impossible to constantly feed and
monitor. Therefore, this study designed an automatic feeding system and
water quality monitoring in a louhan aquarium using a microcontroller from
ESP32 which was then integrated with the DS3231 Real-Time Clock Module,
DS18B20 Temperature Sensor, pH-4502C Sensor, TDS-SEN0244 Sensor,
SG- 90 Servo Motor, Relay, Buzzer, Heater¸ Liquid Crystal Display, and
other electronic components. The results of testing the feed automation
system, namely the accuracy of the Real-Time Clock Module, obtained a
delay of 10 seconds and the Servo Motor movement system can issue an
average feed weight of 0.072 grams in 1 rotation of the motor with an angle
of 150° and a delay of 1 second. The test results on the water quality
monitoring system were carried out on three indicator variables, namely
temperature, degree of acidity (pH), and total dissolved substances (TDS) in
water which had a relatively small average error rate. The results of testing
the heater in stabilizing the aquarium temperature, the average period of the
heater in the on condition until it turns off reaches 13.11 minutes.
Keywords— ESP32, Feeder, Internet of Things, Louhan.

Paper ID 055
Design and Build a Prediction System State of Charge (SoC) Battery

Lithium-ion for Battery Based Management Internet of Things
(IoT)
Muhammad Alwi Sya'ban1, Adam Faroqi1, Lia Kamelia1, Devi Eryanti2

1
2
Department of Islamic Counseling Guidance
Bandung, Indonesia
1187070048@student.uinsgd.ac.id, adam.faroqi@uinsgd.ac.id,
lia.kamelia@uinsgd.ac.id, devieryanti@uinsgd.ac.id
Abstract -Batterylithium-ion is the most widely used type of battery

because it has superior durability and efficiency compared to other
types of batteries. In long-term use, methods are needed to extend its
lifespan, and also the rise of IoT-based innovations makes it
interesting to collaborate in a study using the Telegram application.
This study discusses the prediction of SoC using the methodOpen
Circuit Voltage, the INA219 sensor as a voltage reader, and the
ESP32 microcontroller as a processor and sender of data to Telegram
to be able to find out the remaining capacity of the battery so that the
battery is not used in unhealthy conditions. The research begins with
preparing and designing the components needed for research, then
the components are assembled and after that tests are carried out on
Telegram sensors and bots to find out the value of accuracy anddelay
that occurs during use. The results in testing with three voltage
variations at 12.6 V, 12.0 V, and 11.4 V obtained an accuracy value
of 99.78% for each 12.6 V variation, then 99.56% for the 12.0
variation V, and 99.84% for the 11.4 V variation. Meanwhiledelay on
the Telegram bot is at an average of 6.079 seconds.
Keywords— Batterylithium-ion, SoC Prediction, IoT,Open Circuit

Voltage.

Paper ID 056
Design and Build of IoT-Based Dragon Fruit Plant Maintenance

System Using the Nodemcu ESP8266 Microcontroller
Abdul Haris1, Adam Faroqi1, Edi Mulyana1, Siti Sanah2

1
Departement of Electrical Engineering
2
Department of Arabic Education
Bandung, Indonesia
harisj993@gmail.com, adam.faroqi@uinsgd.ac.id
edim@uinsgd.ac.id, siti.sanah@uinsgd.ac.id
Abstract—The yield of naturally occurring dragon fruit plants is still

quite low. Therefore, a tool capable of conducting maintenance
procedures without harming the dragon fruit plant itself was
developed. The device has an autonomous watering pump that
operates depending on temperature and soil conditions. The water
pump will operate if the soil is too dry, as determined by the air
temperature, which at a specific degree causes the soil to lose
moisture. Additionally, the device is fitted with a fluid-feeding pump
that is controlled by the Telegram program. The Telegram
application activates and monitors each step of feeding. The system
employs lights that switch on and off at specified intervals. In the
watering system, the soil moisture sensor readings for soil conditions
correspond to the accuracy of the sensor readings for soil conditions,
and the DHT11 sensor's temperature readings are accurate to 98.0%.
The functioning of the combination of the two sensors has been
effective, with watering occurring when the soil moisture sensor's
values surpass 430 and the DHT11 sensor's readings exceed 36°C.
Lights are programmed to turn on and off at certain intervals. Both
the DHT11 sensor's temperature readings and the soil moisture
sensor's readings are accurate to within 0.2% for use in the watering
system, while the DHT11 sensor's accuracy for soil temperature
readings is 99.2%. When the soil moisture sensor reads more than
430 and the DHT11 reads more than 36 degrees Celsius, the watering
process is triggered. This combination of sensors has shown to be
functional.
Keywords— watering, radiation, nutrition, sensor, SNTP.

Paper ID 061
Behavior in Financial Technology: Study of Intention to Use of

Mobile Payment
Leni Susanti1, Siti Hajar Mohamad2, Doni Purnama Alamsyah1

Creativepreneurship Department
1
BINUS Business School Bina Nusantara University
Jakarta, Indonesia
2
Department of Business Management and Law Management and
Science University
Selangor, Malaysia
leni.susanti@binus.ac.id sitihajar_mohamad@msu.edu.my
doni.syah@binus.ac.id
Abstract—Research on mobile payments continues to increase,

considering the positive impact on the issue of sustainability of
financial technology. This research aims to examine attitudes toward
using mobile payments through support from perceived
compatibility, perceived benefits, and social influence. The research
was conducted quantitatively with a survey of 331 respondents. The
data from respondents was processed using SmartPLS through two
tests, namely the PLS Algorithm and Bootstrapping Process. The
research analysis is presented in line with the results of the hypothesis
test. In the study, it is known that perceived compatibility and
perceived benefits have an important role in supporting the attitude
toward using mobile payments. The compatibility of smartphones
and applications is the main point that has a positive impact on the
use of mobile payments. Other findings explain that social influence
does not have a significant impact on attitude to use, allegedly
because users pay more attention to mobile payment compatibility.
The research findings are important information for service
providers, which need to be studied in more depth on the level of
compatibility and benefits of mobile payment services. This is an
effort to support the sustainable use of mobile payment services.
Keywords—Mobile Payment, Compatibility, Financial Technology

Paper ID 065
Convolutional Neural Network with VGG-16 Architecture for

Object Image Classification in Arabic Language
Iqbal Putra Ramadhan, Dian Sa'adillah Maylawati, Diena Rauda

Ramdania, Agung Wahana, Cepy Slamet, Rifqi Syamsul Fuadi
Bandung, Indonesia
iqbalputra2892015@gmail.com, diansm@uinsgd.ac.id,
diena.rauda@uinsgd.ac.id, wahana.agung@uinsgd.ac.id,
cepy_lucky@uinsgd.ac.id rifqi@uinsgd.ac.id
Abstract—Arabic is one of The challenging languages to learn. It

takes 1320 intensive hours to reach a proficient level. The media plays
an active role in enhancing learning and achieving students' goals,
which significantly influences the success and efficiency of education,
especially for school students who find it challenging to learn the
Arabic language. Based on this, a classification model is needed to
interpret objects in Arabic, making The learning process more
efficient and varied. The design of this system uses The Convolutional
Neural Network (CNN) method as an image data processor. All
research data uses image pictures as input and output, including
Tables, Pens, Chairs, Whiteboards, Clocks, People, Mobile Phones,
Plants, Books, and Bags. All research data consists of 7,341 images of
class objects, including training and testing data. The training data
uses 6085 images, and the test data uses 1256 images each. This
research employs four testing scenarios with variations in learning
rate and dropout parameters. Learning rate variations include 0.001
and 0.0001, while dropout variations are 0.3 and 0.5. The test results
from all scenarios show that a learning rate of 0.001 and a dropout of
0.5 have The best accuracy value at 90%. This value proves that The
CNN algorithm performs well in detecting class objects.
Keywords— Convolutional Neural Network, Deep Learning, Object

Detection

Paper ID 067
Microstrip Antenna with Mushroom-like EBG for Gain

Enhancement at 2.3 GHz Frequency
Tubagus Arya Wirangga1, Ahmad Saepul Milah1, Nanang Ismail1,

Harfan Hian Ryanu2
1
Departemen of Electrical Engineering
2
Bandung, Indonesia
tubagusaryawirangga@gmail.com, ahmadsaefulmilah56@gmail.com,
nanang.is@uinsgd.ac.id, harfanhr@telkomuniversity.ac.id
Abstract—This paper proposed a design and realization of microstrip

antenna to enhance gain of the antenna using mushroom-like
Electromagnetic Band Gap (EBG) method. The operational
frequency range for this antenna is 2.35 GHz (2,300 - 2,400 GHz). The
designs of two antennas are one without a mushroom-like EBG
technique and the other with mushroom-like EBG compared in this
working paper, both of the antenna’s designs work at the same
frequency. Parameters compared are antenna’s return loss, VSWR,
bandwidth, and radiation pattern. The result is the microstrip
antenna using mushroom-like EBG successfully enhanced the gain of
the antenna both in simulation and the pabricated antenna. The
simulation result show that the microstrip antenna using mushroom-
like EBG enhanced the gain by 8,69%, while the pabricated
microstrip antenna using mushroom-like EBG successfully enhanced
the gain up to 33,15% from 1.849 dBi to 2.462 dBi.
Keywords—Gain, Microstrip, Mushroom-like EBG

Paper ID 068
The Effect of Uniplanar-Type EBG on Microstrip Antenna Gain at

the 2.3 GHz Frequency
Reza Anggi Rizkia1, Ahmad Saepul Milah1, Nanang Ismail1, Harfan

Hian Ryanu2
1
Departemen of Electrical Engineering
2
Bandung, Indonesia
Arreza244@gmail.com, ahmadsaefulmilah56@gmail.com,
nanang.is@uinsgd.ac.id, harfanhr@telkomuniversity.ac.id
Abstract—This study deals with increased microstrip antennas gain

using the uniplanar-type Electromagnetic Band Gap (EBG) method.
The antenna designed to increase the gain, works at a 2.35 GHz
working frequency, in line with the 5G band that is available in
Indonesia. The cutting technique used is the feedline cutting
technology, and the substrate material used is FR-4 Epoxy with a
permittivity value of 4.4 with a thickness of 1.6 mm. The study
analyzed the comparison of two manufactured antennas, an antenna
without EBG and with EBG operating at the same frequency. The
results showed that the better gain than antennas without the use of
EBG, with a percentage increase of 35.95%.
Keywords—Antenna microstrip, electromagnetic band gap, uniplanar

type, gain, 5G.

Paper ID 078
Classification of Cassia Cinnamon and Ceylon Cinnamon using

Deep Learning
Agus Pratondo1, Nanang Ismail2, Brian Nayaka Prasetya1, Muhammad

Fatah Al Khadami1
1
Department of Multimedia Engineering
School of Applied Sciences
Telkom University
2
Dept. of Electrical Engineering Faculty of Engineering
Bandung, Indonesia
pratondo@gmail.com, nanang.is@uinsgd.ac.id,
brianprasetya08@gmail.com, fatah.alkhadami17@gmail.com
Abstract— Cinnamon, a widely utilized spice with diverse culinary

and medicinal applications, consists of various variants, notably
Cassia and Ceylon. Accurate classification of these variants is crucial
for ensuring product authenticity and quality control in the spice
industry. In this study, we propose a classification framework
utilizing deep learning techniques, specifically transfer learning with
the VGG16 architecture. A dataset comprising 306 images of Cassia
cinnamon and 317 images of Ceylon cinnamon was used to train and
evaluate the classification model. The results demonstrate a
promising accuracy of 82.66%, indicating the efficacy of the proposed
approach in distinguishing between Cassia and Ceylon cinnamon
variants. This classification method holds significant importance in
enhancing quality assurance processes and ensuring consumer
confidence in cinnamon products. The findings suggest that deep
learning-based classification methods offer a reliable and efficient
solution for the authentication and classification of spice variants,
thereby contributing to improved product quality and consumer
satisfaction in the spice industry.
Keywords—Classification, Cinnamon, VGG 16, RandomWeight,

Pretrained-Weight.

Paper ID 080
PIT Performance Measurement using NFD Pipeline Parameters in

Named Data Networking (NDN)
Adi Sucipto1,2, Syaiful Ahdan1,2, Iskandar1, Nana Rachmana Syambas1

1
Bandung, 40132, Indonesia
2
Lampung 35132, Indonesia
adi.sucipto@teknokrat.ac.id, syaifulahdan@teknokrat.ac.id,
iskandar@stei.itb.ac.id, nana@stei.itb.ac.id
Abstract—Although NDN is an entirely new architecture, its design

principles are based on the success of the current Internet, they
represent our knowledge of the advantages and disadvantages of the
current Internet architecture, and they can be implemented
gradually using the operations of the current Internet. The Pending
Interest Table (PIT), the Forwarding Information Base (FIB), and
the Content Store (CS) are the three data structures found inside the
NDN router. Interest sent upstream to the content source is tracked
by the Pending Interest Table (PIT), for data to be transmitted to
requesters downstream. PIT is also used for measurement, loop
detection, and the containment of recently satisfied interest. In this
study, we did the simulation to measure and analyze the performance
of PIT in NDN using NFD Pipeline Measurements. The NDNSim
simulator was used in the research to observe PIT performance. The
findings of the simulation and analysis indicated that in an NDN
network, Satisfied Interest was influenced by the granularity of a
prefix. At the number of interests 1000, the results showed a decrease
in packets, including a decrease of around 21% in the Ininterest packet, and
around 30% inSatisfiedInterest and InData. The effect of Prefix length also
decreased in Ininterest ranging from 0.014-0.015%, while in
inSatisfiedInterest and InData it ranged from 4.2-8.5%. The PIT's
performance was influenced by the length of the prefix and the
number of interest messages the customer sent.
Keywords—Named Data Networking, NDNSim, NFD, Pending

Interest Table, Prefix.

Paper ID 083
Implementation of Smart Electric Meter for Solar System Based

on Internet of Things
Rizky Mahesa Ramadhan, Lia Kamelia, Rin Rin Nurmalasari, Rangga

Julfian Hakim, Rizky Nurhadhi
Bandung, Indonesia
1207070112@student.uinsgd.ac.id, lia.kamelia@uinsgd.ac.id,
rinrinnm@uinsgd.ac.id, ranggajulfianhakim07@gmail.com,
1207070113@student.uinsgd.ac.id
Abstract—Electricity has become a necessity in human activities. One of the

renewable energy sources (EBT) that can be used as an easily implementable
alternative electricity provider on a residential scale is the solar power plant
(PLTS). However, PLTS cannot be remotely monitored for its performance.
Currently, the use of electric power in Indonesian households still relies on
kWh meters provided by the state electricity company (PLN). These meters
only provide electricity usage data without displaying the corresponding
tariff costs. This research aims to develop a device to monitor electricity
consumption and control electrical loads' on/off status via the Internet to
conserve energy. The device has an automated timer to turn electronic
devices on and off and an automatic transfer switch (ATS) in the solar power
system based on time settings or scheduling. The device uses NodeMCU as
the microcontroller and a PZEM-004T sensor for voltage, current, and
power measurements. Based on the test results, the device functions properly.
Electricity consumption data, including power, current, voltage, and cost,
can be displayed on LCD and IoT MQTT Panel App that can be accessed on
smartphones. Testing the PZEM-004T sensor by comparing current
measurements using a clamp meter showed an average error of 5.038%, and
comparing power measurements with the power in the electrical load
specifications used showed an average error of 4.7%. This shows the
suitability of the PZEM-004T sensor in the system because the error value is
relatively low. The test results for the relay switch response had a maximum
response of 1.94 seconds but were still relatively fast. The lamp is turned on
or off in all tests according to the switch.
Keywords— Solar Power, NodeMCU, Smart Electric Meter, Internet of Things

(IoT), System Control

Paper ID 084
Indonesian Sign Language Translation System Using ResNet-50

Architecture-Based Convolutional Neural Network
Zalfa Humaira Salsabila, Rin Rin Nurmalasari, Lia Kamelia

Bandung, Indonesia
zalfa33.zha@gmail.com, rinrinnm@uinsgd.ac.id,
lia.kamelia@uinsgd.ac.id
Abstract—Research on Convolutional Neural Networks (CNN) continues to

be developed to increase efficiency and speed in processing data and reduce
the need for large amounts of training data. CNN has several network
architectures. In this research, the ResNet-50 architecture is used. The
ResNet-50 architecture, usually called transfer learning, is used in the
learning process. The model trained using the ResNet-50 architecture is used
as a basis for learning other tasks. The classification layer in a ResNet-50
CNN has several essential components. First, a flattened layer converts the
spatial representation of features into one dimension with global average
pooling, producing an image feature vector. Then, although dropout is rarely
used in ResNet50 architecture, we can add it if necessary to prevent
overfitting. After that, the feature vector is passed to the Dense layer with
128 neuron units and ReLU activation to learn a more complex feature
representation. Finally, the feature vector is passed to the final Dense layer
with 25 neuron units and softmax activation, which generates probabilities
for each possible class, allowing the model to provide the most likely class
predictions for a given image. The dataset used in this research is 200 images
for one label, so for 25 labels, there are a total of 5000 images. This dataset
will then be divided into two, namely training data and testing data. To
produce the highest accuracy, the model parameters will be developed. The
model parameters developed in this research are the dataset ratio, the
number of epochs, and the learning rate. The results of testing the ResNet-50
CNN model produced the highest accuracy of 99.15 and the lowest loss, 0.25,
with the dataset divided using a value of learning rate 0.01, the ratio of 70%:
30%, epochs 50 times with 110 batches.
Keywords— Convolutional Neural Network, ResNet-50, Transfer Learning.

Paper ID 085
A Comparative Analysis of Machine Learning Algorithms: K-

Nearest Neighbor (KNN) and Random Forest for Predicting Sports
Poses
Nur Syifa Maulida, Lia Kamelia, Rin Rin Nurmalasari

Bandung, Indonesia
nurasyifamaulida615@gmail.com, lia.kamelia@uinsgd.ac.id,
rinrinnm@uinsgd.ac.id
Abstract—The aging process often leads to a decline in physical

function, emphasizing the importance of maintaining physical
activity, mainly through exercise. However, a significant portion of
the elderly tends to become less physically active. In response to this
challenge, solutions such as using artificial intelligence, specifically
Machine Learning, have been proposed for monitoring exercise
activities within the household environment. This study compares the
effectiveness of Machine Learning algorithms, specifically K- Nearest
Neighbor (KNN) and Random Forest, in predicting exercise poses.
The dataset used consists of a total of 15,372 images merged into a
single video frame. Subsequently, training was conducted to identify
body landmarks. The dataset was divided into two variables, labeled
C and L. Variable C contains 8,238 images representing correct
exercise poses. In contrast, variable L contains 7,134 images
illustrating incorrect exercise poses. The results indicate that the
Random Forest model exhibited the highest accuracy during the
training phase at 99%. Conversely, the KNN model demonstrated the
highest accuracy at 97% during evaluation. This difference can be
attributed to various factors, including model complexity.
Index Terms—Machine Learning K-Nearest Neighbor, Random

Forest, Landmark, Accuracy

Paper ID 086
Implementation of 13.56 MHz Crystals in Oscillators for Wireless

Power Charging Applications
Fahmi Fathurrahiman1, Azwar Mudzakkir Ridwan1, Aan Eko Setiawan1,

Zenal Aripin2, Nurul Fahmi Arief Hakim3
1
2
Telecommunication & Microwave Lab.
Dept. of Electrical Eng. Education
3
Bandung, Indonesia
1197070022@student.uinsgd.ac.id,
azwarmudzakkirridwan@uinsgd.ac.id, aaneko37@hmail.com,
Azka3103@gmail.com, nurulfahmi@upi.edu
Abstract —This research analyzes the performance improvement of

a 13.56 MHz crystal oscillator for wireless power charging
applications. The circuit used is a pierce oscillator circuit with CMOS
74HC04N, amplified by a common collector amplifier. The method is
to adjust the resistor and capacitor values and use other transistors.
In its manufacture, capacitors made from ceramic and mylar are
used to analyze the output. The performance results of the crystal
oscillator using a ceramic capacitor produce a signal with a strength
of 10 dBm and a peak amplitude of 3.16V, while the crystal oscillator
using a mylar capacitor produces a signal with a strength of -2 dBm
and a peak amplitude of 2.22V at a frequency of 13.56 MHz. These
results show that the crystal oscillator using ceramic capacitors
performs better than the crystal oscillator using mylar capacitors.
Keywords—common collector amplifier, oscillator, wireless power

charging

Paper ID 089
Text Classification Using NLP by Comparing LSTM and Machine

Learning Method
Zaidan Muhammad Mahdi1, Retno Fauziah Istiqomah2, Alfarelzi1, Sri

Astuti1, Ibnu Asror2, Ratna Mayasari1
1
School of Electrical Engineering
2
School of Computing
Telkom University
Bandung, Indonesia
zaidanmahdi@student.telkomuniversity.ac.id,
retnofauziahiii@student.telkomuniversi ty.ac.id,
farelzii@student.telkomuniversity.ac.id,
sriastuti@telkomuniversity.ac.id, iasror@telkomuniversity.ac.id,
ratnamayasari@telkomuniversity.ac.id
Abstract— Natural Language Processing (NLP) has seen significant

advancements recently, leading to various applications across
different domains. This research focuses on text classification by
comparing the performance of Long ShortTerm Memory (LSTM)
networks with several traditional machine learning algorithms,
including Naive Bayes, Random Forest, Support Vector Machine
(SVM), K-Nearest Neighbors (KNN), and XGBoost. The novelty of
this study lies in its comprehensive comparative analysis using a
balanced dataset. The data balancing was achieved through the
Synthetic Minority Over-sampling Technique (SMOTE), ensuring
robust model training. Experimental results reveal that the SVM
algorithm achieves the highest accuracy of over 96%, surpassing
other models in performance. This indicates that despite the
advancements in deep learning, traditional algorithms like SVM
remain highly effective for text classification tasks. The findings
provide valuable insights into the strengths and weaknesses of
different NLP approaches, contributing to the ongoing development
of more accurate and efficient text classification models. Future work
could explore hybrid models and it application to diverse datasets to
further enhance classification accuracy.
Keywords—Natural Language Processing, LSTM, Text Classification,

machine learning, SVM.

Paper ID 090
Design of 13.56 MHz Transmitter Coil using Miniaturization

Technique with Planar Inverted F Structure and Defected Ground
Structure
Hafidz Fadilah1, Azwar Mudzakkir Ridwan1, Aan Eko Setiawan1,

Ahmad Saepul Milah1, Zenal Aripin3, M. Idham Kurniawan2
1
2
Faculty of Da’wah and Communication
3
Telecommunication & Microwave Lab.
Bandung, Indonesia
hafidzzfad@gmail.com, azwarmudzakkirridwan@uinsgd.ac.id,
aaneko37@hmail.com, ahmadsaefulmilah56@gmail.com,
Azka3103@gmail.com, muh.idhamkurniawan@uinsgd.ac.id
Abstract—This research designs a transmitter coil with a spiral

meandering technique for wireless charging applications. The
proposed transmitter coil is designed on an FR4 substrate with a
thickness of 1.6 mm, which can be applied at a frequency of 13.56
MHz. The design utilizes a Planar Inverted F Structure and Defected
Ground Structure (DGS) to improve the return loss performance and
reduce the size of the resulting transmitter coil. The applied method
results in a return loss value of -37.71 dB, and using the DGS
structure can reduce the transmitter coil size by 432 mm. Applying
Inverted F Structure and DGS results in a transmitter coil design of
230 mm x 230 mm.
Keywords—transmitter coil, meandering, spiral, wireless charging,

planar inverted F structure, defected ground structure, return loss

Paper ID 093
Monitoring Harvest Power System Using Piezoelectric Disk

Electric-Based IoT
Elisa Apriliyani1, Enjang A Juanda2, Azwar Mudzakkir Ridwan1, Aan

Eko Setiawan1, Nurul Fahmi Arief Hakim2, Raihan Nurhakim2
1
2
Dept. of Electrical Eng.Education
Bandung, Indonesia
elisaapriliani146@gmail.com, juanda@upi.edu,
azwarmudzakkirridwan@uinsgd.ac.id, aaneko37@gmail.com,
nurulfahmi@upi.edu, raihannurhakim@upi.edu
Abstract—Innovative technologies are being developed to harness

power from a variety of natural and mechanical. One such
prospective technology is the utilization of piezoelectric materials to
produce electricity. Piezoelectric energy harvesters can generate
electricity from various sources. The Internet of Things (IoT) is a
network of interconnected devices that are capable of accumulating,
exchanging, and processing data. In the context of energy harvesting,
IoT devices that are equipped with communication modules and
sensors can perpetually monitor the performance of piezoelectric
systems. The electrical voltage scale produced by piezoelectric
sensors is relatively small, namely 0.1 v - 1.459 v for 1 circuit. Then,
the electric voltage scale produced by piezoelectric sensors is 2.0 v -
2.866 v for two circuits. The last electric voltage scale piezoelectric
sensors produces is 3.0 v - 3.443 v for three circuits. From these
results, it can be said that giving vibration to the piezoelectric sensor
component will affect the output obtained; large and evenly
distributed vibrations can produce a greater voltage.
Keywords— Piezoelectric, Energy harvesting, IOT, Vibration

Paper ID 095
Tomato Harvesting Robot Prototype: Fuzzy-Controlled Arm with

Vision-Based Tomato Detection
Hardiansyah Firdaus, Rina Mardiati, Aan Eko Setiawan

Bandung, Indonesia
hardif66@gmail.com, r mardiati@uinsgd.ac.id, aaneko37@gmail.com
Abstract—Advances in technology support activities that are required to

facilitate humans to support human mobility, especially in agriculture. In this
study, a prototype tomato-picking robot arm based on a Pixy2 CMUcam5 to
detect the position of tomatoes is developed, utilizing fuzzy logic methods to
adjust the robot arm angles for precise tomato harvesting. The robot arm
detects ripe tomatoes and moves them to a predetermined location. This
prototype utilizes Arduino Uno, Pixy2 CMUcam5, Servo MG90S, Servo
SG90, jumper cables, a 5V power supply unit (PSU), and a laptop. The
software employed includes Arduino IDE, Matlab, and PixyMon. Due to the
requirement for precise movement carried out slowly, the Mamdani fuzzy
logic control method is adopted. The Mamdani fuzzy logic method is utilized
to achieve accurate results by considering various indicators as input. Test
parameters include the x-axis and y-axis values between the camera sensor
and the object. Simulations are conducted using a case study with an x value
of 158 and a y value of 153. Testing involves simulation software and the
Arduino IDE system, with manual calculations conducted for comparison.
Results from testing with simulation software yield values of 72.6 for the base
servo and 69.8 for the elbow servo. In the Arduino IDE software, values of 72
for the base servo and 69 for the elbow servo are obtained, while manual
calculations yield 72.7 for the base servo and 69.8 for the elbow servo. Testing
is also performed to evaluate the system’s efficiency using the fuzzy method
compared to without fuzzy logic, detecting objects in various x-axis (left,
center, and right) and y-axis (top, center, and bottom) positions in nine tests.
Results show that the system employing the fuzzy method is more efficient in
object detection. The Pixy camera functions effectively as a visual sensor, and
the servo moves according to angles set by the Mamdani fuzzy logic method.
Keywords—pixy camera, fuzzy logic control, mamdani, servo, arm robot, vision
sensor

Paper ID 104
LoRa Based Underwater Wireless Network System
Naufal Zaidan Nabhan, Joko Suryana

Radio Telecommunication and Microwave Laboratory School of
Electrical Engineering and Informatics
Bandung, Indonesia
23223018@mahasiswa.itb.ac.id,joko@itb.ac.id
Abstract—To apply the Internet of Things (IoT) underwater, a robust

protocol is needed to connect devices. LoRa, known for its long-range
and low-power characteristics, is a widely used protocol in IoT. Its
significance in underwater environments lies in enabling reliable data
transmission between sensors and Autonomous Underwater Vehicles
(AUVs), forming an IoT network. However, implementing LoRa
underwater is challenging due to high attenuation at its typical
operating frequency of 433 MHz. This paper contributes by testing
the performance of underwater communication systems at a
frequency of 40 MHz using a planar hexagonal loop antenna, which
experiences lower attenuation. Our method involves transmitting text
underwater via the LoRa protocol using microcontroller and LoRa
module, employing Down Conversion and Up Conversion techniques
to shift the frequency from 433 MHz to 40 MHz and vice versa, using
Software Defined Radio (SDR). The results demonstrate successful
reception and display of the transmitted text, indicating the potential
of the proposed system for effective underwater communication.
Index Terms—internet of things (IoT), radio communication,

nderwater communication, lora, and wireless network system.

Paper ID 015
Local Oscillator Phase Shifter for IQ Mixer in A Synthetic Aperture

Radar Application Based on Quadrature Hybrid Branch-line
Coupler
Abdurrasyid Ruhiyat1, Farohaji Kurniawan2, Catur Apriono1

1
Universitas Indonesia
Depok, Indonesia
2
Research Center for Aeronautics
Technology National Research and Innovation Agency
Bogor, Indonesia
abdurrasyid.ruhiyat@ui.ac.id, faro001@brin.go.id
catur@eng.ui.ac.id
Abstract—Synthetic Aperture Radar (SAR) is a powerful instrument

in microwave remote sensing applications because of its capability for
all-weather time operation and producing high-resolution images.
The IQ mixer is a vital component of the RF (radio frequency) SAR
subsystem, which increases the bandwidth of the transmitted signal
and improves the resolution of the SAR system. The IQ mixer is
composed of a quadrature LO phase shifter, two matched mixers,
and a power combiner-splitter. This paper proposes a design for a
quadrature LO phase shifter based on a microstrip quadrature
hybrid branch-line coupler. The coupler provides different good
phase outputs around 90◦. The output and input transmission lines
have been placed on the outer side of the coupler to offer a flat gain
over a range bandwidth (150 MHz), positively influencing the output
response. The fabricated coupler satisfies all the specification
requirements and can be used as an LO phase shifter.
Index Terms—coupler, IQ mixer, phase shifter, synthetic aperture

radar, local oscillator

Paper ID 016
Robust PI Controller Design of Electro-hydraulic Power Steering

System for Commercial Vehicles
Rina Ristiana1, Sunarto Kaleg1, Rina Mardiati2, Alexander Christantho

Budiman1, Abdul Hapid1, Muhammad Arjuna Putra Perdana1
1
Research Center of Transportation Technology
National Research and Innovation agency
2
West Java, Indonesia
rina008@brin.go.id 0000-0002-7842-1126, suna024@brin.go.id 0000-
0002-5800-8713, r_mardiati@uinsgd.ac.id 0000-0002-3949-7830,
alex003@brin.go.id 0000-0002-3034-4352, abdu019@brin.go.id 0000-
0002-0599-4206, marj001@brin.go.id
Abstract— A controller design has been developed that is applied to electro-

hydraulic power steering (EHPS) systems for commercial vehicles. The
EHPS system is built in three subsystems, namely mechanical, hydraulic, and
electric motor subsystems. An EHPS model was built based on these three
subsystems to evaluate, validate and determine the EHPS control design,
especially for EHPS motor speed control. The model takes into consideration
modifications in the input of the driver steering torque given to the steering
wheel angle and changes in the supply of the hydraulic pump pressure in the
piston cylinder, which produces a shift of the sector gear distance. The EHPS
model is separated into three parts: linearization mechanical, nonlinear
hydraulic, and linear electric motor. Based on the EHPS model, a PI
(Proportional-Integral) controller is designed to control the speed of the
EHPS motor by assuming the part 1 and 2 models are in numerical form, so
that the control design is only based on the EHPS part 3 model, resulting in
a simpler controller design that still takes into account important variables
in the EHPS system. The PWM (pulse width modulation) technique is used
to control the speed of the EHPS motor by specifically regulating the motor's
input voltage. The goal of this PI Controller design is to obtain optimal EHPS
characteristics from changes in steering wheel angle, pressure, and motor
speed due to differences in steering wheel torque input, which results in the
displacement of the ball nut, which explicitly provides movement to the
vehicle wheels, which is known as turning maneuver, and to ensure that the
steering feels comfortable at high speeds while also providing good steering
at low speeds.
Keywords: PI Controller, EHPS modelling, commercial vehicles, hydraulic

pressure, mechanical steering, BLDC motor

Paper ID 017
Estimated The Assist Map of The Electro-Hydraulic Power Steering

System for An Electric Vehicle Microbus
Muhammad Arjuna Putra Perdana1, Rina Ristiana1, Rina Mardiati2,

Sunarto Kaleg1, Aam Muharam1, Amin1
1
2
Department of Tarbiyah and Teacher
Bandung, Indonesia
marj001@brin.go.id, rina008@brin.go.id 0000-0002-7842-1126,
r_mardiati@uinsgd.ac.id 0000-0002-3949-7830, suna024@brin.go.id
0000-0002-5800-8713, aam.muharam@brin.go.id 0000-0001-8803-
7999, amin007@brin.go.id 0000-0001-8600-9054
Abstract—The assist map of the electro-hydraulic power steering

(EHPS) system depicts the power and portability rules of the steering
in the shape of a curve. The assist map characteristic determines the
assist torque value depending on the vehicle’s longitudinal speed,
vehicle lateral acceleration, EHPS motor speed, and steering wheel
torque. The assist map is required to ensure portability and
performance, as well as to recognize disturbances in the steering
system caused by variations in steering drive torque and vehicle
lateral acceleration. To obtain precise steering performance, an assist
map estimate is required, which utilize the extended Kalman Filter
approach. The assist map curve is produced while the vehicle is
stationary or at rest; however, variations in steering drive torque and
vehicle lateral acceleration may be observed and approximated to
determine the magnitude of the steering wheel angle, hydraulic
pressure, and EHPS motor speed. According to the assist map
concept, the assist torque rules may be determined based on changes
in vehicle lateral acceleration and EHPS motor speed, which are then
supplied to the EHPS system to support the driver with navigation,
particularly when the vehicle turns.
Keywords: Extended Kalman Filter, Assist Map Curve, EHPS system,

Assist Torque, Light Vehicle, Steering Navigating, Vehicle Turn
Maneuvers.


Electrical, Computer, & Biomedical Engineering

Paper ID 006
Dynamic Adaptive Video Streaming: A Comparative Examination

of NDN and IP Architectures in Real-world Scenarios
Ega Helmi Mubarok, Komang Maesa Dithanugraha, Hasan Nur Arifin,

Muhammad Alief Fauzan Bariadi, Aditya Arya Putranda, Nana
Rahmana Syambas, Intan Putri Maharani Prasetyaningrum, Fikry Adam
Bandung, Indonesia
18121002@telecom.stei.itb.ac.id, 18121031@telecom.stei.itb.ac.id,
hasanarif@telkomuniversity.ac.id, nana@stei.itb.ac.id
Abstract—The study compares the dynamic adaptive video streaming

performance of Named Data Networking (NDN) and Internet
Protocol (IP) in real-world scenarios. Key parameters include
bandwidth, Round-Trip Time (RTT), Retransmission TimeOut
(RTO), and video resolution. The evaluation aims to understand data
transmission efficiency, signal travel time, packet retransmission
timing, and overall video streaming quality. The study implies that
one architecture outperforms the other based on the analyzed
parameters, indicating that the superior architecture is Named Data
Networking (NDN). The research contributes valuable insights into
NDN and IP architectures' strengths and weaknesses in dynamic
adaptive video streaming applications, aiding in understanding their
comparative performance in practical situations.
Keywords— Named Data Networking, TCP/IP, Dynamic adaptive

streaming, Shaka Player, QOE

Paper ID 007
Performance Analysis of Forwarding Strategy in NDN-Based

Vehicle Networks
Zakariyya Gambetta Muhammad K1, Bagus Isa Ahmad1, Felisha Dita

Zafirah1, Nindya Dwi Larasati1, Naura Tiara Riesani1, Fildza Haniifa
Butsaina1, Syaiful Ahdan2, Eki Ahmad Zaki Hamidi2, Nana Rachmana
Syambas2
1
Dept. of Telecommunication Engineering
2
Bandung, Indonesia
syaifulahdan@teknokrat.ac.id, ekiahmadzaki@uinsgd.ac.id,
nanasyambas@itb.ac.id
Abstract—The growth of Named Data Networking (NDN) could lead

to the improvement of data storing efficiency in a high mobility
environment. Vehicular NDN (VNDN) is a vehicular ad-hoc network
that implements NDN architecture. A high mobility environment
needs an efficient forwarding strategy, which is why in this paper we
tested and compared the performances of four forwarding strategies;
Adaptive Smoothed RTT-based forwarding strategy (ASF),
Multicast VANET, Self-Learning forwarding strategy, and the Best
Route forwarding strategy. The performance is measured by its
number of Forwarding Pipelines activities. The result in this paper
stated that the ASF has the best performance than the other three due
to its higher number of satisfied interests and data sent, while having
the least amount of latency.
Keywords—Vehicular Named Data Networking, Named Data

Networking, NDN4IVC, VNDN Forwarding Strategy

Paper ID 009
Leveraging The Existing Sensor Signals on Komatsu D31P

Bulldozer Through IoT-Enabled Embedded System for Heavy
Equipment Unit
Aditya Kurniawan1, Kholilatul Wardani2, Eki Ahmad Zaki Hamidi3

1
Heavy Equipment Engineering Department
Akademi Teknik Alat Berat Indonesia
2
Telecommunication Engineering Department
Politeknik Kota Malang
Malang, Indonesia
3
Electrical Engineering Department
Bandung, Indonesia
aditya@akabi.ac.id, wardani@poltekom.ac.id
ekiahmadzaki@uinsgd.ac.id
Abstract—This research aimed to create an embedded system for a

IoT enabled device on Dozer Komatsu D31P by designing and
implementing NodeMCU-based controller on Dozer electrical circuit.
Choosing NodeMCU-based is a viable solution to enhance the
connectivity of raw signal of dozer system. The system incorporates
existing sensor on dozer such as engine oil pressure signal (EOP)
coded SAD00520, alternator charge signal (ACS) coded SAD00522,
air cleaner clogged signal (ACCS) coded SAD00997, water
temperature signal (WTS) coded SAP00527, fuel level signal (FLS)
coded SAP00528 and engine preheating signal (EPS) SAD00708. The
implementation of IoT enabled device therefore enhancing the unit
with remote monitoring and wireless data collection in real-time for
analysis, further optimizing equipment performance, leading to
enhanced efficiency in maintenance management system. Proactive
maintenance and predictive analytics could be applied in the future
to reduce downtime and enhance productivity. The research
demonstrated that the designed embedded system functioned
appropriately in accordance with standard. Furthermore, using the
IoT enabled embedded system, analysis on PA (Physical Availability)
will be available in real time.
Keywords—IoT enabled device, NodeMCU, Dozer D31P

Paper ID 010
Deep Imitation Learning Behavior Fusion Algorithm for Visual

Based Autonomous Driving Scenario
Handoko Supeno, Bambang Riyanto Trilaksono, Widyawardana

Adiprawita, Nur Ulfa Maulidevi
Bandung, Indonesia
33220026@std.stei.itb.ac.id, handoko@unpas.ac.id
bambang.riyanto@itb.ac.id, widyawardana@itb.ac.id
ulfa@itb.ac.id
Abstract—The decision and planning system for autonomous driving

in urban environments is hard to design. Current methods
predominantly rely on manually designed driving policies, which are
costly to develop and maintain. Imitation learning offers a promising
alternative by enabling systems to learn and improve driving policies
automatically through data collection. However, existing imitation
learning methods for autonomous driving are hardly perform well
for complex urban scenarios. In this paper, we introduce a
framework designed to efficiently learn driving policies in urban
scenarios. Instead of employing the traditional methods of designing
separate modules or training the agent on multiple tasks end-to-end,
our research adopts a novel approach. We propose training the agent
on specific behaviors per task and subsequently integrating these
behaviors, guided by camera-based environmental perception. We
conducted testing within dynamic traffic scenarios using the Carla
simulator to validate our framework. The experiments show that our
method can achieve high performance in various traffic density
scenarios, as well as a high level of comfort and safety.
Keywords— Autonomous agent, deep imitation learning, Behavior

Fusion

Paper ID 013
Geographic Information System-Based Recommendations For

Analyzing The Distribution Of Critical Land In The Citanduy River
Basin Using The Modified SWAT Method
Teguh Nurhadi Suharsono1, Rama Rosdiana2, Rina Mardiati3

1
Faculty of Engineering
2
Magister of Civil Engineering Department
Sangga Buana University
3
Bandung, Indonesia
teguh.nurhadi@usbypkp.ac.id,ramarosdiana@ymail.com,
Abstract—A watershed is a geographical area through which a river and its

tributaries flow, forming an interconnected network. It serves as a collection
point for rainwater and other sources of water, which then flow through the
tributaries and main river. The watersheds in most regions of Indonesia
experience a significant strain because to the combination of a dense
population and ongoing exploitation of natural resources. The escalating
occurrence of landslides, the accelerating rate of soil erosion and
sedimentation, as well as the occurrence of floods and droughts in various
places, are indications of the deteriorating state of the watershed. The study
employs an adapted SWAT (Soil and Water Assessment Tool) approach to
effectively address many issues in the Citanduy watershed, with a particular
focus on analyzing crucial land-related challenges. The utilization of GIS for
mapping the distribution of crucial land in the Citanduy watershed has not
been extensive. This study intends to assess the criticality of the Citanduy
watershed in achieving sustainable watershed management through
mapping. The final outcome of each map is overlaid to accurately determine
the extent of crucial land and create a standardized map that may serve as a
reference for the government in policy-making. The research findings
indicate that 50.01% of the land, covering an area of 1820.484 Km2,
experienced a high rate of land erosion. Additionally, 48.71% of the land,
spanning 1773.14 Km2, had a medium land erosion rate. The remaining
1.28% of the land, which covered an area of 46.61 Km2, had a low land
erosion rate. The average land erosion in the Citanduy River Basin is 160,566
tons per hectare per year.
Keywords—The Citanduy River Basin, critical land, GIS, The Modified SWAT
Method

Paper ID 014
Exploring Deep Learning Models for Air Quality Forecasting: A

Comparative Review
Aang Aliyudin1, Aradea1, Rianto1, Rina Mardiati2

1
Departement of Informatics
Siliwangi University,Tasikmalaya, Indonesia
2
Bandung, Indonesia
aliyudinaang@gmail.com, aradea@unsil.ac.id, rianto@unsil.ac.id,
Abstract—Burgeoning global attention to the problem of increasing air

pollution has become an urgent focus, particularly in terms of its significant
implications for the sustainability of human health and the ecosystem.
Economic growth and urbanization in metropolitan cities have raised energy
consumption escalating the level of air pollutant emissions. This phenomenon
is particularly manifested in increased air pollution in urban environments
indicating a noteworthy negative impact on public health in a relatively short
period. The main parameters adopted as air quality indicators are sulfur
dioxide (𝐒𝐒𝐒𝐒𝐒𝐒), carbon monoxide (CO), particulate matter with a diameter of
less than 10 micrometers 𝐏𝐏𝐏𝐏𝐏𝐏𝐏𝐏 , particulate matter 2.5 micrometers (𝐏𝐏𝐏𝐏𝐏𝐏.𝟓𝟓),
ozone (𝐎𝐎𝐎𝐎), and nitrogen dioxide ( 𝐍𝐍𝐍𝐍𝐍𝐍) . The parameters signify salient
variations in the types and sources of air pollutants, including contributions
from transportation, industries, and human activities. The approach of
adopting deep learning to predict air quality is one of the innovations to
maintain human health. Modeling using deep learning is very sensitive to the
employment of hyperparameters becoming a challenge when building models.
Multi-index of supervised learning (MMSL) based on Long Short Term-
Memory (LSTM) is one of the optimal deep learning approaches in
predicting air quality components. Modeling applies MMSL to predict with
hyperparameters, namely batch size of 2^9, hidden layer used is 1 and
optimization adopts Adam with 25 epochs to predict 𝐒𝐒𝐒𝐒𝐒𝐒, CO and 𝐎𝐎𝐎𝐎.
Utilizing a model with the right hyperparameters produces an optimal model
contributing to maintaining health from air pollution.
Keywords—Deep Learning, Hyperparameter, Air Pollutione

Paper ID 042
Low-Cost Stereo Vision System Using ESP32-CAM and YOLO
Danish Muhammad Hafidz∗ , Daven Darmawan Sendjaya† ,

Muhammad Ogin Hasanuddin‡ , Pranoto Hidaya Rusmin§
Bandung, Indonesia
Email: { ∗13221080, †13221062}@mahasiswa.itb.ac.id, { ‡moginh,
§pranoto.hr}@itb.ac.id
Abstract—Today, distance estimation algorithm (DEA) is very

important to enable machines to map the world and interact with it,
independent of human controls. These algorithms are beneficial for
various usages, such as the ADAS or Humanoid Robots, to control
the movements of each devices. In order to implement DEA, we can
integrate the concept of stereo camera with ESP32-CAM connectivity
capability to implement distance estimation algorithm in three-
dimensional space, by simulating how human eyes work. The
objectives of this experiment are to build a Stereo Camera with
ESP32-CAM module, design a Distance Estimation Algorithm (DEA),
implement DEA to estimate the distance of objects in three-
dimensional space, and measure the error and processing delay
generated by DEA. The stereo camera implementation stage involves
designing the device with ESP32-CAM, uploading the camera web
server program, and calculating camera parameters. Distance
Estimation Algorithm (DEA) algorithm is executed using Python and
YOLO to analyze images from both camera eyes. DEA provides
optimal performance for objects with distances below 7 meters after
calibration, the calculation of object distances above 7 meters
produces significant errors. The DEA implementation also shows
limitations in real-time, taking 0.1 - 2 seconds from ESP32-CAM
video recording to object calculation results due to delays from the
internet network and computer processing limitations.
Keywords—Distance Estimation Algorithm, Stereo Camera, ESP32-

CAM, YOLO

Paper ID 048
Fisher-Yates Approach for Daily Qur'an Chatbot
Syahrul Safarudin Hasan1, Wildan Budiawan Zulfikar1, Nur Lukman3,

Agung Wahana2, Aldy Rialdy Atmadja1
1
Department of Informatics UIN Sunan Gunung Djati
Bandung, Indonesia
2
Department of Computer Science IPB University
Bogor, Indonesia
3
Department of Information Communication Technology
Asia e University
Kualalumpur, Malaysia
syahrul.safarudinh9@gmail.com, wildan.b@uinsgd.ac.id,
c70109230012@aeu.edu.my, wahana.agung@apps.ipb.ac.id,
aldyrialdy@uinsgd.ac.id
Abstract—The Al-Qur'an was once only in the form of a book, now it

is available in the form of a cellphone application and is easy to access,
then there is also this Al-Qur'an brought to social media, but from
what the author sees, especially from a telegram platform, most of
them are no longer active and only a few have Indonesian translations,
while based on a report from We Are Social which always presents
data on internet trends, social media and others mention that active
social media users in Indonesia from 2020 to January 2022 increased
from 160 million people in 2020 increased by 19% to 191 million
people from 277.7 million people in January 2022, so the author
thought of bringing back this Al-Qur'an on the telegram platform
using technology in the form of telegram bots to bring the Al-Qur'an
to the Telegram social media platform. FisherYates is used to
generate random permutations of a finite set. The randomization
process does not provide the possibility of repetition because in
determining the next random variable it will not produce the same
number until there are no variables left. The result of this research is
that fisher yates can be used in randomizing letters and verses of the
Qur'an and the randomization results do not cause repeated numbers,
and scheduller system regularly so that the Qur'an verses can be
delivered to users who use bots.
Keywords – Al-Quran, Scheduller, Fisher Yates, Telegram, Chatbo

Paper ID 060
Systematic Review on Interpretability in Computer Aided Alcohol

Use Disorder Diagnosis
Nur Zahrati Janah, Adhistya Erna Permanasari, Noor Akhmad Setiawan

Department of Electrical and Information Engineering
Universitas Gadjah Mada
Yogyakarta, Indonesia
nur.zahrati@mail.ugm.ac.id, adhistya@ugm.ac.id,
noorwewe@ugm.ac.id
Abstract—Numerous studies have investigated the detection of AUD

using EEG signals. However, the real-world implementation of
diagnostic systems is hindered by users’ difficulty in understanding
the rationale behind a decision. This study presents a systematic
literature review based on PRISMA guidelines, utilizing journal and
conference articles from 2019 to 2023 obtained from Scopus. Among
66 collected studies, feature engineering techniques dominated by
EEG sub band decomposition processes, followed by time-domain
feature extraction, EEG conversion to images, and brain connectivity
features. With advancements in computation, comprehensive hybrid
features and raw EEG data are emerging as inputs for classifiers.
Some studies also use dimensionality reduction techniques such as
statistical analysis, optimization methods, feature reduction methods,
and manual selection. Moreover, while black box classifier models
are widely used, intrinsically interpretable classifiers exhibit no
significant differences in accuracy performance. Future work aims to
develop AUD detection systems with enhanced interpretability while
maintaining robust classification performance, utilizing features
informed by experts’ domain knowledge to facilitate decision
explanation.
Keywords—alcoholism, alcohol use disorder, computer-aided diagnosis,

machine learning, interpretability.

Paper ID 062
Enhancing Electricity Usage Monitoring and Prediction with IoT-

based Multiple Linear Regression Algorithm
Fahni Amsyari1, Dian Sa’adillah1 Maylawati1, Agung Wahana1, Wisnu

Uriawan1, Nur Lukman1, Hidayatul Fikra2
1
2
Magister of Psychology Universitas Padjajaran
Bandung, Indonesia
fahniamsyari.work@gmail.com
diansm@uinsgd.ac.id https://orcid.org/0000-0002-1193-3370
wahana.agung@uinsgd.ac.id https://orcid.org/0000-0002-6468-0741
wisnu_u@uinsgd.ac.id https://orcid.org/0000-0001-6922-4705
n.lukman@uinsgd.ac.id https://orcid.org/0000-0003-2674-6460
fikraarza2903@gmail.com https://orcid.org/0000-0003-3927-2305
Abstract—Throughout 2021, the number of IoT users in Indonesia

will be greater than that of connected smartphone users. It is
estimated that the number of IoT device users in 2025 will reach 658
million devices after the arrival of 5G services. With an IoT-based
electricity usage monitoring system, users can monitor electricity
consumption in real-time and get more accurate information
regarding energy usage trends. Apart from that, the use of machine
learning algorithms that study data to obtain predicted results in the
future is useful for knowing the electricity consumption patterns used.
Therefore, the aim of this research is to monitor and predict
electricity use by combining machine learning algorithms, namely
multiple linear regression with the least squares method and IoT
technology. The research results show that the multiple linear
regression algorithm in predicting electricity usage has good
performance with an error tolerance of 5%, an RMSE value of 0.0493,
and an average prediction accuracy of 91.98%. Apart from that, the
performance of monitoring electricity usage using IoT has an
accuracy value of 98.07% with a data transmission time tolerance of
less than 3 seconds.
Keywords— electricity monitoring, electricity usage, internet of things,

linear regression, machine learning

Paper ID 063
Application of Convolutional Neural Network Algorithm for

Analyzing Sentiments on the Kampus Merdeka Policy
Mohamad Irfan1, Theo Vectra Riyadi1, Aldy Rialdy Atmadja1, Rifqi

Syamsul Fuadi1, Abdul Muin2
1
Bandung, Indonesia
2
UIN Sultan Maulana Hasanuddin
Banten, Indonesia
irfan.bahaf@uinsgd.ac.id, vectrar.theo@gmail.com
aldyrialdy@uinsgd.ac.id, rifqi@uinsgd.ac.id,
abdul.muin@uinbanten.ac.id
Abstract—Sentiment analysis examines public opinions on the

Kampus Merdeka policy by analyzing texts from various sources.
The study follows the Cross Industry Standard Process for Data
Mining (CRISP-DM) method, encompassing stages such as business
understanding, data understanding, data preprocessing, model
implementation, and evaluation. The study utilizes preprocessing
techniques, such as converting emoticons and emojis, text filtering,
removing stopwords, stemming, word normalization, tokenization,
and sequencing. The data for analysis is sourced from Twitter and
YouTube, comprising 428 datasets. The accuracy, which measures
the similarity between predicted and actual values, is 76%.
Additional tests demonstrate that incorporating emoticon and emoji
conversions in the text can increase sentiment analysis accuracy by
5%, resulting in 81%. These findings indicate the effectiveness of the
Convolutional Neural Network algorithm employed in this research.
Keywords— Convolutional Neural Network, Converting emoticons and

emojis, Kampus Merdeka, Sentiment Analysis, Twitter, Youtube

Paper ID 064
Performance Comparison of Vedic and Karatsuba Algorithm for

64-Bit Barrett Modular Multiplication
Handy Jonarta, Rafael Aditya Cahyo W., Muhammad Ogin Hasanuddin,

Infall Syafalni, Nana Sutisna, Wildan Trusaji, Trio Adiono
Bandung, Indonesia
Email: {13221070, 13221066}@std.stei.itb.ac.id, moginh@itb.ac.id
Abstract—Modular multiplication finds application in various fields

that rely on cryptographic and finite-field arithmetic. Basically,
modular multiplication are consist of two main parts namely integer
multiplication, and modular reduction. In this paper we examine two
of well known integer multiplication namely, vedic and karatsuba for
integer multiplication, and we use barrett reduction for modular
reduction part. For comparison we use 64-bit as parameter size for
modulus bit size, and we implement using target board alveo 250. As
the result show that our implementation of the Vedic-based modular
multiplier demonstrates a 23.08% improvement in area usage
compared to the Karatsuba method, while the Karatsuba-based
approach achieves a 29.41% improvement in efficiency. These results
highlight the potential for optimizing modular multiplication in
cryptographic applications by selecting appropriate algorithms
based on specific performance requirements.
Index Terms—Karatsuba Multiplication, Vedic Multiplication, Barrett

Reductio

Paper ID 071
Implementation Of Convolutional Neural Network (CNN) in The

Android-Based Application for Detecting Coffee Bean Maturity
Eka Imamul Muhlisin1, Rin Rin Nurmalasari1, Lia Kamelia1, Ramdani

Wahyu Sururie2
1
2
Sharia and Law Faculty
Bandung, Indonesia
1197070042@student.uinsgd.ac.id, id rinrinnm@uinsgd.ac.id,
lia.kamelia@uinsgd.ac.id,ramdani.wahyu@uinsgd.ac.id
Abstract—The research aims to explore the implementation of

artificial intelligence in developing an application that can accurately
detect the maturity level of roasted coffee beans solely using a
smartphone device. The research employs an AIbased approach by
utilizing deep learning techniques to train a model on a dataset
containing various roasted coffee bean maturity levels. The training
data comprises visual images of coffee beans at different stages of
maturity, enabling the model to discern patterns and characteristics
associated with maturity levels. The application is built upon a
classification system model for the maturity level of roasted coffee
beans, employing an algorithm or convolutional neural network
(CNN) method using the ResNet50 architecture. The utilization of the
CNN method can enhance the accuracy of the model. This system is
developed by employing digital image processing with data
represented as coffee bean images. The smartphone application
system uses Visual Studio Code, Flutter, and Android Studio
software, incorporating image processing, colour feature extraction,
and CNN classification methods. The model evaluation demonstrates
an accuracy level of 100%, with precision, recall, and F1-Score values
also reaching 100%. However, the accuracy of the ResNet50 model in
classifying the maturity level of roasted coffee beans in real-time
applications, based on the confusion matrix, is approximately 83.3%.
Keywords— Coffee beans, Artificial intelligence, smartphone, deep

learning.

Paper ID 079
Overfitting Defeat with Dropout for Image Classification in

Convolutional Neural Networks

Bandung, Indonesia
33222024@mahasiswa.itb.ac.id, hend@itb.ac.id, iskandar@itb.ac.id
Abstract—Data is a crucial part of machine learning, used to create

models whose patterns are defined with specific algorithms. The
primary purpose of the machine learning process is to generate the
best prediction models. However, in general, models that trained well
in the training phase differ when used in the testing phase. This
phenomenon is called overfitting, a common problem in machine
learning, mainly when developing image classification models using
Convolutional Neural Networks (CNNs). The main task of the output
layer in CNNs is to classify the truth labels from the data input images.
One machine learning performance is defined by how well the models
can predict the truth labels. Overfitting is one of the problems that
can influence the model's performance. Dropout is one of the
regularization techniques intended to solve this problem. Our
purpose in this research is to know the relationship between the use
of dropout and CNN's performance in image classification from the
point of view of overfitting problems. The experimental results stated
that this technique can alleviate overfitting by 35% with a learning
rate value 0.001.
Keywords—Overfitting, Convolutional Neural Networks, Dropout,

Learning rate, Image Classification

Paper ID 081
Implementation of the Hybrid Base ABR Algorithm in Named Data

Networking Based Video Streaming System
Hasan Nur Arifin, Nana Rachmana Syambas, Tutun Juhana

Bandung, Indonesia
33222024@mahasiswa.itb.ac.id, nana@itb.ac.id, tutun@itb.ac.id
Abstract—This study aims to implement the Hybrid Base algorithm

in video streaming based on Named Data Networking (NDN) with a
focus on improving performance and user experience (Quality of
Experience). The Hybrid Base algorithm combines the advantages of
Buffer Base and Throughput Base algorithms to provide optimal and
adaptive video playback quality, striving to minimize delay and
maximize bandwidth utilization. This approach allows the algorithm
to adjust video bitrate based on user buffer conditions and current
network throughput. Through an experimental approach, this study
compares the performance of the Hybrid Base algorithm with
existing Adaptive Bit Rate (ABR) algorithms, showing superiority in
terms of video quality, network performance, and playback
performance for the Hybrid Base algorithm.
Keywords— Hybrid Base Algorithm, Video Streaming, Named Data

Networking

Paper ID 088
Semantic Segmentation of Iris from Pterygium for Biometric

Process using U-Net
Ambhara Putri Rie Kikhaawa1, Khoerun Nisa Syaja’ah1, Yudha Satya

Perkasa1, Rin Rin Nurmalasari2, Wei-Hua Yang3
1
2
Bandung, Indonesia
3
Shenzhen Eye Institute
Shenzhen Eye Hospital
Guangdong Province, China
rkikhaa@gmail.com, nisasya@uinsgd.ac.id, yudha@uinsgd.ac.id,
rinrinnm@uinsgd.ac.id, Guangdong Province, China
benben0606@139.com
Abstract— There is an alternative advanced identification system

called biometrics. Biometrics uses unique human body parts to
identify a person. One part of the human body considered unique and
can be utilized by a biometric system is the iris. The use of biometrics
on the eye's iris is also known as the Iris Recognition System (IRS).
Segmentation is one of the stages of the IRS that is very influential in
the accuracy of individual identification. The segmentation stage also
has challenges, such as pterygium tissue occlusion in pterygium
sufferers. In this research, segmentation was carried out for these
challenges using the U-Net method. U-Net is a method that can be
used for biomedical image segmentation tasks because it only
requires a small amount of label data and a very reasonable time for
model training. Segmentation of pterygium sufferers shows
subjectively inaccurate prediction image results at any epoch.
However, the evaluation metrics show increasingly better results as
the epoch increases. The best results were obtained by epoch 300,
namely an IoU Coeff of 0.96 and a Dice Coeff of 0.92.
Keywords— Segmentation, U-Net, Iris Recognition System (IRS),

biometrics.

Paper ID 096
Qibla-Finding Robot based on Vector and Fuzzy Logic Methods

for Accurate Directional Alignment
Najin Khoerus Syakirin, Rina Mardiati, Aan Eko Setiawan

Bandung, Indonesia
najinkhoerussyakirin@gmail.com, r mardiati@uinsgd.ac.id,
aaneko37@gmail.com
Abstract—Several tools have been created that use spherical trigonometric

formula as a method to identify Qibla direction. However, since it involves
many angles in trigonometry on the surface of the sphere, the spherical
trigonometry formula is difficult to calculate. spherical trigonometric
formula is difficult to calculate. An alternative approach to improvised
formula for determining the Qibla direction has been using vector algebra
(VA). To use vector algebra for Qibla direction, a prototype Qibla direction
robot is being developed using a vector approach and fuzzy logic. is being
developed by using the vector approach and fuzzy logic control. The robot’s
microcontroller receives location data from global positioning system (GPS)
that it carries. After encoded into a user-friendly format, the location data is
is processed using the angle formula between two vectors to obtain theta.
Theta is the angle obtained from the point of the gps coordinate point vector
with the north pole(⃗) and the gps coordinate point vector with the kabah(⃗u).
After getting theta, the Qibla direction is 360-theta. Obtained Qibla at 293.07
degrees, there is a difference of 2.13 degrees with the Qibla determined by
KEMENAG. After determining the Qibla direction, a fuzzy logic control
model is created with 2 inputs and an output. The inputs given are theta and
360-theta, for input using a calibrated Gy-271 compas module. The robot has
a stepper motor with a speed that has been set by fuzzy logic, the speed that
has been set will make the robot stop at 293 degrees to 295 degrees (Qibla
according to KEMENAG). Setting the speed on the stepper motor aims so
that there is no excess motor rotation so that the robot will be precise.
Keywords—vector, fuzzy logic control, mamdani, motor stepper, Qibla-Finding

Robot, Gy-271

Paper ID 103
Intelligence Malware Detection System with LinearSVC
Zahrina Maryam, Hendrawan, Setia Juli Irzal Ismail

Bandung, Indonesia
23220028@mahasiswa.itb.ac.id, hend@itb.ac.id,
Abstract—Malware, a form of malicious software, is engineered to

disrupt computer functionality, exfiltrate sensitive information,
surveil user activity, and perpetrate other illicit malicious actions. It
continues to proliferate daily. This escalation poses significant
challenges to the field of malware analysis and detection.
Consequently, the development of automated malware classification
systems becomes imperative. Machine learning emerges as a
promising approach for this purpose. This research proposes a
malware classification system employing the Support Vector
Machine (SVM) algorithm, with a Linear SVC approach, evaluated
using the EMBER dataset. The proposed system achieves a highest
accuracy of 98.9% with a total time of 59 seconds.
Keywords—Malware, EMBER, Machine learning, SVM, LinearSVC

Paper ID 105
Synchronization Technique for Orthogonal Time Frequency Space

(OTFS)
Bella Wahmilyana Asril, E.Y. Hamid

Bandung, Indonesia
23222027@std.stei.itb.ac.id, effrina@itb.ac.id
Abstract—Synchronization in Orthogonal Time Frequency Space

(OTFS) communication systems faces particular challenges,
especially in high mobility wireless environments. Carrier Frequency
Offset (CFO) is one of the primary issues that can disrupt OTFS
performance. The main objective of this study is to develop an
effective synchronization mechanism for precise CFO estimation to
ensure robust signal reception at the receiver. The proposed
methodology involves two synchronization methods: using Cyclic
Prefix (CP) and preambles with two different schemes, namely
Zadoff-Chu (ZC) sequence and IEEE 802.11a standard. The tests
were conducted by transmitting image data and performing
synchronization before OTFS demodulation. The results indicate
that the IEEE standard preamble method provides the best image
quality with the lowest Bit Error Rate (BER), followed by the ZC
sequence and CP methods.
Keywords—bit error rate, carrier frequency offset, cyclic prefix,

preamble, synchronization

Paper ID 106
Comprehensive Risk Evaluation Model for Data Center Security

Risk Assessment
Sayed Shoaib Hussaini, Budi Raharjo

Management Engineering and information Security
Institute of Technology Bandung
Bandung, Indonesia
23222701@std.stei.itb.ac.id, br@itb.ac.id
Abstract—Data centers play a crucial role in modern organizations

by storing and processing large volumes of sensitive information.
However, the evolving threat landscape and the increasing
sophistication of cyberattacks pose significant security challenges to
data centers. To ensure the integrity and availability of critical data,
this research endeavors to address this pressing need by proposing a
cuttingedge Comprehensive Risk Evaluation Model (CREM). By
systematically assessing and mitigating the various dimensions of risk
associated with data centers, CREM aims to redefines risk
assessment through its dynamic, iterative framework, where ongoing
recursive loops constantly find threats, examine events, and assess
vulnerabilities. The study involves a comprehensive literature review,
basic related concepts, methodology and development of risk
evaluation model architecture that addresses the identified gaps. The
model incorporates modules for threat source identification, threat
event analysis, risk assessment, vulnerability scanning, risk impact
analysis, threat level identification and whole control evaluation
process. This research aims to enhance security practices, enabling
proactive risk identification and mitigation. The findings contribute
to the field of data center security and offer practical guidelines for
improving security posture.
Keywords– Data Center, Risk Evaluation Model, Risk Assessment, Risk

Identification, Risk Analysis, Risk Evaluation.


Poster Session

Paper ID 001
User Clustering Using Modified K-means in PD-NOMA
Rizki Nugraha1, Iskandar2, Hurianti Vidyaningtyas3,4

1
Telecommunication Engineering
2
Radio Telecommunication and Microwave Laboratory
3
4
School of Electrical Engineering, Telkom University
Bandung, Indonesia
18119035@telecom.stei.itb.ac.id, iskandar@stei.itb.ac.id,
Abstract—Power Domain Non-Orthogonal Multiple Access (PD NOMA) is

one of the many multiple access techniques applied to cellular
communications to increase spectrum efficiency and throughput like sum-
rate. In this technique, limited resources, for example, power and frequency,
are allocated to several user devices (user equipment or UEs) in one cell in a
non-orthogonal manner. A user clustering concept is required, where user
devices are clustered or configured to increase the number of system levels.
The aim of this final project is to produce a clustering algorithm or system
that can increase the sum rate in PD-NOMA. This task is carried out using
the K-means algorithm which will then be carried out in two types of
modifications so that it will produce two different algorithms. Then, from the
two algorithms, each sum rate will be calculated and will be compared with
the sum rate of the conventional NOMA algorithm and also compared with
the sum rate of the Orthogonal Multiple Access (OMA) multiple access
technique which in this case uses TDMA. We proposed 2 modified algorithms
to cluster user. The first one called modified-silhouette K-means and the
second one is near-far K-means. The results show that the first modified
algorithm is able to show a sum-rate value that is higher than the sum-rate
of OMA and conventional NOMA. Meanwhile, the second one unfortunately
shows a sum-rate value that is lower than NOMA's conventional sum rate.
However, both are still higher than the OMA number. The conclusion of this
paper is that, with the right clustering algorithm, like our proposed
algorithms in this paper, it is possible to increase the sum-rate on PD-NOMA
which is better than the sum-rate on OMA so that it is very suitable for use
on 5G.
Keywords—user clustering, NOMA, sum-rate, K-means.

Paper ID 002
Development and Deployment of Web Application For User

Clustering Using Modified K-Means On PD-NOMA
Putra Rafii Pradana Santoso2, Iskandar1, Hurianti Vidyaningtyas2,3

1
Radio Telecommunication and Microwave Laboratory
2
3
School of Electrical Engineering, Telkom University
Bandung, Indonesia
18119036@telecom.stei.itb.ac.id, iskandar@stei.itb.ac.id,
Abstract— Power Domain Non-Orthogonal Multiple Access (PD-NOMA) is

a multiple access technique that supports the system to achieve ultra-low
latency and ultra-high connectivity in 5G cellular communications. NOMA
has a drawback, if imperfect successive interference cancellation (SIC)
occurs, error happened when deciding to reduce the wrong modulated signal,
it will produce residual interference. A user clustering system is needed so
that it can prevent errors and interference that ultimately increase the sum-
rate value. Modified k-means clustering is used as the clustering model. In
the experiment, two modified k-means models are proposed by using
silhouette coefficient as a determinant of k value, the first model uses
optimum distance and the second model uses a combination of k-means
clustering with near-far pairing. Furthermore, it will be compared with
TDMA and near-far pairing. The results show that the first modified k-
means model, with optimum distance, produces a better sum-rate than other
clustering models. In line with this, a web application was also built which is
used as a place to visualize and deliver information on the results of the
analysis of the user clustering process using modified kmeans. Users can
provide input in the form of user coordinate positions to find out how the
process works and the analysis results generated from the program. The web
application uses the streamlit framework and is deployed using the streamlit
community cloud and google cloud. There are two main features that can be
used, file uploads and random generators, from both of which user position
data will be obtained which will later be processed in the program. Users can
also download reports on the output of the clustering process.
Keywords—NOMA, k-means, user clustering, web application.

Paper ID 008
Load Balancing Algorithm Application on Named Data

Networking
Syah Muafa Al Kautsar1, Albert1, Gatra Akhira1, Muhammad Arrifqi1,

Achmad Kabir Rifai1, Geo Perdana Saragih1, Ade Nurhayati1,2, Galih
Nugraha Nurkahfi1, Nana Rachmana Syambas1
1
Bandung, Indonesia
2
Faculty of Engineering Telkom University
Jakarta Barat, Indonesia
adenurhayati@telkomuniversity.ac.id,
galih.nugraha.nurkahfi@lipi.go.id, nana@stei.ee.itb.ac.id
Abstract—Load balancing is a technique used to distribute traffic across

multiple servers. This can help to improve performance and reliability by
reducing the load on any individual server. In this paper, we discuss the
application of load balancing algorithms in NDN. NDN is a new network
architecture that is designed to be more efficient and scalable than the
traditional TCP/IP architecture. We discuss three common load balancing
algorithms: weighted round robin, weighted least count, and least response
time. We describe how these algorithms can be implemented in NDN, and we
discuss their advantages and disadvantages. We also discuss the challenges
and limitations of load balancing in NDN. One challenge is that NDN is a
content-centric architecture, which can make it difficult to track the load on
each server. Another challenge is that NDN uses a name-based forwarding
scheme, which can make it difficult to use traditional load balancing
algorithms. Finally, we discuss the future prospects and developments of load
balancing algorithms for NDN. We discuss the potential of using machine
learning to improve load balancing performance, and we discuss the
potential of using a combination of multiple load balancing algorithms.
Keywords—named data networking, load balancing, weighted round robin
algorithm, mini-NDN

Paper ID 018
Comparative Evaluation of Forwarding Strategies in NDN-

Simulated SEA-ME-WE Network Topology
Patricia Adventia Pratama1, Bevan Julio Krisna1, Nana Rachmana

Syambas1, Samuel Tommy Setiadjie1, Rima Saefiana1, Ratna Mayasari2,
Dennys Yudhistira Hutapea1, Daru Kristian Aji1, Ridha Muldina
Negara2
1
2
Telecommunication Engineering School of Electrical Engineering
Telkom University, Indonesia
nanasyambas@itb.ac.id, 18121023@telecom.stei.itb.ac.id
18121041@telecom.stei.itb.ac.id,ratnamayasari@telkomuniversity.ac.id
18121029@telecom.stei.itb.ac.id,18121053@telecom.stei.itb.ac.id
ridhanegara@telkomuniversity.ac.id
Abstract—Named Data Networking (NDN) is a proposed solution for

the future internet, where the internet is shifting from a host-based
model into an information-centric model. Forwarding strategies are
a method of how a node forward the received interest packet in NDN.
This research compares established forwarding strategies in ndnSIM
on NDNsimulated Submarine Optic Fiber Network, the SEA-ME-
WE topology. We assess the performance of different forwarding
strategies regarding transmission overhead, average total delay,
cache hit ratio, and cache eviction rate. Based on our research, the
most optimal forwarding strategies used in SEAME-WE simulated in
ndn is Best-route with the best output based on several performance
matrices and the second best for the other matrices with values not
far from the best.
Keywords— Named Data Networking, Forwarding Strategy, SEA-ME-

WE Topology

Paper ID 058
Layer 2 Communication Performance Comparison of Ad-hoc Mesh

and Software-Defined Vehicular Network OpenFlow
Ribka Tiarma1, Nana Rachmana Syambas1, Eueung Mulyana1, Galih

Nugraha Nurkahfi2
1
2
Research Center for Telecommunication National Agency for Research
and Innovation (BRIN)
181290078@telecom.stei.itb.ac.id, nanasyambas@itb.ac.id,
eueung@itb.ac.id, 133221305@std.stei.itb.ac.id,
galih.nugraha.nurkahfi@brin.go.id
Abstract -Intelligent Transportation System (ITS) is an advanced

application that aims to provide innovative services related to
different modes of transport and traffic management. Software-
Defined Network (SDN) is a networking architecture that gives the
ability to do centralized configuration and monitoring just using
software. Its ability open opportunities to incorporate ITS and SDN.
Layer 2 communication of computer networks constitutes a crucial
role in facilitating reliable communication between devices within the
same network segment. This communication is fundamental for
computer networks. Departing from that concept, this experiment
focuses on examining the worthiness of applying SDN in VN by
comparing l2 communication performance between ad-hoc mesh and
SDVN OpenFlow. As a result, SDVN OpenFlow is superior to ad-hoc
mesh acknowledged from every communication aspects performed.
Keywords: Layer 2 Communication, Vehicular Network, Ad-Hoc

Mesh, Software-Defined Vehicular Network OpenFlow

Paper ID 059
Optimizing Customer Experience Through Low Latency Search: A

Delay Analysis of Heuristic and linear search Methods for E-
Commerce Applications
Agil Fuad Gumelar, Tutun Juhana, Usman Zakaria

Bandung, Indonesia
18120024@telecom.stei.itb.ac.id, tutun@stei.itb.ac.id,
18120038@telecom.stei.itb.ac.id
Abstract—The usage of electronic commerce applications has

significantly expanded, especially in Indonesia, and it has become
increasingly important to provide an optimal user experience to
handle a large number of users. One of the crucial aspects of
electronic commerce applications is the ability to perform effective
and efficient product searches. Product search is highly demanded,
and the loading speed of a product search can significantly slow down
when multiple users are simultaneously using it. Both heuristic
algorithms and linear search are reliable in handling product
searches. Linear search requires exponential computational time as
the data size increases, while heuristic algorithms have relatively
constant computational time. This research aims to compare the
performance of heuristic algorithms and linear search in product
searching within an electronic commerce application and the
objective is to develop an optimized technique that effectively
balances the trade-offs inherent in these approaches.
Keywords—Android application, Heuristic search, linear search,

electronic commerce

Paper ID 087
Selection of a Simple 5G Handover Algorithm: Preparation for

Named Data Networking (NDN)
Ade Nurhayati1,2, Syaiful Ahdan1,3, Tutun Juhana1, Nana Rachmana

Syambas1
1
2
Faculty of Applied Science
Telkom University
Jakarta, 11710, Indonesia
3
Faculty of Engineering & Computer Science
Lampung, 35132, Indonesia
adenurhayati@telkomuniversity.ac.id, syaifulahdan@teknokrat.ac.id,
tutun@stei.itb.ac.id, nana@stei.itb.ac.id
Abstract—Named Data Networking as an alternative network for 5G

network traffic is required to be able to provide better performance
compared to other networks such as internet protocol networks. In
NDN wireless, it is known that there is producer mobility and
consumer mobility, so optimal mobility management is needed,
including the handover process. In this paper the author proposes a
simple RSSI-based 5G handover algorithm with distance as a
reference RSSI value. This algorithm is a preparation for the
handover algorithm on the NDN wireless network. By using the
assumed distance value MS to BTS as the basic value for calculating
RSSI, handover determination is determined by comparing the RSSI
value to the RSSI threshold value. With the scenario of different
numbers of users and cluster sizes, from the handover process
involving 8 BTS it can be calculated how big the handover probability
is for each user. From the simulation results, it can be concluded that
the more users and the larger the cluster size, the greater the
handover probability value. This is linear, with the wider the cluster, the
probability of getting an RSSI signal from the original BTS becomes smaller,
making it possible to get an RSSI signal from neighboring BTS, so that a
handover occurs.
Keywords: NDN, 5G, Handover Probability, RSSI

Paper ID 107
A Systematic Review Network Slicing: Revenue Optimization for

Slice Admission Control Objectives
Innel Lindra1, Hendrawan1, Agus Subekti2

1
Bandung Institute of Technology (ITB)
2
Research Center for Telecommunications
National Research & Innovation Agency (BRIN)
Bandung, Indonesia
33223009@std.stei.itb.ac.id, hend@itb.ac.id, agus.subekti@brin.go.id
Abstract – Network slicing has been introduced as a promising

paradigm for 5G/B5G technology. The slice admission control
process is one of the first steps in network slicing. The decisions taken
in the slice admission control process will affect subsequent network
performance. Therefore, choosing the right slice admission control
mechanism is very necessary, one of the goals of which is to increase
operator revenue. In addition, the elasticity of a slice and the
optimization methods used in calculating operator revenue will also
be investigated.
Keywords – network slicing, slice admission control, slice admission

elasticity, slice admission control optimization method

Program Book ICWT 2024

Uploaded by

Copyright:

Available Formats

Program Book ICWT 2024

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Program Book ICWT 2024

Uploaded by

Copyright:

Available Formats

10th ICWT 2024, Batam, Indonesia

July 04-05, 2024

July 04-05, 2024, Nagoya Hill, Hill Batam i

10th International Conference Wireless and Telematics (ICWT) 2024

Copyright and Reprint Permission: Abstracting is permitted with credit to

All right reserved. Copyright ©2024 by IEEE.

Conference Record Number 62080

July 04-05, 2024, Nagoya Hill, Hill Batam ii

Table of Contents ................................................................ iii

July 04-05, 2024, Nagoya Hill, Hill Batam iii

This page is intentionally left blank.

July 04-05, 2024, Nagoya Hill, Hill Batam iv

Message from the General Chair

ICWT has been expected to become one of the important

Since 2015, this conference has brought together a tremendous and

July 04-05, 2024, Nagoya Hill, Hill Batam 1

Sponsors; and particularly I want to thank all members of the

Finally, I wish all participants a successful and fruitful conference.

Prof. Dr. Ir. Hendrawan, M.Sc.

July 04-05, 2024, Nagoya Hill, Hill Batam 2

Organizers and Organizing Commitee

International Advisory Committee

July 04-05, 2024, Nagoya Hill, Hill Batam 3

 Muhammad Ammar Wibisono (Universiteit Twente, the

Technical Program Committee

July 04-05, 2024, Nagoya Hill, Hill Batam 4

This page is intentionally left blank.

July 04-05, 2024, Nagoya Hill, Hill Batam 5

July 04-05, 2024, Nagoya Hill, Hill Batam 6

This page is intentionally left blank.

July 04-05, 2024, Nagoya Hill, Hill Batam 7

10th ICWT 2024

Thursday, 4 July 2024 (Day 1)

July 04-05, 2024, Nagoya Hill, Hill Batam 8

Thursday, 4 July 2024 (Day 1)

July 04-05, 2024, Nagoya Hill, Hill Batam 9

Thursday, 4 July 2024 (Day 1)

July 04-05, 2024, Nagoya Hill, Hill Batam 10

Thursday, 4 July 2024 (Day 1)

July 04-05, 2024, Nagoya Hill, Hill Batam 11

Thursday, 4 July 2024 (Day 1)

July 04-05, 2024, Nagoya Hill, Hill Batam 12

July 04-05, 2024, Nagoya Hill, Hill Batam 13

July 04-05, 2024, Nagoya Hill, Hill Batam 14

July 04-05, 2024, Nagoya Hill, Hill Batam 15

July 04-05, 2024, Nagoya Hill, Hill Batam 16

July 04-05, 2024, Nagoya Hill, Hill Batam 17

Track: Electrical, Computer, & Biomedical Engineering

July 04-05, 2024, Nagoya Hill, Hill Batam 18

July 04-05, 2024, Nagoya Hill, Hill Batam 19

July 04-05, 2024, Nagoya Hill, Hill Batam 20

Track: Wireless Technology, Mobile Communication, & Radar

July 04-05, 2024, Nagoya Hill, Hill Batam 21

Track: Electrical, Computer, & Biomedical Engineering

Enhancing Electricity Usage Monitoring and

Comprehensive Risk Evaluation Model for Data

July 04-05, 2024, Nagoya Hill, Hill Batam 22

Track: Telematics, Networking, & Telematics Application

Implementation of the Hybrid Base ABR

Multiple Faces Recognition Model for Automatic