Project Report

Name: Roll no: Class:

Qaisar Khan 190779 BSCYS-F-19-

B556
Muhammad 190803 BSCYS-F-19-
Adnan Haider B556
Muhammad 190819 BSCYS-F-19-
Sohail B556
Aqeel Shahzad 190851 BSCYS-F-19-
B556

Subject: Network Security

1|Page
IoT Network Security and Applications via Long Range Technology
In this project report, we worked on make a long-range technology based IoT network security
system and the application which also worked on LoRa technology. The Long-range technology
advancement which is used to make a security system that system which is enlisting, where we
are used to constructing up an organization laborer have that assembles and cycles position signal
from the multi-sensor cycles and examination taking care of module, and in a part of moment
recognizes range modules ca part of momentum through the wireless sensor network, the Arduino
United Nation UNO which is used to make the easy for using the software and hardware, and the
multi-sensory combination handling stations, which send the pieces of information to the central
system which checks the information through the inaccessible contraptions of the Long-range
technology organization. The course of action of this venture contains LoRa has, which get
signals from distinctive center points and are related with a multi-sensory combination math
system through an inaccessible network.

INTRODUCTION:
IoT stand for internet of thing where number of devices connected to sever. The device and sever are
connect by using IP addresses. And the communication between the sever and devices are wireless
communicate. This wireless technology is also knowing as LoRa technology. LoRa stand for long-range
technology. Basically, our project is based on this technology to secure the IoT device which are connect
to sever.
There is the following attack on IoT system:
 Botnets
 Blockchain
 Denial of Service
 Identity and data theft
 Social Engineering
 Ransomware
In this project, we also discussed about the LoRa technology and its localization system security analysis.
In this report, we also discussed about the range free, range-based attack. To prevent the data information,
We used encryption and decryption system. And discussed about the system architecture which help us to
Understand the structure of our system.

LoRa IoT Localization System Security Analysis:

LoRa IoT node location system:
The node location used one hub to gets its geographical area data. It depends on the value, power
consumption, or utilization and another factor that is helpful in the working process, most user’s
hub area frameworks utilize a guide hub helped hub area plan, that is, the organization contains a
couple of signal hubs, which acquire their area data via conveying the GPS area component, and
which send the information which contains the data of the area which assemble the coordinate
system.

2|Page
 Fig(a)

Fig(b)
(a) Long-range technology network architecture. (b) Long-range technology (LoRa) class
comparison

Analysis of attacks on the node localization system

Assaults on a hub confinement framework essentially happen in the position connection
estimating furthermore, assessment stages. The assault target is typically the guide hub or the
remote connection for communicating reference point messages. As various confinement
frameworks depend on various physical characteristics and confinement measures, the methods
for an assault are firmly identified with the area method utilized by the framework, which is broke
down, as follows.

Attack on range-based localization

Range based localization technique in WSN requires the use of special distance estimation
hardware in each node to accurately measure the distance and angle between the nodes. They use
different distance estimation technique to know their position in the network.

Attack on range-free localization

The range-Free Localization technique is not required to know the accurate distance between the
nodes using the distance estimation technique because of which it is simpler and cost-efficient
compared to the Range-based localization technique. Here, some reference nodes will be there
and some nodes will be fitted with powerful antennas, which can transmit the signal and the

3|Page
whole network can be divided into triangular regions and the localization effort will be to
determine whether each node falls inside or outside the triangle formed. This technique is known
as APIT. For example, replay, make, change, discard signal messages, wormhole attacks, and
Sybil attacks. Sybil's attack on the restriction framework means that harmful hubs will play
various roles. Subsequently, different centers that did not exist in the organization appeared,
which disrupted regular childbirth activities. To be more explicit, in the centroid algorithm, the
area of the abstruse hub is resolved as the polygonal centroid shaped of k adjoining signal hubs:

In any case, the security issue is less thought to be in the underlying plan of the LoRa IoT hub limitation
framework. For a while, space research has focused on the most proficient method to improve.

Fig. 2. Wormhole attack towards APIT algorithm.

Fig. 3. (Color online) Network-layer attacks towards DV-based localization

algorithms.

Sensor Localization Techniques and Analysis

In general, ethernet is connected by a short cable which connects the source of Internet to multiple
home devices. (which can access Internet using different protocols). As frequency increases, the
RF noise automatically increases. Also, the electromagnetic interference started to exist. So
ethernet wasn't able to work efficiently for higher frequencies which means in obtaining high data
rates. Also, another drawback is that, there was no way of controlled allocation of bandwidth

4|Page
when multiple sources were used to transfer the data between each other. Latency and
synchronization are another two problems with the ethernet. That's why we have moved to
automotive ethernet. In automotive industry robots find it hard to move because of poor mapping
and localization techniques. Mapping is nothing but making key areas of interest using the
topological area where the bot moves around. Localization is nothing but when the bot moves to a
certain location, it has to know what is the next location and what location it is at present. So
based upon the sensor readings it identifies the position of itself against the geometrical or
topological map and also how far it has moved from its previous position.
So, difference between localization and mapping is that mapping is having an idea about the
whole ground surroundings. While the localization is having details of particular location where it
is going to go around.
Two main sensors are camera and radar. The camera provides the vehicle visual surroundings and
the radar provides the physical surroundings of the car. Like what objects are present using
phenomenon called as doppler effect. Typically, radar might have ultrasonic sensors which sends
the signals and gets the echoes. Based on the echoes, the objects and their distance to the vehicle
are calculated. Radar can be called as environmental sensors and camera can be as processing
sensor

System Encryption and Decryption Process

Symmetric encryption has less computational cost than Asymmetric encryption with small storage
space and sensor computing power, which is not suitable for asymmetric computation like RSA
algorithm.
Rabin asymmetric key encryption algorithm is used here which have low power consumption and
more in line of LoRa WSN.
If the plain text is longer while using Rabin asymmetric key encryption algorithm, it will take
longer to encrypt it.
Following formula is used for encryption method:
C = M2 mod N.
Here, C denotes cipher text, M denotes plaintext, N denotes public key, where N = P × Q. The
sets of plain texts M1, M2, M3, M4 is calculated by private keys P and Q and public key N.
Following formulas are used for decryption:
(i)
W1 = C (p+1)/4 mod P
W2 = P-C (p+1)/4 mod P
W3 = C (Q+1)/4 mod Q
W4 = Q-C (Q+1)/4 mod Q
(ii)
a = Q × (Q−1mod P) b = P × (P−1modQ)

5|Page
 (iii)
M1 = (a ×W1 + b×W3) mod N
M2 = (a ×W1 + b×W4) mod N
M3 = (a ×W2 + b×W3) mod N M4 = (a ×W2 + b×W4) mod N.
Four sets of plain text used to solve a cipher text by using Rabin asymmetric key encryption
algorithm.
The original text will be added to original text once the correct plain text is found.
The technique utilized in this article is to take a letter from the first plain text as the comparison
data and append it to the first plain text to frame another plain text for encryption.

System Architecture:
Positioning of nodes is vital here require an optimized positioning system to know the signal
strength, packet reliability, etc. This optimized system can be possible by using SIBMSFCL.
SIBMSFCL algorithm diagnoses the abnormal packets that are deviated from minimum security
standards it improves the
positioning of the system. Lora is
flexible by being tolerant to attacks
such in a situation of packet delays
and low power consumption
technology that make it
advantageous over the ZigBee
network (another low power
network but short in range as
compared to the Lora network)
because the probability rate of
transmission delay in Lora as
compared to ZigBee is
(10.5%<34.6%) (ref p.127). Lora
network is openly deployed and has
node resource constraints, it is
difficult to remove the natural or other external threats as well as tampering attack since it inherits
the radio broadcasting feature. Concluding that heart operation i.e., localization of system is the
core problem in security of LoRa network.

This system contains three parts as following:

(Fig 9 p.128)
1. Nodes or sensor
2. Programmed chip (Raspberry pi)

6|Page
3. Network server host (CMS)
Indoor Test: - we have used the HBE
(High balanced Enhancement) values in range
0f 1-4 m and proper method is proposed from
Eq 1(where d1 has effect of HBE while d2 is
without effect of HBE). In this test, every 1 m
is calculated multiple times and then taken
average for the purpose of relationship between
HBE and distance.

7|Page
Experimental Results and Analysis

Indoor and Outdoor localization test

(ref.p.129 for eq1-3)

This above table (referred to table2 p.129) shows that error correction of localization algorithm is
more accurate than RSSI primal algorithm (by 50%) and RSSI weighting method (by 8%).

Outdoor Test: - Like Indoor test we have used the reference points and HBE values in
range of 10-50 m and using the eq1-3 to determine the relationship between HBE and distance.
For comparison illustrate the following table (referred to table3 p.130)

8|Page
 Also, if we consider the comparison between high energy consumption system (GPS) result with
the localization algorithm (using SIBMSFCL) result we find the error below 5% for localization
result.

Jamming test of LoRa and other WAN

As it is discussed earlier that LoRa is very attack tolerant that it can bear and survive almost in
both homogenous and heterogenous environment of frequency bands. For instance, if frequency
bands are different for other WAN, it has no effect on LoRa network. This network is tested in
915 MHz frequency band environment so in such situation with other WAN like ZigBee it is not
being affected. Also, it can be multi-hop in order to increase it range and reduce expenses. And if
packet is delayed or lost, there is mechanism of retransmission which improves it integrity which
means critical event will be generated in order to analyze the unknown packet is sent.

Accuracy and Reliability test comparison

Accuracy Test: - Below figure shows the result for sample test of LoRa network. It says
that there is 100% accuracy of localization result to the actual location when it is in range of 21m
however increasing the distance reduces the accuracy for instance in range of 32 m accuracy rate
is about 85%

Reliability Test: - For this test there were different environment test. In which we had test it on
different days (sunny, rainy, cloudy etc.) as well as the distance between source and destination.so we
found out that by increasing the transmission range there is reduction in the packet reliability. Below
is the table that verify this point.

Table5 (ref.p.132)

9|Page
Conclusion
In comparison to other network of LoRa type like Zigbee there is 30% increase in reliability i.e., both
network support low power and high range technique but LoRa has more advantage over Zigbee.
Therefore, in the era of rapid development of network security, multi-sensor fusion computing, and
LoRa technology, Localization system of national security monitoring system, energy or power
management, environment management, data management, factory monitoring, and renewable
energy management has become very important.
The main disadvantage of the LoRa module is that the bandwidth of LoRa system is low. Moreover,
it does not permit continuous transmitting of data. Hence, it is only used in periodic communication
system. The LoRa module can be used for long range communication. It is responsible to accomplish
the frequencies of communication, data rate and low power in any kind of devices. It mainly has
three components: such as End-note, Gateways and Network server. The End-note system contains
low powered sensors for transmitting the data. The Gateway system collects the

|Page
data from the end-note and sends them to the Network Server. The final component is mainly
responsible to scan the received data.
The following devices that are using this LoRa technology are Tracker’s device, Gas monitored
device and Fire detection modules.

Reference:
1). F. Y. Ren, H. N. Huang, and C. Lin: J. Software 14 (2008) 1282 (in Chinese with English
abstract). http://www. jos.org.cn/1000-9825/14/1282.htm
2). http://www.jos. org.cn/1000-9825/16/857.htm
3). https://www.semtech.com/lora/what-is-lora
4). https://ieeexplore.ieee.org/document/8115843
5). https://lora-alliance.org/wp-content/uploads/2020/11/lorawan_security_whitepaper.pdf

10 | P a g e
6).https://www.researchgate.net/publication/338489155_IoT_Network_Security_and_Application
s_via_Long_Range_Technology

11 | P a g e
7). https://link.springer.com/chapter/10.1007/978-3-319-95189-8

12 | P a g e
13 | P a g e