Biljana Stojkoska

Air pollution monitoring and control is becoming a key priority in urban areas due to its substantial effect on human morbidity and mortality. This paper presents a system architecture for intelligent pollution visualization and future pollution prediction by encompassing pollution measurements and meteorological parameters. First, a pollution model using spatial interpolation is built. By adding meteorological parameters this model is further used to identify the pollution field evolution and the position of potential sources of air pollution. Using deep learning techniques, the system provides predictions for future pollution levels as well as times to reaching alarming thresholds. The whole system is encompassed in a fast, easy to use web service and a client that visually renders the system responses. The system is built and tested on data for the city of Skopje. Although the spatial resolution of the system data is low, the results are satisfactory and promising. Since the system can be seamlessly deployed on an Internet of Things sensing architecture, the improved data spatial resolution will improve performance.

Indoor localization, as a key enabling technology for Inter-net of Things (IoT) location-based applications, is one of the primary services in smart automated systems. It is becoming increasingly important and beneficial in many industrial, commercial and public domains. This paper proposes an iterative trilateration technique for localization in indoor IoT environment, as an improvement of baseline trilateration. In our simulations, the iterative trilateration achieves high localization coverage with accurate location estimation compared to baseline trilat-eration. Additionally, our technique was analyzed considering different parameters like communication range, range error and anchors' fraction in three-dimensional (3D) space under scenarios with and without obstacle. The results show that our iterative trilateration technique is almost resistant to the obstacles in the environment.

This paper presents a development of a low-power device for livestock tracking in an outdoor environment by using RF technologies. The animal tracking device (AnTrack) is self-sustainable with a watertight solar panel(s). The device records the exact location every 15 minutes and when the device is within the radio range of the base station, it automatically sends the data to be relayed to a server Over-the-Air. Analysis can then be done on the location points of each animal. We perform a detailed analysis of the power consumption and prove that our AnTrack is capable of generating enough supply power even when there is no sunshine for a week.

Background

Assessment of health benefits associated with physical activity depend on the activity duration, intensity and frequency, therefore their correct identification is very valuable and important in epidemiological and clinical studies. The aims of this study are: to develop an algorithm for automatic identification of intended jogging periods; and to assess whether the identification performance is improved when using two accelerometers at the hip and ankle, compared to when using only one at either position.

Methods

The study used diarized jogging periods and the corresponding accelerometer data from thirty-nine, 15-year-old adolescents, collected under field conditions, as part of the GINIplus study. The data was obtained from two accelerometers placed at the hip and ankle. Automated feature engineering technique was performed to extract features from the raw accelerometer readings and to select a subset of the most significant features. Four machine learning algorithms were used for classification: Logistic regression, Support Vector Machines, Random Forest and Extremely Randomized Trees. Classification was performed using only data from the hip accelerometer, using only data from ankle accelerometer and using data from both accelerometers.

Results

The reported jogging periods were verified by visual inspection and used as golden standard. After the feature selection and tuning of the classification algorithms, all options provided a classification accuracy of at least 0.99, independent of the applied segmentation strategy with sliding windows of either 60s or 180s. The best matching ratio, i.e. the length of correctly identified jogging periods related to the total time including the missed ones, was up to 0.875. It could be additionally improved up to 0.967 by application of post-classification rules, which considered the duration of breaks and jogging periods. There was no obvious benefit of using two accelerometers, rather almost the same performance could be achieved from either accelerometer position.

Conclusions

Machine learning techniques can be used for automatic activity recognition, as they provide very accurate activity recognition, significantly more accurate than when keeping a diary. Identification of jogging periods in adolescents can be performed using only one accelerometer. Performance-wise there is no significant benefit from using accelerometers on both locations.

Nowadays, finding the Unmanned Aerial Vehicle (UAV) position in the absence of GPS is attractive and challenging problem in the research community. In this paper, we present a novel algorithm for mini UAV indoor localization based on distance measurements between the UAV and the existing infrastructure consisting of WiFi Access Points. Our algorithm uses two well-known techniques from the literature: Multi-dimensional Scaling (MDS) and Weighted Centroid Localization (WCL). Through extensive simulations we have shown that our algorithm is very suitable for indoor localization of mini UAVs. For small radio-range error, our algorithm exhibits a small localization error of less than 5% of the radio range.

Customer churn is one of the main problems in the telecommunications industry. Several studies have shown that attracting new customers is much more expensive than retaining existing ones. Therefore, companies are focusing on developing accurate and reliable predictive models to identify potential customers that will churn in the near future. The aim of this paper is investigating the main reasons for churn in telecommunication sector in Macedonia. The proposed methodology for analysis of churn prediction covers several phases: understanding the business; selection, analysis and data processing; implementing various algorithms for classification; evaluation of the classifiers and choosing the best one for prediction. The obtained results for the data from a telecommunication company in Macedonia, should be of great value for management and marketing departments of other telecommunication companies in the country and wider.

The increasing average age of the population in most industrialized countries imposes a necessity for developing advanced and practical services using state-of-the-art technologies, dedicated to personal living spaces. In this paper, we introduce a hierarchical distributed approach for home care systems based on a new paradigm known as Internet of Things (IoT). The proposed generic framework is supported by a three level data management model composed of dew computing, fog computing and cloud computing for efficient data flow in IoT based home care systems. We examine the proposed model through a real case scenario of an early fire detection system using a distributed fuzzy logic approach. The obtained results prove that such an implementation of dew and fog computing provides high accuracy in fire detection IoT systems, while achieving minimum data latency.

Pervasive and ubiquity computing are expected to expand the development of new business oriented mobile applications. Knowing the exact physical location of the wireless devices is crucial for providing awareness of these applications. Many algorithms have been proposed for wireless localization, but most of them are designed for outdoor localization by using Global Positioning System (GPS) signals. In indoor environments, the GPS is not available, which makes localizations not trivial. This paper surveys state-of-the-art attempts toward efficient indoor localization for smartphones. We define a taxonomy used for better classification of the algorithms. Furthermore, we describe the characteristics of modern indoor positioning systems, as well as the challenges associated with the localization techniques. Finally, we provide real experiments using different smartphone models in order to discover typical problems that occur when signal strength is used as a range measurement technique in indoor lo-calization systems. Keywords: indoor localization · taxonomy · smartphone · RSSI

In this paper we implemented a local geometry alignment algorithm for locating the primary user (PU) in cognitive radio network. Based on the estimated distance between PUs and Secondary users (SUs) for the neighbors within certain communication range, the relative configuration of all users in the network is obtained initially and is refined finally to get the global position of every user in the network. The localization performance of the proposed approach is compared to multidimensional scaling and principal component analysis.

Protein interaction networks (PINs) are argued to be the richest source of hidden knowledge of the intrinsic physical and/or functional meanings of the involved proteins. We propose a novel method for computational protein function prediction based on semantic homogeneity optimization in PIN (SHOPIN). The SHOPIN method creates graph representations of the PIN augmented by inclusion of the semantics of the proteins and their interacting contexts. Network wide semantic relationships, modeled using random walks, are used to map the augmented PIN graphs in a new semantic metric space. The method produces a hierarchical partitioning of the PIN optimal in terms of semantic homogeneity by iterative optimization of the ratio of between clusters dissimilarities and within clusters similarities in the new semantic metric space. Function prediction is done using cluster wide-hierarchy high function enrichment. Results validate the rationale of the SHOPIN method placing it right next to state-of-the-art approaches performance wise.

Pervasive and ubiquitous computing is expected to expend the development of new business oriented mobile applications. Knowing the exact physical location of the wireless devices is crucial for providing awareness of these applications. Many algorithms have been proposed for wireless localization, but most of them are designed for outdoor localization by using Global Positioning System (GPS). In indoor environments where GPS is not available localization becomes complex and challenging task. Reflection, diffraction or absorption of signal, as well as other characteristics like temperature, antenna orientation or presence of obstacles represents a big challenge. Another challenge is to successfully combine different localization techniques and signals from different radio emitters, like Wi-Fi, Bluetooth or GSM base stations, to achieve accurate omnipresent indoor localization. OMINLOC aims to develop new comprehensive set of methodologies and analysis tools for indoor GPS-free wirele...

In the near future wireless sensor networks are expected to
become increasingly popular due to their low cost and ease
use. As a distributed systems, they are usually deployed in
unattended and hostile environment which makes them
vulnerability to “man in the middle attack” since the medium
they are using could not be restricted. The only way to protect
the content from being exposed to the adversary is by using
cryptography. Most often the public key and the symmetric key
encryption are used. However the cryptographic algorithms
are processor and memory demanding.
When using wireless sensor networks (WSN) we need to
consider the aspects of protecting communication among the
nodes. We are investigating the possible attacks and suggest
our adaptive holistic approach to secure the network while
minimizing the cost.

energy in a given WSN, so the most logical way to
reduce the energy consumption is to reduce the number of
radio transmissions. To address this issue, there have been
developed data reduction strategies which reduce the amount of
sent data by predicting the measured values both at the source
and the sink, requiring transmission only if a certain reading
differ by a given margin from the predicted values. While these
strategies often provide great reduction in power consumption,
they need a-priori knowledge of the explored domain in order
to correctly model the expected values. Using a widely known
mathematical apparatus called the Least Mean Square
Algorithm (LMS), it is possible to get great energy savings
while eliminating the need of former knowledge or any kind of
modeling. In this paper with we use the Least Mean Square
Algorithm with variable step size (LMS-VSS) parameter. By
applying this algorithm on real-world data set with different
WSN topologies, we achieved maximum data reduction of over
95%, while retaining a reasonably high precision.

""As Wireless Sensor Network (WSN) has become a
key technology for different types of smart environment, nodes
localization in WSN has arisen as a very challenging problem
in the research community. Most of the applications for WSNs
necessitate a priory known nodes positions. In this paper, we
propose an algorithm for three dimensional (3D) nodes
localization in surface WSN based on multidimensional scaling
(MDS) technique. Using extensive simulations, we investigated
in details our approach regarding different network topologies,
various network parameters and performance issues. The
results from simulations show that our algorithm produces
small localization error and outperforms MDS-MAP in terms
of accuracy.""

The main constraint of wireless sensor networks (WSN) in enabling wireless image communication is the high energy requirement, which may exceed even the future capabilities of battery technologies. In this paper we have shown that this bottleneck can be overcome by developing local in-network image processing algorithm that offers optimal energy consumption. Our algorithm is very suitable for intruder detection applications. Each node is responsible for processing the image captured by the video sensor, which consists of NxN blocks. If an intruder is detected in the monitoring region, the node will transmit the image for further processing. Otherwise, the node takes no action. Results provided from our experiments show that our algorithm is better than the traditional moving object detection techniques by a factor of (N/2) in terms of energy savings.