Telfor Journal Vol.12 No.2 (2020)

Content

OpenDaylight and OpenNebula Integration: Testing Traffic Management

O. Belkadi, A. Vulpe, Y. Laaziz, and S. Halunga
Topic:
Telecommunications Networks
Abstract
Software-Defined Networking (SDN) and Cloud Computing are now two of the most adopted technologies, on which many organizations are working to enhance every day. For instance, SDN is particularly emerging to solve networking complexity in cloud data centers, so we see many attempts to integrate Network and Cloud Managers. In this paper, we address an integration of these two technologies, particularly a yet undiscussed combination of two popular frameworks: OpenNebula and OpenDaylight. These open source solutions are widely used for cloud management and network management, yet there are no developed modules for communication between the two. Therefore, we propose a simple way for OpenDaylight to manage OpenNebula's compute nodes, using a common component they both support: OpenvSwitch. We compared OpenNebula with the popular OpenStack cloud manager, as it is attracting more attention in both academia and industry, by evaluating some relevant time metrics and discussing the differences of the proposed technologies. Then, we deployed a test topology to conduct some traffic management techniques in this integration. Our results show that OpenNebula's deployment time as well as clean-up time is significantly lower than OpenStack, but OpenStack takes less time to the running state, besides proving the simplicity of traffic management in OpenNebula using OpenDaylight.
Keywords
Software-Defined Networking, Cloud Computing, OpenNebula, OpenDaylight, Traffic management.
Full Text
HyperLink OpenDaylight and OpenNebula Integration: Testing Traffic Management
Page(s)
74-79
Doi
10.5937/telfor2002074B

Analysis and Characterization of IoT Malware Command and Control Communication

Đ. D. Jovanović and P. V. Vuletić
Topic:
Telecommunications Networks
Abstract
The emergence of Mirai botnet in 2016 took worldwide research teams by surprise, proving that a large number of low-performance IoT devices could be hacked and used for illegal purposes, causing extremely voluminous DDoS attacks. Therefore, a thorough inspection of the current state of IoT botnets is essential. In this paper, we analyze the dynamic behavior and command and control channels of two classes of IoT botnets, Mirai and Gafgyt. Based on collected information, a comparative analysis and key phases of botnet communication is provided. Such an analysis will serve as a basis for smart botnet detection mechanisms.
Keywords
Botnets, CnC communication, IoT.
Full Text
HyperLink Analysis and Characterization of IoT Malware Command and Control Communication
Page(s)
80-85
Doi
10.5937/telfor2002080J

Comparing Simulation Model for Objective QoE Video Evaluation with real IPTV Test Scenario During Appearance of Packet Losses

N. Goran, A. Begović, and N. Škaljo
Topic:
Telecommunications Networks
Abstract
The aim of this article is to explain benefits of the simulation model for objective QoE (Quality of Experience) video evaluation in order to find similarities with real IPTV (Internet Protocol TeleVision) system distribution in a DSL (Digital Subscriber Line) network. In DSL networks, it is known that packet losses appear suddenly and might have a "burst" character, at any time during IPTV service delivery. In addition, these packet losses usually appear in groups and lead to huge degradation of the IPTV video service, which decreases a customer’s QoE guaranteed level. Hence, estimation of this degradation in an access network is crucial and this paper explains a simulation model based on SSIM (Structural Similarity Index) analysis, which can be used as one perceptive video quality assessment by imitating a real environment with packet losses. To check this, we compared simulation model cases with the real ones in IPTV video distributed over DSL and exposed to different packet loss appearances.
Keywords
IPTV, packet loss, QoS/QoE simulation, SSIM, QoE assessment.
Full Text
HyperLink Comparing Simulation Model for Objective QoE Video Evaluation with real IPTV Test Scenario During Appearance of Packet Losses
Page(s)
86-91
Doi
10.5937/telfor2002086G

Approach to Implementation of Local Navigation of Mobile Robotic Systems in Agriculture with the Aid of Radio Modules

R. Iakovlev and A. Saveliev
Topic:
Radio Communications
Abstract
In this paper an approach is presented, enabling to solve the problem of local navigation of mobile robotic platforms (MRP), based on utilization of wireless networks with mesh topology. Establishment of wireless networks was ensured, based on the set of radio modules, mounted on unmanned aerial vehicles (UAV), comprising a swarm. This paper presents a developed algorithm for establishment of such wireless networks, aided by LoRa-technology, as well as an algorithm for MRP localization, based on analysis of signal level, where the incoming signals are fed from MRP group radio modules to radio modules of wireless data transfer network. An algorithmic model is given for task distribution among UAV and to implement navigational capabilities of MRP swarm. In some experiments descending dependencies of absolute error value, pertinent to MRP, from the number of UAV in action were revealed, as well as of averaged deflection value of MRP positions in motion along their paths from the number of UAV in action. Thereby the averaged value of MRP localization error, depending on the number of UAV in action, was from 8.14 to 17.13 m, and the averaged value of MRP position deflection – from 16.38 to 57.12 m, respectively.
Keywords
robot navigation, LoRa, radio communication, mesh topology, mesh networks.
Full Text
HyperLink Approach to Implementation of Local Navigation of Mobile Robotic Systems in Agriculture with the Aid of Radio Modules
Page(s)
92-97
Doi
10.5937/telfor2002092I

Implementation Challenge and Analysis of Thermal Image Degradation on R-CNN Face Detection

N. Latinović, T. Vuković, R. Petrović, M. Pavlović, M. Kadijević, I. Popadić, and M. Veinović
Topic:
Signal Processing
Abstract
Face detection systems with color cameras were rapidly evolving and have been well researched. In environments with good visibility they can reach excellent accuracy. But changes in illumination conditions can result in performance degradation, which is the one of the major limitations in visible light face detection systems. The solution to this problem could be in using thermal infrared cameras, since their operation doesn't depend on illumination. Recent studies have shown that deep learning methods can achieve an impressive performance on object detection tasks, and face detection in particular. The goal of this paper is to find an effective way to take advantages from thermal infrared spectra and provide an analysis of various image degradation influence on thermal face detection performance in a system based on R-CNN with special accent on implementation on a hardware platform for video signal processing that institute Vlatacom has developed, called vVSP.
Keywords
face detection, image degradation, R-CNN, thermal images, Video Signal Processing, GPU.
Full Text
HyperLink Implementation Challenge and Analysis of Thermal Image Degradation on R-CNN Face Detection
Page(s)
98-103
Doi
10.5937/telfor2002098L

Detection of Cyber-attacks in Systems with Distributed Control based on Support Vector Regression

D. M. Nedeljkovic, Z. B. Jakovljevic, Z. Dj. Miljkovic, and M. Pajic
Topic:
Signal Processing
Abstract
Concept of Industry 4.0 and implementation of Cyber Physical Systems (CPS) and Internet of Things (IoT) in industrial plants are changing the way we manufacture. Introduction of industrial IoT leads to ubiquitous communication (usually wireless) between devices in industrial control systems, thus introducing numerous security concerns and opening up wide space for potential malicious threats and attacks. As a consequence of various cyber-attacks, fatal failures can occur on system parts or the system as a whole. Therefore, security mechanisms must be developed to provide sufficient resilience to cyber-attacks and keep the system safe and protected. In this paper we present a method for detection of attacks on sensor signals, based on  insensitive support vector regression (ε-SVR). The method is implemented on publicly available data obtained from Secure Water Treatment (SWaT) testbed as well as on a real-world continuous time controlled electro-pneumatic positioning system. In both cases, the method successfully detected all considered attacks (without false positives).
Keywords
Cyber Physical Systems, Cyber Security, Industrial Control Systems, Industrial Internet of Things, Support Vector Regression.
Full Text
HyperLink Detection of Cyber-attacks in Systems with Distributed Control based on Support Vector Regression
Page(s)
104-109
Doi
10.5937/telfor2002104N

Transfer Learning for Domain and Environment Adaptation in Serbian ASR

B. Z. Popović, E. T. Pakoci, and D. J. Pekar
Topic:
Speech Processing
Abstract
In automatic speech recognition systems, the training data used for system development and the data actually obtained from the users of the system sometimes significantly differ in practice. However, other, more similar data may be available. Transfer learning can help to exploit such similar data for training in order to boost the automatic speech recognizer’s performance for a certain domain. This paper presents a few applications of transfer learning in the context of speech recognition, specifically for the Serbian language. Several methods are proposed, with the goal of optimizing system performance on a specific part of the existing speech database for Serbian, or in a noisy environment. The experimental results evaluated on a test set from the desired domain show significant improvement in both word error rate and character error rate.
Keywords
Automatic speech recognition, Kaldi speech recognition toolkit, transfer learning, noise adaptation, Serbian.
Full Text
HyperLink Transfer Learning for Domain and Environment Adaptation in Serbian ASR
Page(s)
110-115
Doi
10.5937/telfor2002110P

Striping Input Feature Map Cache for Reducing off-chip Memory Traffic in CNN Accelerators

R. Struharik and V. Vranjkovic
Topic:
Applied Electronics
Abstract
Data movement between the Convolutional Neural Network (CNN) accelerators and off-chip memory is critical concerning the overall power consumption. Minimizing power consumption is particularly important for low power embedded applications. Specific CNN computes patterns offer a possibility of significant data reuse, leading to the idea of using specialized on-chip cache memories which enable a significant improvement in power consumption. However, due to the unique caching pattern present within CNNs, standard cache memories would not be efficient. In this paper, a novel on-chip cache memory architecture, based on the idea of input feature map striping, is proposed, which requires significantly less on-chip memory resources compared to previously proposed solutions. Experiment results show that the proposed cache architecture can reduce on-chip memory size by a factor of 16 or more, while increasing power consumption no more than 15%, compared to some of the previously proposed solutions.
Keywords
Cache memory, convolutional neural network, energy-efficiency, low-power computing.
Full Text
HyperLink Striping Input Feature Map Cache for Reducing off-chip Memory Traffic in CNN Accelerators
Page(s)
116-121
Doi
10.5937/telfor2002116S