International Journal of Vehicle Information and Communication Systems (37 papers in press)
Journey in vehicular ad-hoc network: a survey of message dissemination approaches and their delays
by Puja Padiya, Amarsinh Vidhate, Ramesh Vasappanavara
Abstract: Vehicular Ad-Hoc Network (VANET) is currently an active area of research and aims to improve vehicle, road safety, traffic efficiency, convenience and comfort for drivers as well as passengers. This paper provides a state of the art overview of VANET standards, architectures, channel access methods and message dissemination approaches. A detailed survey based on delays, especially those that occur in reactive message dissemination approaches, with a short survey of predictive message dissemination approaches, has been presented. We also highlight our view on some of the open issues to be addressed.
Keywords: VANET; vehicular ad-hoc network; data dissemination; road safety; routing; delays; quality of service; standards; architectures; channel access.
Cloud-assisted multi-tier hierarchical safety routing strategy for collision avoidance in a vehicular ad hoc network
by Nalina V, P. Jayarekha
Abstract: The Vehicular Ad hoc network (VANET) comprises a number of moving vehicles that establish wireless communication between them directly or using fixed infrastructure. Generally, the vehicles in VANET can obtain various services such as safety and comfort by establishing cooperative communication among them or from global servers through the internet. The main intention of VANET is to protect human lives from a dangerous situation and avoid chain collisions by alerting vehicles through emergency messages. Disseminating emergency messages in a hazardous area through single hop or multi-hop is a fundamental approach for efficiently delivering the emergency alert messages to all vehicles. However, the dissemination approaches incur a high redundant rate and inefficient use of network resources. The design of safety message dissemination protocols has to ensure reliable data delivery with strict delay deadline and also use the network resources in an efficient way. Taking into account the multicriteria information in dissemination oriented decision making is an appropriate solution for critical message communication. This paper proposes a Cloud-assisted Multi-tier Hierarchical Safety Routing (CM-HSR) strategy to avoid chain collisions with efficient resource usage. Initially, the CM-HSR divides the vehicles into a logical multi-tier hierarchical structure based on multiple information retrieved from cloud and roadside infrastructure. For effectively handling an emergency situation and network dynamism, the CM-HSR dynamically changes the multi-tier structure with the help of roadside infrastructures. To ensure reliable delivery with minimum redundant rate, the CM-HSR incorporates two mechanisms that are Accident severity level based Dangerous region Formation (ADF) and Multicriteria Decision Making (MDM). Finally, the simulation results demonstrate that the proposed CM-HSR attains better performance in terms of latency, duplicate packets, number of collisions, number of transmitted data packets, reachability, overhead, and number of secondary collisions in evaluation.
Keywords: Cloud-assisted VANET; logical multi-tier hierarchical structure; multicast message dissemination; multicriteria decision making; optimal forwarder vehicle selection.
End-to-end delay and backlog bound analysis for hybrid vehicular ad hoc networks: a stochastic network calculus approach
by Shivani Gupta, Vandana Khaitan
Abstract: This paper studies a hybrid Vehicular Ad-hoc Network (VANET) that incorporates two different technologies, i.e., IEEE 802.11p and the long-term evolution. End-to-end delay and backlog are acknowledged as major performance measures of vehicular networks that characterise its Quality of Service (QoS), therefore, in this paper we focus on obtaining some measures of these two attributes. We obtain the probabilistic upper bounds on the end-to-end delay and backlog instead of evaluating the delay and backlog in view of the fact that providing the probabilistic bounds is more reasonable as in some real-life scenarios it may be intricate to obtain the closed-form results. To obtain the probabilistic bounds on delay and backlog, a queueing network model is proposed that represents the message dissemination scheme used in the hybrid VANET architecture. The novelty of this paper lies in the fact that instead of considering a Markovian queueing network, the arrival and service processes in the proposed queueing network are assumed to be self-similar and heavy-tailed distributed, as such characteristics are extensively reported in communication networks. The mathematical analysis of the proposed queueing network follows the stochastic network calculus approach for the reason that it supports generally distributed arrival and service processes. Further, the probabilistic upper bounds on the end-to-end delay obtained using heavy-tailed arrival and service times are compared with the delay bounds obtained using exponential arrival and service times to validate the appropriateness of using heavy-tailed characteristics of the network traffic. In addition, a comparative study of hybrid VANET with the other two conventional architectures of VANET, i.e. ad-hoc network only and cellular network only, is also provided in the paper.
Keywords: hybrid VANET; queueing network; stochastic network calculus; end-to-end delay; backlog; probabilistic bounds; heavy-tailed traffic.
Camouflage-based location privacy preserving scheme in vehicular ad hoc networks
by Leila Benarous, Benamar Kadri, Saadi Boudjit
Abstract: Location privacy is critical and preserving it is essential. The tracking exposes the real time location, history of visited places and parsed trajectories. Metaphorically speaking, it is the cyber equivalent of physical stalking and as dangerous as it is. In vehicular networks in particular, this issue is serious because autonomous vehicles timely transmit their locations, headings, speed and identity to neighbouring vehicles and/or service infrastructures. To preserve the location privacy, various pseudonym-based approaches exist, mainly focusing on unlinkable pseudonym change strategies. In this paper, we propose a camouflage-based solution that prevents the linkability of pseudonyms even within low density roads where the tracking chances are high. The solution is simulated using NS2 against a global passive attacker that executes the semantic and syntactic linking attacks. The results demonstrate the effectiveness of the solution in protecting privacy.
Keywords: autonomous vehicle; vehicular network; privacy; attacker; simulation; linkability.
3D object detection based on image and LIDAR fusion for autonomous driving
by Guoqiang Chen, Huailong Yi, Zhuangzhuang Mao
Abstract: 3D object detection is the fundamental task of autonomous driving. The existing approaches are very expensive in computation owing to the high dimensionality of point clouds. We use the 3D data more efficiently by representing the scene from the RGB image and the Birds Eye View (BEV). The whole network is composed of two parts: one is the 2D proposal network for 2D region proposal generation, and the other is the 3D region-based fusion network to predict the 2D locations, orientations, and 3D locations of the objects. First, we fuse the BEV feature map and the RGB image to enhance the input. Second, we adopt the 3D encoding form with 2D-3D bounding box consistency constraints and design ROI-wise feature fusion to predict location information. Our experimental evaluation on both the KITTI as well as a large scale 3D object detection benchmark shows significant improvements over the state of the art.
Keywords: 3D object detection; image and LIDAR; deep learning; multisensory fusion; autonomous driving.
Modular arithmetic and subset sum problem: a state-of-art technique in information security issues towards a smart vehicular system
by Anirban Bhowmik
Abstract: Nowadays, vehicles are used at a large scale in modern society. But in many countries the current traffic-safety statistics are very terrifying. Many people are killed and injured in road accidents. To reduce this problem, governments and manufacturers have launched different initiatives, such as the use of safety belts, airbags, antilocking brake systems and smart vehicular transportation systems. Upcoming traffic safety initiatives in smart transportation systems depend on information technology, and this technology also helps to authenticate and track vehicles in the system. Recent smart vehicular systems use different types of network, such as VANETs, AI-based applications, etc., that aim to provide a safer, coordinated, smooth and smart mode of transportation. This article focuses on the communication security issues in smart vehicular applications. Transmitting messages efficiently and accurately among vehicles is the key issue in this system. At present, the communication in smart transportation systems is vulnerable to various types of security attack because it uses an open wireless connection. The different types of attack are secrecy attack, routing attack, data authenticity attack, and attack on authentication; besides these, in a dense environment, the vehicle may receive multiple messages at the same time. Therefore, how to complete the authentication of multiple messages in a short time is an urgent problem. To address these problems, here we have introduced a technique using the concepts of approximation algorithm and linear congruence. The different types of experiment on our technique and their results confirm that our scheme is very secure, robust and efficient for data transmission in smart vehicular.
Keywords: smart vehicular system; intermediate key; linear congruence; approximation algorithm; subset sum problem; session key; nonlinear function; CLS.
Efficient clustering for wireless sensor networks using modified bacterial foraging algorithm
by Dharmraj Biradar, Dharmpal D. Doye, Kulbhushan A. Choure
Abstract: The energy efficiency and clustering are directly related to each other in Wireless Sensor Networks (WSNs). A significant number of methods have been introduced for energy-efficient clustering in the last couple of decades. To limit energy use and improve network throughput, various methods for the clustering algorithm were introduced using an optimisation algorithm, fuzzy logic, and thresholding techniques. The optimisation algorithms such as Particle Swarm Optimisation (PSO), Genetic Algorithm (GA), Ant Colony Optimisation (ACO) and their variants were presented, but the challenge of selecting the efficient Cluster Head (CH) and cluster formation around it with minimum overhead and energy consumption is unresolved. In this paper, energy proficient and lightweight clustering algorithm for WSNs is proposed using the Modified Bacterial Foraging optimisation Algorithm (MBFA). The aim of designing the MBFA is to limit energy use, control overhead, and improve network throughput in this paper. The process of CH selection using MBFA is performed via a novel fitness function. The wellness capacity is planned using key parameters, for example, remaining energy, node degree, and geographical distance between sensors to base station. The MBFA selects the sensor node as CH using the fitness value. The proposed clustering protocol is simulated and evaluated with state-of-art protocols to justify efficiency.
Keywords: bacterial foraging optimisation; clustering; cluster head selection; energy efficiency; particle swarm optimisation.
Innovative approach to prevent wormhole attack on reactive routing of vehicular ad-hoc network by using clustering and digital signatures
by Shahjahan Ali, Parma Nand, Shailesh Tiwari
Abstract: Owing to wireless nature and dynamic topology, a vehicular ad-hoc network is sensitive towards various types of assault, in which wormhole assault is one of them. This attack disturbs the route discovery process of any routing protocol. Here an innovative approach is proposed to prevent the wormhole attack on reactive routing such as AODV, DSR of VANET by using the clustering and digital signatures concept. In this research work, SUMO 0.32.0 and NS-3.24.1 simulators are used. The simulation results show that the proposed approach is able to prevent the wormhole assault on reactive routing in VANET. The novelty of this research work is that till now the approach based on clustering and digital signatures to prevent the wormhole assault on reactive routing in VANET has not been used. By using this proposed approach the VANET can be made more secure. The secure VANET is essential to implement the intelligent transport system.
Keywords: vehicular ad-hoc network; routing; AODV; DSR; wormhole; SUMO 0.32.0; NS-3.24.1; digital signatures.
Enhanced video-based traffic management application with virtual multi-loop crate
by Manipriya Sankaranarayanan, Mala Chelliah, Samson Mathew
Abstract: The growth in urban population leads to gridlock of vehicles in city roads. The quality of transportation is improved by the latest technologies of Intelligent Transportation Systems (ITS) applications. Any ITS application relies heavily on sensors for data collection for efficient management, control and planning of transportation. In this paper, the video-based traffic data collection systems and their techniques are improved by using the proposed Virtual Multi-Loop Crate (VMLC) framework. VMLC uses the all the spatial colour information for image processing without losing information. The results of the proposed framework are used to estimate traffic statistics and parameters that are essential for ITS applications. The parameter values obtained from VMLC are analysed for accuracy and efficiency using Congestion Level (CoLe) estimation application. The results show that the VMLC framework improves the quality of data collection for any video-based ITS applications.
Keywords: vehicle detection; traffic statistics and parameters; spatial colour information; image processing data management; video-based traffic data.
Improved stability through self-localisation scheme in heterogeneous vehicular clustering
by Iftikhar Ahmad, Zaheed Ahmed, Muhammad Ahmad Al-Rashid
Abstract: Intelligent local data processing within vehicular ad hoc networks (VANET) may increase the capabilities of the Internet of Vehicles (IoV). To share data effectively, vehicular clusters should be synchronised and stable. A vehicle needs an uninterrupted Global Positioning System (GPS) signal for synchronisation purposes, especially in the urban environment. GPS interruption leads to an unstable connection that is a big hurdle in developing cost-effective solutions for navigation and route planning applications. To solve this problem, a self-location calculation scheme within the vehicular clustering process is proposed. The proposed self-location calculation algorithm enables vehicles to calculate their coordinates in the absence of GPS signals. A clustering mechanism is developed for sharing traffic information system (TIS) data among multiple vehicles over a particular road segment. Sharing of vehicular data in real-time helps vehicles to synchronise well. The developed scheme is simulated and compared with existing known approaches. The results show the better stability of our proposed mechanism over others.
Keywords: VANET; stability; vehicular clustering; synchronisation; localisation.
An improved spectral temporal average method for mitigating Doppler effects in V2V communications
by Wahyu Pamungkas, Jans Hendry, Walid Maulana Hadiansyah
Abstract: Vehicle-based communication systems have the main characteristic of high-speed users. This movement causes the Doppler effect and the communication channel to change very rapidly. One parameter that determines the mitigation performance of the Doppler effect is the ability of channel estimation to overcome the very fast channel changes. In this study, the communication channel used was V2V with a moving scatterer and a greater number of scatterers than previous studies, where the characteristics of this channel generate a greater Doppler effect than the channel model without a scatterer. In the spectral temporal average (STA) Doppler effect mitigation method, the channel estimation section still uses the least squares (LS) method. Limitations in the shapes that linear models may assume over long ranges, probably weak extrapolation properties, and sensitivity to outliers are the key disadvantages of linear least squares. To solve this problem, this study proposed the piecewise method for the channel estimation section as part of the STA Doppler effect mitigation method. Both methods were tested in low-speed and high-speed scenarios with 100 symbols. The simulation results show that the performance of the STA method using the piecewise method had an advantage of up to 99.3% compared with the LS method at a speed of 10 m/s and SNR of 17 dB, and 98.5% at a speed of 100 m/s and SNR of 20 dB with 100 symbols. Increasing the speed from 10 m/s to 100 m/s decreased the BER performance by up to 38 times.
Keywords: V2V channel; Doppler effect; least squares; piecewise method; spectral temporal average.
Special Issue on: Big Data Innovation For Sustainable VANET Management
An automatic moving vehicle detection system based on hypothesis generation and verification in a traffic surveillance system
by Smitha Jolakula Asoka, N. Rajkumar
Abstract: An intelligent transportation system has a major topic called traffic surveillance. In a complex urban traffic surveillance system, booming of vehicle detection and tracking is an problematic dilemma. To overcome this, a two-stage approach for a moving vehicle detection system is proposed in this paper. The proposed system mainly consists of two stages namely, hypothesis generation and hypothesis verification. At the first step, hypotheses are generated using the concept that shadows beneath the vehicles are darker than the road region. The second step verifies whether a generated hypothesis is correct or not using an optimal artificial neural network (ANN). The weights corresponding to the ANN are optimally selected using the grasshopper optimisation algorithm. Through experimental results, it is shown that the proposed moving vehicle detection system performs with better accuracy than other methods.
Keywords: traffic surveillance system; moving vehicle detection; tracking; hypothesis generation; hypothesis verification; feedforward neural network; grasshopper optimisation algorithm.
Special Issue on: Research Challenges and Emerging Technologies in Autonomous Systems
Fuzzy-based local agent routing protocol for delay-conscious MANETs
by C. Venkataramanan, B. Senthilkumar
Abstract: Owing to the demand on multimedia applications, most researchers still concentrate on the area of Mobile Ad hoc NETworks (MANETs) to ensure the quality of services. MANET is an infrastructure-less network, where the devices (i.e. nodes) are self-configuring together and form the network without any central coordinator. Owing to the absence of central monitoring, MANET experiences various issues such as packet loss, topological control and delay. In order to address those problems in this paper, the enhanced version of Ad hoc On Demand Distance Vector (AODV) routing protocol is proposed. According to this proposed approach, each node in the network has to find the number of packets in the queue and calculate the weight value, which is used to predict the best routing path for ongoing transmission. The local agents are nominated for collecting and processing the information. The local agent performs the decision-based routing by using fuzzy inference model (AODV-FLA).
Keywords: AODV; energy usage; fuzzy; MANETs; routing; QoS.
An experimental analysis of quad-wheel autonomous robot location and path planning using the Borahsid algorithm with GPS and Zigbee
by Siddhanta Borah, R. Kumar, Subhradip Mukherjee, Fenil. C.Panwala, A. Prasanna Lakshmi
Abstract: This paper presents a hardware system structure and wireless navigation system for both localisation and path navigation of a mobile robot, implementing a 32-bit ARM processor (LPC2148 Board) into the design process of a mobile robot integrated with GPS and a ZigBee wireless communication device. A novel path-navigation algorithm (Borahsid algorithm), with less complexity than the existing algorithms adopted for mobile robot realistic work, uses GPS localisation as well as ZigBee communication. For simulation purpose MATLAB programming language has been used to simulate the mobile robot localisation and path navigation, and the results show the effectiveness of the model and the feasibility of the Borahsid algorithm. However, the entire control structure is executed and the experimental results were obtained in a real time system. The experimental results authenticate the performance and steadiness of the implemented control system process.
Keywords: ARM processor; GPS; ZigBee-based communication; Borahsid algorithm; MATLAB.
Special Issue on: ICBCC-2019 Intelligent Transportation Systems for Smart Cities
Improved coverage measurements through machine learning algorithms in a situational aware channel condition for indoor distributed massive MIMO mm-wave system
by Vankayala Chethan Prakash, G. Nagarajan, V. Subramaniyaswamy, Logesh Ravi
Abstract: In a massive MIMO (Multiple Input Multiple output) mm (millimetre)-wave system, the channel conditions are measured and analysed for a better placement of reflectors or antennas. In order to increase the coverage area and to reduce interference among users factors such as pathloss and power delay profile are extracted from the channel impulse response (CIR) i.e. from the received signal with respect to transmitter and receiver channel propagation conditions. In a distributed indoor massive MIMO mm-wave system, pathloss and power delay profile are evaluated for line of sight (LoS) and non-line of sight (NLoS) environments at frequencies such as 28 and 39 GHz. Based on these factors, a dataset is constructed for 28 GHz. Algorithms such as Support Vector Machine (SVM), KNN and Fine Tree are considered. These algorithms are trained with a set of datasets and are tested for performance metrics such as Mean Absolute Error, Correlation Coefficient, Root Mean Squared Error, Relative Absolute Error, and Root Relative Squared Error, which are evaluated. Simulation results show an accuracy of 94% and 95% using SVM, 93.8% and 94.5% using KNN, and 93.2% and 93.8% using Fine Tree algorithm for pathloss and power delay profile respectively.
Keywords: Fine Tree; KNN; massive MIMO; mm-wave; pathloss; power delay profile; support vector machine.
Multivariate short-term traffic flow prediction based on real-time expressway toll plaza data using non-parametric techniques
by Annu Mor, Mukesh Kumar
Abstract: Accurate real-time traffic flow prediction is a vital component of an Intelligent Transportation System (ITS).The real-time traffic flow prediction helps transportation authorities as well as travellers for better route guidance. In this study, a novel approach is proposed for accurate toll plaza traffic prediction by introducing heterogeneous data sources other than traffic volume data. Toll data is analysed with exogenous factors, such as weather conditions and holidays. Here, ten non-parametric techniques is applied for traffic prediction on a real-time multivariate dataset. The proposed approach is validated using data collected from Pinjore-Kalka Toll Plaza, Chandigarh, India. The performances of the non-parametric models are compared on the basis of mean square error, absolute mean square error, coefficient of determination and correlation. The experimental results revealed that the random forest regression technique outperforms other techniques, achieving an accuracy of 90%. The proposed approach can be used for further proxy measure of level-of-service to design the existing infrastructure more efficiently for application in smart cities.
Keywords: traffic flow; intelligent transportation system; non-parametric technique; multivariate time series data; proxy measure.
Effect of feature and sampling ratio on tool wear classification during boring operation using tree-based algorithms
by Surendar Selvasubramaniam, Elangovan Mahadevan, Akshay Elangovan, Vijayakumar Varadarajan
Abstract: The tool condition monitoring (TCM) system is used to predict the tool wear during the machining process. The predominant wear is the flank wear which has its impact on the surface roughness of the workpiece that is being machined. The quantum of flank wear is to be ascertained so that a decision could be made whether the time has come for the insert to be replaced. Although since the wear is continuous, it may be divided into three stages and may be classified as to which stage the tool wear falls into. Wear prediction may be carried out by extracting information from the vibration signals acquired during machining and interpreting them using machine learning. This paper confers on monitoring the uncoated carbide tooltip during boring operation using tree-based classifier algorithms such as random forest, J48, logistic model tree and gradient-boosted tree, in order to study the effect of feature and sampling ratio on tool wear classification when tree-based algorithms are used. Also, the statistical features and histogram features were compared for various cutting tool conditions to explore a better classifierfeature combine.
Keywords: J48; random forest tree; gradient-boosted tree; logistic model tree; Knime analytics platform.
Dynamic formulation of a two link flexible manipulator and its comparison analysis with a knuckle joint cantilever
by Prasenjit Sarkhel, Nilotpal Banerjee, Nirmal Baran Hui
Abstract: In the present study, a dynamic modelling technique for a two-link flexible planar manipulator is presented. The developed manipulator model can include an arbitrary number of flexible links. The gross equation of motion of each flexible link has been obtained by composing the rigid body motion and a small elastic deformation of each link. In order to bring simplicity, the modelling technique has been applied to model a two-link flexible manipulator. The analytical derivation and final expression for the equation of motion has been shown for the two-link flexible manipulator using the proposed modelling method. Later on, a comparison analysis has been studied between the proposed manipulator model and a simple cantilever with a knuckle joint. Finite element analysis of the beam has been done in ANSYS 17.1 considering different types of element under different loading conditions. The various numerical results have been generated at different nodal points by taking the origin of the Cartesian coordinate system at the fixed end of the beam. Finally, a comparison analysis has been made to find a way to validate the proposed mathematical modelling.
Keywords: flexible manipulator; dynamic modelling; equation of motion; cantilever beam; knuckle joint.
Special Issue on: AIST2019 Empowering Intelligent Transportation Using Artificial Intelligence Technologies
Network traffic analysis using machine learning techniques in IoT network
by Shailendra Mishra
Abstract: End-node internet-of-things devices are not very intelligent and resource-constrained; thus, they are vulnerable to cyber threats. They have their IP address, and once the hacker traces the IP, it becomes easy to get into the network and exploit the other devices. Cyber threats can become potentially harmful and lead to infection of machines, disruption of network topologies, and denial of services to their legitimate users. Artificial intelligence-driven methods and advanced machine learning-based network investigation protect the network from malicious traffic. The support vector machine learning technique is used to classify normal and abnormal traffic. Network traffic analysis has been done to detect and protect the network from malicious traffic. Static and dynamic analysis of malware has been done. Mininet emulator is selected for network design, VMware fusion is used for creating a virtual environment, the hosting OS is Ubuntu Linux, and the network topology is a tree topology. Wireshark was used to open an existing packet capture file that contains network traffic. Signature-based and heuristic detection techniques were used to analyse the signature of the record, which is found using a hex editor, and proposed rules are applied for searching for and detecting these files that have this signature. The support vector machine classifier demonstrated the best performance with 99% accuracy
Keywords: network traffic analysis; IoT; cyber threats; cyber attacks; machine learning.
A novel framework for efficient information dissemination for V2X
by Ravi Tomar, Hanumat G. Sastry, Manish Prateek
Abstract: This paper is focused on presenting a robust framework for information dissemination in vehicular networks using both Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) communication modes. The framework is designed to first prioritise the generated information and then, based on the priority, the message is disseminated over the network using one of the techniques for V2V or V2I. The paper first discusses the need for information dissemination and further proposes the novel framework for efficient information dissemination. The framework comprises two techniques for disseminating the information through V2V or V2I. The two techniques are presented and supported by the experimental, simulation and statistical analysis results. The results obtained are compared with existing mechanisms for information dissemination and are found to be performing better than standard information dissemination mechanisms.
Keywords: information dissemination; V2V; V2I; priority based.
Automated storyboard generation with parameters dependencies for regression test cases
by Nishant Gupta, Vibhash Yadav, Mayank Singh
Abstract: In recent trends and advancement of agile technology, the industry demand is for an effective and useful specification from the customer to reduce the effort, time and cost of software development. The storyboard is an effective tool to cater for the customer's requirements in an efficient manner. Our proposed framework and tool STORB will provide the platform where customer and business analyst may use the tool to generate a storyboard based on provided functionalities and parameters. The tool will provide detailed information about the customers requirements and generate the storyboard. Further, test data can also be generated for testing test cases. The tool has been used for three functionalities and their parameters on login functionalities of web application. The tool also defines the dependencies among parameters so that regression test cases can be generated. The result shows a useful significance of the tool in the software industry for the current trend of agile development.
Keywords: agile testing; regression testing; storyboard;test cases; functionalities.
Machine learning techniques applied to call admission control in 5G mobile networks
by Charu Awasthi, Prashant Kumar Mishra
Abstract: Highly reliable applications with low latency are key feature in 5G networks. In the prevailing scenario of efficient mobile network systems, the Quality of Service (QoS) depends on the regulation of traffic volume in wireless communications, known as the Call Admission Control (CAC). 5G networks are also very important for Intelligent Transportation Systems (ITS) as they can be used for quick detection and controlling of traffic, hence can be informative, sustainable, and more effective. Machine learning is the concept of providing the power to learn and develop mechanically, by practising. It also provides the power to attain learning and development in the absence of classical methods such as programming. It also permits wireless networks such as 5G to be increasingly dynamic and predictive. With this feature, the formulation of the 5G vision seems possible. With the use of machine learning and neural networks, this paper proposes various CAC methods deployed for 5G multimedia mobile networks. This can be achieved by delivering the best from all the attributes of soft computing that are deployed in the current mobile networks for ensuring recovery of efficiency of the prevailing CAC methods.
Keywords: artificial intelligence; machine learning; neural networks; 5G mobile networks; wireless networks; intelligent transportation system.
PALCT: vehicle-to-vehicle communication based on pseudonym assignment and encryption scheme using delay minimisation cover tree algorithm
by Righa Tandon, P.K. Gupta
Abstract: Vehicle-to-Vehicle communication is one of the new paradigms of networking which should be secure, fast and efficient. In this paper, we propose a framework that implements the pseudonym-based authentication scheme in which communication among vehicles is encrypted by using matrix array symmetric key (MASK), digital signature algorithm (DSA) and intelligent water drop (IWD). The proposed security scheme also ensures handling of many security attacks such as key-guessing, non-repudiation, replay and modification. In the proposed scheme, to preserve the vehicles identity, we have provided different pseudonyms to each vehicle in the network, which ensures secure communication among vehicles. Furthermore, the proposed delay minimisation cover tree algorithm ensures the issue of time-delay during vehicle to vehicle communication. In this algorithm, we have used Dijkstras algorithm for finding the optimal shortest path during vehicular communication. Obtained results show that the proposed scheme is effective and efficient as it reduces the time-delay by 4% for 140 vehicle nodes and 28.4% for 1000 vehicle nodes.
Keywords: pseudonyms; vehicle-to-vehicle communication; security; time-delay.
Special Issue on: Intelligent Edge Computing for Connected and Autonomous Vehicles Trends and Challenges
Predictive mechanism of a modified bug controller for mobile robot path navigation
by Subhradip Mukherjee, R. Kumar, Siddhanta Borah
Abstract: Bug approaches are popular for mobile robot navigation in challenging environments. The traditional bug approach follows a virtual straight line from source to target location and exhibits obstacle boundary following behaviour if an obstacle is present between source and target location. This behaviour of the Bug approach consumes more travel time towards target location. In addition, obstacles with larger size situated in central-left or central-right side of the virtual straight line between source and target location create navigation problem for Bug approaches. A modified bug (m-Bug) controller has been proposed and realised for mobile robot path navigation with a better solution in MATLAB and V-REP simulation environments. With range sensors, the proposed controller was found to exhibit optimised travel time and path length in the given environments. Different static environments with single and multiple obstacles have been considered to test the proposed controller. Various simulation results and comparative analysis highlight the superiority of the controller.
Keywords: bug approach; obstacle boundary following behaviour; modified bug controller; V-REP; travel time; path length.
A queuing theory based delay efficient packet scheduler for machine type communication
by T.N. Sunita, A. Bharathi Malakreddy
Abstract: Lately, there has been a drastic change in the telecommunication industry owing to the emergence of the Internet of Things (IoT). IoT is the network through which various objects/devices are connected to accomplish a particular task or a goal with very little intervention of humans. Machine to Machine (M2M) plays a very important role in enabling IoT. Right now, Long-Term Evolution (LTE) is the best supporting technology for Machine Type Communication (MTC), because of its flexibility, compatibility and high availability of radio resource. As the LTE is mainly suitable for Human-to-Human (H2H) communication, MTC faces some challenges in LTE such as radio resource management and uplink-scheduling algorithms being unsuitable for MTC, because MTC have different traffic characteristics and are mainly uplink dominant. Moreover, these issues can be handled efficiently through virtualisation technology, which would give us the capability to better manage the network resources through technologies such as Network Function Virtualisation (NFV). In this paper, we use a queuing theory and propose a packet scheduler, which would model the regular and event-driven traffic and schedule the jobs in queues for request processing. Then using the Markov chain process, we calculate the queue length and determine the blocking probability. If there is high blocking probability and still there are some requests in a queue, then those requests will be processed in a virtual server for maximum network usage.
Keywords: machine-to-machine; machine type communication; delay; packet scheduler; blocking probability; queuing theory.
Area efficient and high speed Galois field multiplier for mobile edge computing devices
by N. Sharath Babu, Gunti Hemanth Santosh, S.R. C.H. Murthy Tommandru, M. Shiva Kumar
Abstract: Multiplication is one of the most important arithmetic operations in communication devices and it is implemented using finite field or Galois field arithmetic in this work. The performance of the finite field multiplication operation is closely related to the finite field elements representation. The proposed polynomial based finite field bit parallel systolic array multiplier is able to achieve almost double the speed of existing multiplier with little extra area. There is a considerable reduction in area and power for the proposed word-level normal basis finite field multiplier compared with the existing multiplier. The results clearly indicate that the proposed methods improve the efficiency of finite field multipliers in terms of area, delay or power consumption. Mobile edge computing is becoming one of the prominent technologies that assist to achieve formidable specifications of 5G technology in terms of reliability, latency, scalability and throughput. In order to probe and determine the real time data efficiently, local computing and data offloading are carried out in evolving a joint computation algorithm.
Keywords: high speed multipliers; finite field multiplication; parallel systolic array; edge devices; mobile edge computing.
Development of environment friendly nanoelectronic sensing elements for hybrid electric vehicles
by Dayanand B. Jadhav, Rajendra D. Kokate
Abstract: Nanomaterials and nanodevices have a major impact in the development of many innovative systems and they are the potential replacement for CMOS technology devices. Nanoelectronic sensors are essential in the development of hybrid electric vehicles and autonomous systems. This research paper focuses on the synthesis of ZnO nanoparticles and development of nanoelectronic sensing elements for automotive applications and hybrid electric vehicles. Environment friendly ZnO nanoparticles are used in this work to avoid harmful gases. ZnO nanoparticles are synthesised using Aloe vera extracts through mediated synthesis route. Structural and morphological characteristics are experimentally studied using various spectroscopic and microscopic measures. ZnO nanomaterial elements and LPG sensing properties are systematically investigated to check the suitability for electric vehicle applications. The effects of operating temperature on gas response, resistance and sensitivity characteristics are analysed using various experiments. The performance of our nanoelectronic sensing element is found to be superior to similar sensing elements in terms of working temperature, percentage response, response time and recovery time.
Keywords: nanoparticle; nanoelectronic sensors; LPG gas sensor; hybrid electric vehicles; autonomous systems; nanotechnology.
IoT enabled machine learning framework for social media content based recommendation system
by Adinarayana Salina, E. Ilavarasan, K. Yogeswara Rao
Abstract: Analysing huge volumes of data from the social media tweets on product reviews provides a better understanding of any product. Exploring customer opinions from tweets is helpful to find the strengths and weaknesses of different products and features. There are several studies on product recommendations from Twitter product reviews. In this paper, Internet of Things based two-level product recommendation framework (TLPRF) is proposed to efficiently handle large amounts of Twitter users product reviews data. TLPRF consists of a Raspberry Pi microcomputer as an IoT mining machine and it is programmed to generate a feature level opinion summary. Feature level opinion is found to be useful in accomplishing the product ranking. Based on the customer interest in the product purchase request, a normalised ranking of each matching product is calculated from the feature-wise opinion summary, and the product with maximum ranked score is recommended to Twitter users. The proposed TLPRF is found to be superior to similar other approaches in terms of accuracy, precision, recall and f-measure.
Keywords: internet of things; machine learning; ranking; summarisation; social networks; recommendation systems.
Vote mapping based improved human tracking for intelligent surveillance systems
by Kavita Wagh, Dipak B. Khandgaonkar
Abstract: Human tracking is a challenging task and significant part in the design of intelligent surveillance systems. Though the existing tracking techniques accomplished reasonable outcomes in terms of accuracy and robustness, there is scope for improving the tracking performance. In this paper, vote mapping of patched confidence methodology is used with the consecutively increasing number of patches. The system aims to provide robustness to occlusion and global scene changes by using the number of patches from the bounding box of an image. An individual patch is tracked by kernelised correlation filter and applied to the vote mapping methodology. The consecutively increasing number of patch approach and vote mapping provides robustness to the occlusion in real time tracking scenarios. The qualitative and quantitative analysis reveals the superiority of the proposed vote mapping-based tracker over the existing kernel-based trackers.
Keywords: Correlation; human tracking; vote mapping; kernel; regression; surveillance systems.
Handy text reader based text to speech technology for visually impaired persons
by D. Haripriya, Prathibhachand Bellamkonda, Marka Chandrika, Sachu Alekhya
Abstract: Visually impaired persons are not able to read the text content on their own and most of them struggle to live independently. Braille reading system, audio tactile and digital speech synthesiser have been developed for the visually impaired persons. But it is not always possible to have Braille, audio and video form of printed work. Many researchers used optical character recognition (OCR) tools with IoT technology to improve the reading capability of poor eyesight persons. However, this approach is not comfortable for the reader as there is a possibility of missing certain words by reader during image to text conversion. A handy and cost effective text reader is proposed in this work to help the people to read on their own using text to speech technology. A camera interfaced with Raspberry Pi is used to capture the alphabets and numbers from the document and convert them to text by PyTesseract OCR. The accuracy of the reader is 98% with the distance of 15 cm to 23 cm during day and night. The text document can be a handwritten script with minimum font size of 10 and it can be placed in any orientation.
Keywords: speech synthesiser; PyTesseract OCR; Braille reading system; text to speech technology; text reader; speech communication system.
Reduction of false positives in network intrusion detection using a hybrid classification approach
by H.M. Shreevyas, G.K. Ravikumar, B.N. Shobha
Abstract: Nations and their assets are becoming extremely vulnerable to cyber attacks, such as Distributed Denial of Service (DDoS). For most technologically developing nations this could have catastrophic effects on critical infrastructure and severely damage national economies. Proactive methods to identify network anomalies are most crucial to secure the network or host at defence organisations and also in the corporate sector. DoS attacks are quite popular since the last decade. Traditional DoS attacks can be countered easily now by several existing defence mechanisms, and the source of such attacks can be easily suspended with improved tracking capabilities. However, effective detection of DDoS is still a challenging task and the impact of such attacks is usually very damaging. According to recent studies, there has been a perception that machine learning can have remarkable impact on network security, mainly in network traffic analysis. It is useful to study and analyse network traffic behaviour constantly using large and real time datasets and train them to build a network model using advanced machine learning techniques. These have demonstrated capabilities for detecting both known and unknown attacks. Here we propose a hybrid classifier model that can detect both known and unknown attacks by using two-stage classifiers. The results obtained on benchmark datasets convincingly indicate that the proposed model is a highly useful classifier for detecting different types of TCP flooding based DDoS attack with a reduced number of false positives.
Keywords: cyber security; network intrusion detection; TCP flood attacks; DDoS; naive Bayes classifier; HMM; decision trees.
Special Issue on: Emerging Technologies for Internet of Vehicles
Effectiveness evaluation method for traffic data acquisition based on vehicle-borne network
by Minglei Song, Binghua Wu, Rongrong Li
Abstract: In order to reduce the probability of traffic accidents and enhance the safety of vehicle traffic, a method for evaluating the effectiveness of traffic data collection based on a vehicle network is proposed. A virtual coil is set on the driving lane to detect vehicles through three features of texture change, foreground area and pixel movement within the coil. A traffic detector is introduced to collect traffic data for a long time. Based on the cognitive assessment theory, a comprehensive assessment index system for the effectiveness of traffic data collection is constructed to complete the assessment. The experimental results show that the evaluation time of this method is less than 18 s, and the evaluation energy consumption is lower than other methods above 20 J, which proves that the evaluation time of this method is shorter, the error is smaller and the energy consumption is lower.
Keywords: vehicle-borne network; traffic data acquisition; traffic data evaluation.
Path planning method of automatic driving for directional navigation based on particle swarm optimisation
by Xian Luo, Rongtao Liao, Huanjun Hu, Yuxuan Ye
Abstract: In order to overcome the low planning efficiency of the automatic driving trajectory planning method for directional navigation, a particle swarm optimisation (PSO) based trajectory planning method is proposed. The kinematic characteristics of the vehicle are analysed and the vehicle dynamic equation is constructed. The position coordinates, speed and other motion parameters of the directional navigation vehicle are transformed into a Frenet coordinate system. The trajectory quality evaluation model of automatic driving vehicle for directional navigation is constructed. The trajectory quality evaluation index is taken as the constraint, and each variable is iteratively optimised by the PSO algorithm, so as to effectively realise the trajectory planning of automatic driving of directional navigation. Simulation results show that the proposed method can effectively improve the efficiency of autopilot trajectory planning and enhance the safety of the whole method.
Keywords: particle swarm optimisation; directional navigation; automatic driving; path planning.
Moving target tracking method in intelligent transportation system based on vehicle networking environment
by Dong-yuan Ge, Xi-fan Yao, Wen-jiang Xiang, Ri-bo He
Abstract: In order to improve the anti-jamming ability of moving target tracking and to avoid noise interference, a moving object tracking method based on vehicle network environment is proposed in this paper. The method of internet of vehicles is used to collect the echo signal of moving vehicle target, and the wavelet entropy feature is selected by multi-wavelet scale feature decomposition. According to the correlation feature tracking and identifying inf, the information model of target signal detection under the environment of internet of vehicles is established. The time-frequency characteristics of target signal and the high-order statistical characteristics of signal are analysed by using discrete orthogonal wavelet transform. The adaptive ability is enhanced and the moving target tracking and recognition is realised. The simulation results show that the method has strong anti-jamming ability, improves tracking accuracy, and has good recognition and notification capabilities.
Keywords: vehicle networking; intelligent transportation system; moving target; tracking; signal detection.
Research on data forwarding delay estimation of intelligent transportation system based on internet of vehicles technology
by Jian Gao, Daxin Tian
Abstract: In order to solve the problems of high error rate and long time in traditional data forwarding delay estimation methods, a data forwarding delay estimation method based on internet of vehicles technology for intelligent transportation system is proposed. Based on the Node-Link-Arc-Rord model and internet of vehicles technology, the simulation traffic network is constructed to optimise the intelligent transportation system. Based on the optimised intelligent transportation system, the average delay of the link is calculated according to the message data timestamp information, and the continuous vibration time and vibration period of the signal are estimated by using the sliding rectangular window, and the estimation results of the data forwarding delay of the intelligent transportation system are obtained. The experimental results show that the error rate of time delay estimation is less than 7%, the maximum estimated time is only 0.3 s, and the number of forward queued tasks is reduced.
Keywords: internet of vehicles; intelligent transportation system; data forwarding; time delay estimation; simulation traffic network; sliding rectangular window.
A new prediction method of short-term traffic flow at intersection based on Internet of vehicles
by Ying Zheng, Ying Zhou
Abstract: In order to overcome the problems of large error and long time-consuming in the prediction of short-term traffic flow at intersections, a new short-term traffic flow prediction method based on the internet of vehicles (IoV) is proposed in this paper. In the IoV environment, the training samples are input into the prediction model of the IoV, the output value is calculated, and the error is obtained. Then, the weights and wavelet factors of the network are modified by a gradient descent algorithm. When the network error reaches the set accuracy or reaches the maximum training time, the training is stopped to get the predicted short-term traffic flow. The experimental results show that the mean square percentage error is about 0.01%, and the longest prediction time is 0.878 min. The fitting degree between the predicted value and the actual value of traffic flow is high, and the prediction effect is ideal.
Keywords: internet of vehicles; intersection; short-term traffic flow; prediction.
Special Issue on: ICBCC-2019 Intelligent Transportation Systems for Smart Cities
Collaborative decision making system in intelligent transportation system using distributed blockchain technology
by Bhabendu Kumar Mohanta, Debasish Jena, Utkalika Satapathy, Somula Ramasubbareddy
Abstract: Intelligent Transportation System (ITS) is one of the promising
applications of the Internet of Things(IoT) as the IoT system provides an easy
way to collect and monitor the system. One of the critical components to make
a city smart is by managing the traffic. The modes of transportation in the city are
different, such as bike, car, bus, auto, and rickshaw. Most of the vehicles are integrated with Information Communication Technology (ICT). As the vehicles share and
access information from the ITS infrastructure, some security issues exist, including trust management, privacy, data linking, and computational problems. This paper identifies the security issues present in the ITS model, then proposes a distributed architecture of the ITS system using blockchain. Then the Consensus algorithm is used to perform computations in a distributed platform. The Ethereum platform used to create a distributed network. The implementation and security analysis are given at the end.
Keywords: secure decision making; blockchain; IoT; intelligent transportation system; Ethereum.