Vellore Institute of Technology (VIT) is a private university located in&nbsp;Tamil Nadu, India. Founded in 1984, as Vellore Engineering College, the institution offers 20 undergraduate, 34 postgraduate, four integrated and four research programs. It has campuses in Vellore, Amravati, Bhopal and Chennai.

VIT is one of the top ranked private universities in India according to NIRF, THE and QS Rankings.&nbsp;Govt. of India has recognized&nbsp;VIT, Vellore as an&nbsp;Institution of Eminence. This has allowed VIT to take independent quality initiatives and move up in world ranking.

&nbsp;

&nbsp;

VIT University

Department of Database Systems

In the Wireless Sensor Networks the nodes’ energy efficiency is a primary concern that has to be addressed for enhancing network lifetime. If nodes are mobile in nature, its energy consumption goes high and node lifetime gets more vulnerable with the localization adding up on the energy constraint of the battery. This is the main reason behind location estimation algorithms in WSN gaining popularity. Precision in node’s location identification assists in minimizing energy consumed by nodes which would improve network lifetime and balances load across nodes. In this paper the authors intend to propose an estimation algorithm Redefined Vector Based Glowworm Swarm Optimization (R-VBGSO) that would identify the location of nodes with minimum error rate. In this algorithm, anchor node’s location with respect to unknown nodes is identified with help of passing few signals in the form of messages that are exchanged to locate the nodes position in the network. With the simulation results it is evident that proposed algorithm enhances network lifetime significantly with reduced location estimation error. Furthermore, area of coverage through the sensor nodes is enhanced as the static nodes of such a region would approach the region in the network where the network activity is reduced. Simulation results also demonstrate that R-VBGSO based sensor placement approach will enhance the coverage with less mobility given to the sensors. © Alpha Publishers. All rights reserved.

World University root node

School of Computer Science and Engineering

Vellore Campus

Journal of Green Engineering

Redefined vector based glowworm swarm optimization (R-vbgso) to improve the nodes’ energy efficiency in wireless sensor networks (wsn)

Driving may be a skill that needs our full attention to safety control the vehicle and reply to event happening on the road. Distracted driving becomes a bigger threat per annum and has been the leading causes of accidents for the past decades. There are 3 forms of distractions i eyes off the road visual, ii mind off the road cognitive, iii fatigue or drowsiness while driving. in step with the study released by National Highway Traffic Safety AdministrationNHTSA and also the Virginia Tech Transportation InstituteVTTI, 80 percent of collision and 65 percent of near collision involve some kind of driver distraction. To deal with these challenges, we introduce a system to observe the motive force in terms of fatigue, distraction and activities. This method deals with automatic driver activity detection supported visual information and MTCNN. We propose an algorithm to trace and analyze drivers eyes PERCOLSpercentage of eyelid closure, heartbeat rate. © 2020 SERSC.

International Journal of Advanced Science and Technology

Detection of driver drowsiness using multi-task cnn framework

Efficient task scheduling plays a critical role in cloud computing environment. In this paper, we proposed a novel trust-based fruit fly optimisation algorithm (TFOA) for task scheduling. Even though traditional scheduling algorithms, namely first come first serve, round robin, ant colony optimisation and so on are used broadly in cloud computing process but still efficient scheduling is not achieved. In general cloud service provider desires to receive the customer task in a faster rate allocated on the most trustworthy resource by using TFOA. Simulation outcomes show that the proposed algorithm performs better than the existing transitional algorithms like round robin and particle swarm optimisation (PSO) in terms of reduced makespan and turnaround time and efficient resource utilisation. © 2020 Inderscience Enterprises Ltd.

Department of Software Systems

International Journal of Internet Manufacturing and Services

Trust-based fruit fly optimisation algorithm for task scheduling in a cloud environment

This paper seeks to evaluate the performance of the state of the art object classification algorithms for the purpose of cyclist detection using the Tsinghua–Daimler Cyclist Benchmark. This model focuses on detecting cyclists on the road for its use in development of autonomous road vehicles and advanced driver-assistance systems for hybrid vehicles. The Tiny YOLO v2 algorithm is used here and requires less computational resources and higher real-time performance than the YOLO method, which is extremely desirable for the convenience of such autonomous vehicles. The model has been trained using the training images in the mentioned benchmark and has been tested for the test images available for the same. The average IoU for all the truth objects is calculated and the precision-recall graph for different thresholds was plotted. © 2020, Springer Nature Singapore Pte Ltd.

Department of Communication Engineering

School of Electronics Engineering

VIT Unitversity

Advances in Intelligent Systems and Computing Soft Computing for Problem Solving

Cyclist Detection Using Tiny YOLO v2

An autonomous vehicle needs a lot of smoothly running tasks to keep it on the road without interfering with other vehicles. Lane detection and object detection are some of the most important tasks which needs to be handled by the vehicle. There are many such works in the field to perform them. Instead, this paper focuses on the real time detection of emergency vehicles such as ambulances and fire engines. This would make roads safer and more convenient for such vehicles by allowing the vehicles to make way. In the future of autonomous vehicles, this would streamline traffic as well as give priority to the vehicles that need it. The proposed solution uses a camera which will capture a live video stream, process it and detect vehicles simultaneously alerting nearby vehicles and providing a means for them to make way. © 2020 IJSTR.

International Journal of Scientific and Technology Research

Edge computing solution for smart city - Emergency vehicle detection

Intrusion detection systems assume a noteworthy job in the provision of security in wireless Sensor networks. The existing intrusion detection systems focus only on the detection of the known types of attacks. However, it neglects to recognise the new types of attacks, which are introduced by malicious users leading to vulnerability and information loss in the network. In order to address this challenge, a new intrusion detection system, which detects the known and unknown types of attacks using an intelligent decision tree classification algorithm, has been proposed. For this purpose, a novel feature selection algorithm named dynamic recursive feature selection algorithm, which selects an optimal number of features from the data set is proposed. In addition, an intelligent fuzzy temporal decision tree algorithm is also proposed by extending the decision tree algorithm and integrated with convolution neural networks to detect the intruders effectively. The experimental analysis carried out using KDD cup data set and network trace data set demonstrates the effectiveness of this proposed approach. It proved that the false positive rate, energy consumption, and delay are reduced in the proposed work. In addition, the proposed system increases the network performance through increased packet delivery ratio. © The Institution of Engineering and Technology 2020

Computer Science

Chennai Campus

Department of Information Technology

School of Information Technology and Engineering

Fulltext

IET Communications

Intrusion detection using dynamic feature selection and fuzzy temporal decision tree classification for wireless sensor networks

Opinion mining is an ongoing research area in e-commerce which aims at analyzing the people's opinions, sentiments and emotions. Moreover, the existing e-commerce systems allow the users to share their feedback in the form of textual reviews regarding the products and services. It also allows the consumers to give ratings for products that help in future recommendation of products. In this research work, a computational framework for efficiently predicting the consumer review ratings on the products has been proposed. The proposed framework integrates dimensionality reduction, genetic algorithm, fuzzy c-means and adaptive neuro-fuzzy inference techniques to overcome the limitations of the existing systems. Experiments have been conducted in this work using Amazon dataset consisting of reviews for different products. This system provides better performance and prediction accuracy for review ratings when it is compared with the related work. © 2020 Inderscience Enterprises Ltd.

International Journal of Business Intelligence and Data Mining

Fuzzy-based review rating prediction in e-commerce

In wireless sensor networks (WSNs), energy optimization and the provision of security are the major design challenges. Since the wireless sensor devices are energy constrained, the issue of high energy consumption by the malicious nodes must be addressed well in order to enhance the network performance by making increased network lifetime, reduced energy consumption and delay. In the past, many researchers worked in the provision of new techniques for providing improved security to WSN in order to enhance the reliability in the routing process. However, most of the existing routing techniques are not able to achieve the required security through the use of intelligent techniques for safeguarding the sensor nodes from malicious attacks. In order to address these problems, a new fuzzy temporal clustering-based secured communication model with trust analysis and outlier detection has been developed in this research work. For this purpose, a new fuzzy temporal rule-based cluster-based routing algorithm with trust modelling and outlier detection for monitoring the nodes participating in the communication has been proposed. In addition, a fuzzy temporal rule- and distance-based outlier detection algorithm is also proposed in this paper for distinguishing the malicious nodes from other nodes within each cluster of the network and has been used in the secured routing algorithm. The proposed secure routing algorithm uses the temporal reasoning tasks of explanation-based learning and prediction as well as spatial constraints for making efficient routing decisions through the application of trust and key management techniques for performing effective authentication of nodes and thereby isolating the malicious nodes from communication through outlier detection. By applying these two proposed algorithms for communication in the proposed work, it is proved through experiments that the proposed secure routing algorithm and the outlier detection algorithm are able to perform secured and reliable routing through genuine cluster head nodes more effectively. Moreover, these two algorithms provide improved quality of service with respect to the reliability of communication, packet delivery ratio, reduction in end-to-end delay and reduced energy consumption. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature.

Soft Computing

Intelligent fuzzy rule-based approach with outlier detection for secured routing in WSN

Wireless sensor networks are used to monitor physical or environmental conditions such as temperature and pressure as well as to study the quality of certain environmental and natural entities like air and water bodies by collecting data about the various components present in the air/water at a given spot and time. But the complete data generated by the nodes in each iteration is not always useful, as most of them give the redundant information or details which does not provide any essential information, just bulge up the amount of data being transmitted. Therefore, this paper aims to formulate an early prevention method (EPM) which not only gives a way to detect failed nodes, but also increases the overall efficiency of the network by reducing the overhead at the sink. Copyright © 2020 Inderscience Enterprises Ltd.

International Journal of Internet Technology and Secured Transactions

An early prevention method for node failure in wireless sensor networks

In the wireless sensor networks energy consumption is broadest and widely explored area of research. The solution for energy optimization encompasses various techniques such as efficient routing protocols, data scheduling, clustering, hardware redesigning, supervised and unsupervised network learning algorithm, and so forth. Compared with all the methods that has been so far discussed, swarm intelligence (SI) is considered to be the optimal way to find solution for reducing energy consumption as it is simple and the network formation is understood by the natural mechanism present in nature. SI approaches include ant colony optimization (ACO), particle swarm optimization, glowworm swarm optimization (GSO), and so forth. In this article, the authors provide the solution for the energy conservation problem through efficient GSO methods combined ACO. The revised beaconing glowworm swarm optimization ant colony optimization algorithm will be applied on the sensor network divided into swarms based on glowworms, the ants are introduced in the network that would parse the network by visiting the swarm heads with the principle of ACO behind it. The algorithm is tested on MATLAB 2015a for performance comparison with the HM-ACOPSO method with depicts energy conservation and efficiency in data collection. © 2020 John Wiley & Sons Ltd

Concurrency Computation

Revised beaconing glowworm swarm optimization ant colony optimization algorithm to localize nodes and optimize the energy consumed by nodes in wireless sensor networks