Showing papers by "Rishi Pal Singh published in 2016"
TL;DR: A localization scheme free from the concept of time synchronization for large scale 3D-UWSNs based on dive and rise mobile beacons floating over the sea surface has been proposed in the given paper.
Abstract: The underwater sensor networks (UWSNs) exhibit different characteristics from terrestrial WSNs. To make the sensed data meaningful, one of basic task is the localization of sensor nodes. In UWSNs, this is not feasible to use global positioning system due to its issue of propagation. This senses localization as a fundamental as well as sound issue in the UWSNs. In spite of continuous and sincere research effort, the time synchronization between the sensor nodes for their localization is also a tough job. To address these issues, a localization scheme free from the concept of time synchronization for large scale 3D-UWSNs based on dive and rise mobile beacons floating over the sea surface has been proposed in the given paper. The analytical analysis of the proposed scheme is also presented in the paper. In the simulation section, the performance of the proposed scheme is evaluated in terms of coverage and the number of localized nodes. A discussion based on various obtained results is also given in the paper.
35 citations
TL;DR: The evaluation of these techniques on diverse medical datasets gave an insight into predictive ability of Machine Learning in medical diagnosis and there is a wide space of improvement.
Abstract: Background/Objectives: Medical science industry has immense measure of information; however a large portion of this information is not mined .Machine Learning takes analytics to the extreme by exploring hidden information in data. Diagnosis of a disease is major objective of medical decision support system which will help the physicians to take effective decision Methods/Statistical analysis: In this research paper Machine Learning techniques, K-Nearest Neighbors (KNN), Decision Tree , Artificial neural networks (ANNs), Radial Basis Function (RBF) neural networks and Support Vector Machine (SVM) are analyzed . Findings: Performance of these techniques is compared through various performance measures such as sensitivity, specificity, accuracy, F measure, Kappa statistics, True Positive Rate, False Positive Rate and ROC on Breast Cancer Wisconsin, Liver Disorder, Hepatitis and cardiovascular Cleveland Heart disease datasets. Research work consists of 10 V-fold cross validation method to measure the fair estimate of these prediction techniques . Application/Improvements: The evaluation of these techniques on diverse medical datasets gave an insight into predictive ability of Machine Learning in medical diagnosis and there is a wide space of improvement.
23 citations
TL;DR: An analytical model for the throughput of the IEEE 802.11p MAC sublayer in the presence of hidden terminal outperforms the existing model and takes into consideration control packets and data packets separately.
Abstract: This paper concentrates on the IEEE 802.11p MAC sublayer to analyze its performance in the presence of hidden terminals. An analytical model for the throughput of the IEEE 802.11p in the presence of hidden terminal is presented in this paper. A 3-D Markov chain is created to model the backoff procedure for each access category. The different contention windows and arbitration interframe space are considered for each access category. The model also takes into consideration control packets and data packets separately. The probabilities of frame blocking, successful transmission, collisions and hidden terminals have been derived and used to calculate the throughput. The simulation results are presented to validate the analytical results of the proposed model. The results of the proposed model are compared with the existing model IEEE 802.11 EDCA (Kosek-Szott et al. in Comput Netw 55(3):622---635, 2011). The results of the proposed model outperform the existing model.
12 citations
TL;DR: The finding is that on medical domains the Proposed Weighted Class Based Clustering outperforms other clustering techniques, which gave an insight into predictive ability of Machine Learning in medical diagnosis.
Abstract: Background/Objectives: Medical Decision Support System (MDSS) is a diagnostic interface which provides computer assisted information retrieval as well as may support excellence decision making, to stay away from human error. Even if human decision-making is frequently most advantageous, but it is poor when there are vast amounts of data to be classified. Also capability and accuracy of decisions will decrease when humans are set into pressure and massive work. Forever there is a need and scope for a better MDSS. Methods/Statistical Analysis: Cluster analysis is a method of grouping of objects keen on different groups, has proved to be a valuable tool for identifying co-expressed genes, biologically related groupings of genes and patterns. K-means, Hierarchical and Fuzzy c-means are various clustering techniques have been employed to work as core part of MDSS. Findings: Proposed Weighted Class Based Clustering (WCBC) method is dependent on classifying properties of medical data itself. Weights are calculated on the basis of class value consequently increases separability by placing more number of instances of same class in same cluster. In this paper, the clustering algorithms K-means, Hierarchical, Fuzzy and Weighted Class based K-Means are examined for medical domains. Our finding is that on medical domains the Proposed Weighted Class Based Clustering outperforms others. Application/Improvements: The application of Proposed Weighted Class Based Clustering on medical datasets gave an insight into predictive ability of Machine Learning in medical diagnosis and there is a wide liberty that proposed approach can be used in RBF Neural Network for center calculation and data base Kernel Learning which is open area of research these days.
7 citations
01 Apr 2016
TL;DR: A new approach based on the frequency of a node selected as cluster head for the selection of new cluster head is proposed which is also more efficient than the former one and results in increased lifetime and thus, in increased connectivity.
Abstract: These days wireless sensor networks handle a widespread range of monitoring tasks using their sensing and vision capabilities. These tasks may consist of very complex functions and large calculations. In order to make network alive for a long duration, these sensors have to take decision in judicious and efficient manner. The increased lifetime results into increased connectivity. Also, the intellectual use of energy resources provides the nodes better connectivity to the base station. The different clustering protocols such as Low Energy Adaptive Clustering Hierarchy (LEACH) are used in order to gain energy efficiency. The LEACH uses probability as the key attribute to elect the new cluster head. In this paper, a new approach based on the frequency of a node selected as cluster head for the selection of new cluster head is proposed which is also more efficient than the former one. The proposed approach for cluster head selection results in increased lifetime and thus, in increased connectivity. Simulation results show that in terms of different parameters like number of live nodes, dead nodes and residual energy our proposed strategy outperforms existing conventional sensor network protocols.
5 citations
Journal Article•
TL;DR: A comparative analysis of threeVehicular Ad-hoc Network algorithms namely; AODV, DSDV and ZRP shows that none of the algorithms performs best in terms of all parameters while AODV andZRP’s performance are quite encouraging in Terms of throughput and PDR.
Abstract: Vehicular Ad-hoc Network has drawn the attention of various researchers all around the world in the recent years. The reason can be attributed towards its capability in solving real world problems like traffic congestion. Although quite useful, the development of such systems is inherited with some challenges. The development of efficient communication protocols is one of the major problems which need to be addressed. This paper presents a comparative analysis of three such algorithms namely; AODV, DSDV and ZRP. The algorithms are implemented using Network Simulator and their performance is compared in terms of throughput, Packet Delivery Ratio and End-to-End delay. The result shows that none of the algorithms performs best in terms of all parameters while AODV and ZRP’s performance are quite encouraging in terms of throughput and PDR. Keywords: AODV, ZPR, VANET
2 citations
TL;DR: The q-composite key pre-distribution technique is presented with new flavor that will enhance the network size as well as the security level in comparison to the existing techniques.
Abstract: In the hostile areas, deployment of the sensor nodes in wireless sensor networks is one of the basic issue to be addressed. The node deployment method has great impact on the performance metrics like connectivity, security and resilience. In this paper, a technique based on strong keying mechanism is proposed which will enhance the security of a non-homogeneous network using the random deployment of the nodes. For this, the q-composite key pre-distribution technique is presented with new flavor that will enhance the network size as well as the security level in comparison to the existing techniques. The technique ensures the k-connectivity among the nodes with a redundant method to provide backup for failed nodes. In the simulation section, the performance of the proposed scheme is evaluated using NS-2 based upon the real model MICAz. A discussion based on various obtained results is also given in the paper.
1 citations