An Efficient Network Threat Detection and Classification Method using Anp-Mvps Algorithm in Wireless Sensor Networks
P. Sherubha1, N. Mohanasundaram2
1P. Sherubha, Department of Computer Science and Engineering, Karpagam Academy of Higher Education, Coimbatore, India.
2N. Mohanasundaram, Department of Computer Science and Engineering, Karpagam Academy of Higher Education, Coimbatore, India.
Manuscript received on 22 August 2019. | Revised Manuscript received on 05 September 2019. | Manuscript published on 30 September 2019. | PP: 1597-1606 | Volume-8 Issue-11, September 2019. | Retrieval Number: F3958048619/2019©BEIESP | DOI: 10.35940/ijitee.K3958.0981119
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Wireless Sensor Networks (WSNs) are deployed generally in a hostile environment, where an adversary captures some nodes that are physically connected in the network. It initially reprograms the nodes and makes them replicate into a number of clones, thereby having control over them. In order to provide a distributed solution to resolve the above specified problem specified above, a framework based on Authentic Node Placement based Message Verification and Passing Strategy (ANP-MVPS) is proposed. Some of the solutions offered by existing techniques are not satisfactory due to Energy and Memory constraints. This turns to be a serious drawback for protocols used in WSN’s resource constrained environment. In this work, three diverse factors are considered for investigation. They are: Firstly, modeling of Authentic Node Placement based Message Verification and Passing Strategy (ANP-MVPS) is performed to identify the distributed mechanism of clone in a network and prevent the replication of clone among them. Secondly, the parameter selection Probability of Occurrence of IP, Mean Time Intervals, Time to Live, ACK value, Time Stamp Field, SYN value, Differentiated Service Field and Sequence Number are considered before performing classification. Thirdly, an efficient Naive Bayesian classifier for security analysis based on trust value (NB-TV) is used to estimate the performance metrics like accuracy, sensitivity, specificity, F-measure, Recall etc. This method shows satisfactory results when compared to existing techniques. The simulation was carried out in MATLAB environment. The proposed method shows better trade off in contrast to prevailing techniques.
Keywords: Wireless Sensor networks; Clone attack; Authentic Node Placement based Message Verification and Passing Strategy; Naive Bayesian classifier for security analysis; Accuracy, Trust values.
Scope of the Article: