Visible to the public NL-IDS: Trust Based Intrusion Detection System for Network Layer in Wireless Sensor Networks

TitleNL-IDS: Trust Based Intrusion Detection System for Network Layer in Wireless Sensor Networks
Publication TypeConference Paper
Year of Publication2018
AuthorsGhugar, U., Pradhan, J.
Conference Name2018 Fifth International Conference on Parallel, Distributed and Grid Computing (PDGC)
KeywordsBlack hole attack, Black hole attackers, Blackhole attack, Collaboration, Conferences, false trust, Intrusion detection, intrusion detection system, Mathematical model, Measurement, mobile ad hoc networks, network layer, NL-IDS, periodic trust value, policy-based governance, protocol layers, pubcrawl, resilience, Resiliency, Routing, Routing protocols, Scalability, secure path, security issues, security system, sensor node trust, telecommunication security, Trust, trust based intrusion detection system, Wireless Sensor Network, Wireless sensor networks, WSN
Abstract

From the last few years, security in wireless sensor network (WSN) is essential because WSN application uses important information sharing between the nodes. There are large number of issues raised related to security due to open deployment of network. The attackers disturb the security system by attacking the different protocol layers in WSN. The standard AODV routing protocol faces security issues when the route discovery process takes place. The data should be transmitted in a secure path to the destination. Therefore, to support the process we have proposed a trust based intrusion detection system (NL-IDS) for network layer in WSN to detect the Black hole attackers in the network. The sensor node trust is calculated as per the deviation of key factor at the network layer based on the Black hole attack. We use the watchdog technique where a sensor node continuously monitors the neighbor node by calculating a periodic trust value. Finally, the overall trust value of the sensor node is evaluated by the gathered values of trust metrics of the network layer (past and previous trust values). This NL-IDS scheme is efficient to identify the malicious node with respect to Black hole attack at the network layer. To analyze the performance of NL-IDS, we have simulated the model in MATLAB R2015a, and the result shows that NL-IDS is better than Wang et al. [11] as compare of detection accuracy and false alarm rate.

URLhttps://ieeexplore.ieee.org/document/8745870
DOI10.1109/PDGC.2018.8745870
Citation Keyghugar_nl-ids:_2018