Home > Published Issues > 2018 > Volume 13, No. 8, August 2018 >

Multivariate Analysis for Fuzzy Correlated Node Behavior Detection in Wireless Sensor Network

Noor Shahidah 1 and A. H. Azni 2
1. Faculty Science and Technology, Universiti Sains Islam Malaysia (USIM), Nilai 71800, Malaysia
2. Islamic Science Institute, Universiti Sains Islam Malaysia (USIM), Nilai 71800, Malaysia

Abstract—Wireless Sensor Network depends highly upon the cooperation among the nodes behavior in transmission of packet data, messages and route discovery. Over open medium environment, nodes are free to move and may change their behavior arbitrarily. In the presence of misbehavior node in some cases, it may instigate its neighboring nodes to compromise with the misbehaved node. Thus, this has resulted to a spreading of correlated node behavior and the impact of this event may result in high severity in network performance. Therefore, fuzzy logic model is proposed to formulate the correlated node behavior in WSN. The formulation of correlated node behavior based on fuzzy logic function of peer nodes real parameter measurement is investigated to determine the status of the node and then the fuzzy neural network will model the correlated node behavior occurrence. The accuracy of the results is established via sensor network simulation. The result of this study is providing a fundamental guideline for network designer to understand the fault-tolerance in network topology.
 
Index Terms—Fuzzy logic system, correlated node behavior, wireless sensor network, fuzzy correlated function

Cite: Noor Shahidah and A. H. Azni, " Multivariate Analysis for Fuzzy Correlated Node Behavior Detection in Wireless Sensor Network," Journal of Communications, vol. 13, no. 8, pp. 430-435, 2018. Doi: 10.12720/jcm.13.8.430-435.