Laplace Stleltjes Transform based Conditional Survivability Coefficient Model for mitigating Selfish Nodes in MANETs

Abstract

In MANETs, the cooperation is considered as an important entity for enabling reliable data dissemination among the mobile nodes. But, the existence of selfish nodes weakens the degree of cooperation and in turn reduces the network performance. Hence, the computation of reputation level for each and every node in the network becomes essential in order to make optimal routing decisions. In this paper, we propose a Laplace Stleltjes Transform based Conditional Survivability Coefficient Model (LCSCM), which manipulates the survivability of the network through a parameter called Conditional Survivability Coefficient (CSC). This Conditional Survivability Coefficient aids in determining the reputation level of mobile nodes as well as quantifies the survivability of the entire network. The performance of this conditional probabilistic approach is analyzed using ns-2 based on the network related parameters such as packet delivery ratio, throughput, total overhead, and control overhead by varying the number of mobile nodes in the network. The results obtained through these extensive simulations make it obvious that, this approach outperforms PCMA model with a successful detection rate of 24%. This LCSCM also facilitates in framing 0.25 as the saddle point for selfish node detection.