An Erlang factor-based conditional reliability mechanism for enforcing co-operation in MANETs

Abstract

Reputation is considered to be one of the vital entities for maintaining collaboration among wireless mobile nodes present in an ad hoc environment. The nodes in MANET are dynamic and could change its behaviour drastically, but establishing maximum level of cooperation between these nodes is highly crucial. Moreover, the presence of selfish nodes has a greater impact on the resilience of the network. Hence, a need arises for formulating a mechanism that deals with these selfish nodes. In this paper, we contribute an Erlang distribution based Conditional Reliability Mechanism (ECRCM) that aids in detecting and isolating the selfish nodes present in an ad hoc environment. This mathematical model makes the routing decision with the help of a parameter called Erlang factor based Conditional Reliability Coefficient (ECRC) determined for each and every mobile node present in the ad hoc network. Extensive simulations were carried out through ns-2 simulator and the analysis was based on performance metrics such as packet delivery ratio, throughput, control overhead and total overhead. ECRCM also helps in framing an optimal threshold range for selfish node detection. From the results, it is obvious that the threshold range derived in our study remarkably identifies maximum number of selfish nodes when compared to the selfishness detecting parameters available in the literature.

Quantile Approximation of the Erlang Distribution using Differential Evolution Algorithm

“…Aside from blocking, Erlang distribution, on the other hand, has been applied in outage detection [19]. Other similar applications are in modeling waiting times in public cloud [20], managing selfish nodes in ad hoc networks [21], optimal power allocation in channel [22] and others [23]. Erlang distribution is applied in modeling the vehicular traffic [24], which are often modeled and studied as networks [25].…”

Hybrid Energy and Trust Efficient Reactive Routing Mechanism for MANETs

“…This Erlang Factor-based selfish node detection approach was also determined to be superior in terms of packet delivery rate, throughput, total overhead and control overhead under a varying number of mobile nodes and selfish nodes in the network. Finally, a Trust-based selfish node detection approach that utilizes the merits of conditional probability was proposed for effective isolation of selfish intent characteristics in the network [14]. This proposed conditional probability-based detection approach was also determined to a significant in exploring all the dimensions that attribute towards effective isolation of intentional characteristics of the network In addition, a significant contextual trust-based selfish nodes detection approach was contributed for eliminating the impacts of selfish behavior from the network [15].…”