BoDMaS: Bio-inspired Selfishness Detection and Mitigation in Data Management for Ad-hoc Social Networks

Abstract

Existing data management protocols for socially-aware networks assume that users are cooperative when participating in operations such as data forwarding. However, selfishness as a non-cooperative act of misbehavior can seriously degrade network performance and fairness, particularly in Ad-hoc Social Networks (ASNETs). Therefore, detecting and counteracting selfishness on performance of cooperative users are crucial to the success of ASNETs. In this paper, we propose BoDMaS, a biologically inspired method, to detect and mitigate the impact of node selfishness on data management performance and efficiency of ASNETs. In design of BoDMaS, we consider social willingness (which depends on depth of social relationship among users) as a social behavior and bacteria chemical products as a counter to achieve optimal ASNETs performance. Counter is a parameter attached to individual user counting successful data operations performed in relation with others. Using social willingness and counter, BoDMaS assesses and classifies users, and counteracts their selfishness. BoDMaS is evaluated from different aspects demonstrating its ability to accurately detect and counteract selfishness in replication operations for ASNET environments.