Genetic Algorithm Optimization for Fault Tolerance and Reliability in Wireless Sensor Networks

Le résumé fourni par la source

The WSNs are frequently considered to be in unpredictable communities where the removal of nodes and disruption of communications may hamper network performance. This study develops a new Genetic Algorithmbased Fault-Tolerance Optimization (GA-FTO) framework that seeks to be proactive in terms of augmenting reliability without the need of deploying redundant nodes or using static recovery schemes. The method includes adaptive chromosome encoding, multi-objective fitness functions middle-ground fault tolerance, energy efficiency, and latency, self-learning mutation operator informed by real-time fault diagnostics. Compared to already implemented GA-based approaches, the proposed system utilizes predictive fault modeling and localized re-clustering that is initiated due to early anomaly gradation. To justify adaptability, scenario simulations are performed with different levels of faults, transmission radius, and energy limits. The experimental results illustrate the positive impacts of a substantial enhancement in the network connectivity maintenance, the extension of lifetime, and the reduction of data loss during the high-fault situations. This paper lays the foundation of a functioning and robust optimization approach to mission-critical applications on WSNs under unfavorable/resource-constrained conditions.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.