Genetic Algorithm-based Optimized Fuzzy Adaptive Path Selection in Wireless Sensor Networks

• Main Authors: Muhammad Akram, Muhammad Ashraf, Tae Ho Cho
• Format: Article
• Language: English
• Published: Sukkur IBA University 2018-06-01
• Series: Sukkur IBA Journal of Computing and Mathematical Sciences
• Subjects: fuzzy optimization; route selection; ; filtering; genetic algorithm; fuzzy
• Online Access: http://localhost:8089/sibajournal/index.php/sjcms/article/view/42

In Wireless sensor networks, energy efficiency can be achieved by adaptive choice of the data forwarding path to balance the energy dissipation in the network. This adaptive path selection is done through a fuzzy rule-based method given the input parameters. Due to uncertainty in reasoning and inferencing process and imprecision in the data, the fuzzy-based system becomes an ideal choice for the selection of the paths.  In fuzzy systems, the membership functions need to be optimized to make the best use of the fuzzy inferencing and improve the performance of the fuzzy system. Genetic algorithm-based fuzzy membership function optimization technique selects the optimal solution in a feasible time and saves from the hassle of manual intervention. Manual optimization efforts are unfeasible for common applications and take unlimited time and human expertise to optimize functions in an exhaustive search field. This technique assesses the fitness of the membership functions through simulation outcomes and optimizes them through genetic algorithm based evaluation process. The proposed scheme consists of three modules; The first module simulates the membership function in the given network model, the second module analyzes the performance efficiency of the membership functions through simulation, and the last module constructs the subsequent membership-function populations using GA techniques. The proposed method automatically optimizes the membership functions in the fuzzy system with little human intervention, requires minimal human expertise and saves ample time in the optimization process.