A New Crossover Methods and Fitness Scaling for Reducing Energy Consumption of Wireless Sensor Networks

Abstract

Efficient energy in the wireless sensor networks (WSNs) is a critical issue because sensor nodes are equipped with one-time or low-energy batteries. In these networks, efficient energy saving methods involve clustering network nodes to avoid long-distance communications with base stations (BS) to conserve their energy over a long period of time and extend their lifetimes. In other words, the choice of cluster heads (CHs) to improve routing and energy efficiency plays a central role in extending the network lifetime. This study proposes a new central cluster algorithm based on an improved genetic algorithm (EGA) that finds appropriate numbers of CHs in networks. This enhancement concerns about the application of two new crossover methods: Whole Arithmetic Crossover (WOX), and Local Crossover (LX) methods. This study explored the impact of the two aforementioned crossover methods on WSN energy efficiency and the effect of applying the scaled fitness function on network lifetime. For evaluation, we conducted a comparison between the crossover methods WOX and LX with three crossover methods; Simple Arithmetic Crossover (SMX), Single Arithmetic Crossover (SNX), and Discrete Crossover (DX) considering fitness scaling or without fitness scaling to identify the method that influences energy consumption. The results were then compared with the Low Energy Adaptive Clustering Hierarchy protocol (LEACH). All the simulation experiments were performed in MATLAB. The simulation results reveal that WOX and SNX with a scaled fitness function lead to a longer network lifetime by selecting CHs with longer lifetimes than the SMX, DX and LX methods. As a result, the proposed method exhibited better performance in terms of the power consumption and throughput rate.

Keywords

• Fitness-scaling
• genetic algorithm (GA)
• local crossover (LX)
• whole arithmetic crossover (WOX)
• wireless sensor network (WSN)