Modified Genetic Algorithm with Sorting Process for Wireless Sensor Network

Abstract

The genetic algorithm is widely used in optimization problems, in which, a population of candidate solutions is mutated and altered toward better solutions. Usually, genetic algorithm works in optimization problem with a fitness function which is used to evaluate the feasibility and quality of a solution. However, sometimes, it is hard to define the fitness function when there are several optimization objectives, especially only one solution can be selected from a population. In this paper, we modified genetic algorithms with a novel-sorting process to solve the above problem. Two algorithms, the classic genetic algorithm and newly proposed recently M-Genetic algorithm, are simulated and altered by embedding the novel-sorting process. Besides, both the algorithms and their alteration versions are applied into wireless sensor network for locating Relay nodes. The sensor node loss and package loss number are reduced in genetic algorithms with our sorting process compared to the original ones.