Memetic algorithm and its application to function optimization and noise removal

Memetic Algorithm is a metaheuristic search method. It is based on both the natural evolution and individual learning with information transmission among them. In the present paper, Genetic Algorithm, due to its good exploration capability is taken as main algorithm and chemotaxis mechanism of Bacterial Foraging Optimization (BFO) is used as local search. The memetic process is realized using BFO by imitating the nutrient information from the bacteria of the best fitness. The proposed variant of memetic algorithm is tested on the standard benchmark functions of various dimensions with unimodal and multimodal property. When the results are compared, the proposed memetic algorithm shows better performance than GA and BFO. The performance of the proposed memetic algorithm is better in terms of speed of convergence and quality of solutions. The developed MA and BFO are used for the Gaussian noise removal using the Blind Source Separation (BSS) based on Independent Component Analysis (ICA).