EEG Source Localization Based On Simulated Annealing Genetic Algorithm

Studies on the human brain characteristic is the Interdisciplinary studies which bases on the physics, the biology, the body study, the psychology and the information theory .It is a front domain of current scientific and technical.To survey, analyze and imitate the function of cerebrum from the project discipline angle is a key research, more and more attaching importance to the international academics.The localization question about brain power source, namely inverse problem based on EEG signal, is realizes the information of human brain interior electricity activity according to the electric potential of scalp observation. It arise people's widespread interest and its research results will make important impact on person's feeling, cognition process and nerve disease diagnosis.In this paper, the methods for solving the EEG inverse problem are investigated in details and the research work is focused on the following several aspects:1) Analyze the insufficient of SA and GA and put forward a Hybrid genetic algorithm, which based on the simulated annealing .It absorbs local search capability of GA and global search capability of SA. It avoids effectively the common defects of early convergence and the worse partial search. Results from case studies show that the improvement is effective.2) Bring forward adaptive arithmetic to avoid prematurity problem and be tested using Schaffer2 function.3) The hybrid genetic algorithm has been developed in solving the parameters of dipoles. The experimental result indicated that this method enhanced solution speed and the precision and provided a way to the EEG problem.4) Advantageous for teachers and students to share material, this article proposed a management system of graduate design material, which based on mixed B/S and C/S Advantageous for teachers and students to share material, this article proposed a management system of graduate design material, which based on mixed B/S and C/S.