Immune Network Theory Based Artificial Immune System and Its Application for Pattern Recognition

Natural immune system is a complex of cells, tissues and organs that work together to protect the host against foreign invaders' attacks. Immune network theory proposed by Jerne indicates that the immune system, even in the absence of foreign stimuli, must display an eigen-behavior resulting from cell-cell interactions within the immune system. In this paper, motivated by the natural immune principles, we propose an artificial immune system model based on the immune network theory. Unlike other immune network theory based models, in which an immune cell at idiotypic level i interacts with populations at levels i 1 and i + 1, we propose the new model, in which antibodies at idiotypic level i interacts with populations at constant levels a1 (index of the stimulation cell) and aN (index of the inhibition cell). Simulation results of pattern recognition have demonstrated the effectiveness of the novel artificial immune system.