Real Value Negative Selection Algorithm with the n-Dimensional Chaotic Map

In the detector generation phase for traditional chaos negative selection algorithm based on the binary, taking chaotic map generates chaotic sequences, and then doing discrete chaotic sequences generates candidate detectors. This method has many problems which are the bad analysis of knowledge and data, the low detection efficiency, and the low generation rate of detectors and so on. So a chaos negative selection algorithm based on real value is proposed. On the one hand, it leads into chaos theory and takes self-map which is the better chaotic feature to construct N-dimensional chaotic map in order to generate the center of candidate detectors. This improves the traditional generation mechanism of detectors and adapts more to handle high dimensional space problems. On the other hand, it optimizes the original V-detector algorithm and determines the detection radius with the idea of combining the directional movement and calculation of the geometric center. To the greatest extent, the aims are the maximization of the radius value, the expansion of the coverage area and the reduction of the detector quantity under the premise of satisfying the predetermined coverage rate. Experiment results show that the algorithm improves the speed of the detector generation and the detection efficiency of the detector set.