A chaotic immune niche genetic algorithm for target signal selection in large scale wireless sensor networks

The advance of micro-sensor, nano-systems and networking technologies shown a great potential of small-size, low energy consumption, low storage, and self-adaptation sensors. Large scale wireless sensor networks (LSWSNs) consists of some them that have sensing, computation, wireless communication, and free-infrastructure abilities. The target signal selection problem in LSWSNs attracts attention of people from academic researchers, industry, and military department. The target signal selection scheme is usually designed for LSWSNs to enhance the percentage of detected targets. However, the target signal selection problem can be formulated as a nonlinear mixed integer optimization problem, which is hard to solve. In this paper, we propose a chaotic immune niche genetic algorithm (CINGA) based target signal selection approach for maximizing the percentage of detected targets. We first formulate our objective function to maximize the percentage of detected targets under multiple constraints. The proposed algorithm combines the advantages of the high efficiency of immune operation and the global search ability of the chaotic generator. An analysis is given to show the correctness of CINGA, and simulations are conducted to demonstrate the performance improvement of CINGA against parallel genetic algorithm (PGA) and ant colony optimization (ACO). Although sub-optimal for LSWSNs, simulation results show that the proposed CINGA allows to reach higher monitoring percentage compared to PGA and ACO approach. Furthermore, it was found that the immune operation helps evolution to avoid local optima.