关键词: 临近空间通信网络, 多目标优化, 非支配排序遗传算法

Abstract: In the deployment of near space communication network for specific area, factors such as resource allocate, coverage rate and load power should be taken into account. Considering the deficiency of exiting methods only optimizing single objective, and using the Non-dominated Sorting Genetic Algorithm 2（NSGA2） in solving practice problems, a multi-objective optimization design method based on improved-NSGA2 is presented. Analyzed the downlink performance, combined the target area’s demand distribution, the model is established with objectives of maximizing match degree and coverage rate, and minimizing power cost. The dynamic opposition-based learning mechanism and the differential local mutation operator are introduced into NSGA2. Tests prove that solutions obtained from improved-NSGA2 perform better both in convergence and distribution. The effectiveness of the method is verified by simulation, which provides a reference for the network’s actual deployment.

Key words: near space communication network, multi-objective optimization, Non-dominated Sorting Genetic Algorithm 2（NSGA2）