中文摘要：

传统遗传算法存在过早收敛及局部搜索能力差的缺点，在求解无线网络协作通信功率优化分配等NP难问题时难以求得最优解。通过小生境策略解决遗传算法过早收敛问题，引入复合形法提高局部搜索能力，构造了兼顾广度搜索与深度搜索的高性能混合算法，并对上述问题进行求解。实验结果表明，所提算法与已有算法相比有一定优势，有效延长了协作网络寿命，稳定性较好，分配的功率波动范围小。

英文摘要：

The traditional genetic algorithm, which has the shortcomings of premature convergence and poor local search ability, is hard to solve the power allocation problem（NP-hard problem） of wireless network cooperative communication. This paper conqueres the premature convergence by introducing niche strategy, and improves the local search capabilities by combining complex method, constructs a high-performance algorithm taking account of the breadth and depth in searching, and then solves the power allocation problem. The simulation results show that the proposed strategy is better than the existing algorithms. It is able to effectively extend network lifetime, generates more stable values, and performs a good stability.