模拟退火遗传算法对无线传感器网络部署研究

摘要

研究无线传感器节点部署优化问题,传感器网络节点的部署在一定程度上决定了网络的性能和使用寿命.传统的遗传算法在无线传感器节点部署优化过程中,由于交叉和变异的概率是固定的,易产生局部最优问题,导致部署不理想,网络生命周期过短.为了更好地优化网络部署,提高网络生命周期,提出了一种基于模拟退火遗传算法的无线传感器节点部署优化方法.方法将传感器节点部署转化为一个组合优化问题,网络节点离散成为网格,通过遗传算法进行最优部署方案的搜索,同时采用模拟退火算法对遗传算法的种群进行更新,提高了最优解的搜索速度.仿真结果表明,模拟退火遗传算法部署的效率高,网络存活的节点数更多,有效地延长了网络的生命周期.%Wireless sensor nodes deployment optimization problem is studied. Wireless sensor network deployment determines its capability and lifetime. Traditional genetic algorithm generates local optimal problem easily in wireless sensor nodes deployment optimization process because the probabilities of crossover and mutation are fixed, thus the results are not ideal, and network lifetime is too short. In order to optimize network deployment and improve the network life, this paper puts forward a wireless sensor nodes deployment optimization method based on genetic algorithm and simulated annealing algorithm. In this method, sensor nodes deployment optimization is transformed into combinatorial optimization problem, and network nodes are expressed as a grid, using genetic algorithm to search the optimal deployment, and simulated annealing algorithm is used to the population change, for the search speed improvement. Simulation results show that the proposed algorithm, compared with the simple genetic algorithm and simulated annealing algorithm, has higher efficiency in deployment and more alive nodes in network, which prolongs the network lifetime.