无线电工程

为了提高无线传感器网络(Wireless Sensor Network, WSN)的覆盖率，提出了一种基于相互学习能力和动态学习因子的改进粒子群优化(Modified Partide Swarm Optimization, MPSO)算法。引入了拉丁超立方采样(Latin Hypercube Sampling, LHS)序列来初始化种群，增加了种群的多样性，为之后优化奠定基础；引入一种相互学习方法，粒子通过随机选择目标粒子来增强自身的学习能力，提升局部寻优性能；利用一种动态学习因子策略，通过改变粒子的学习能力，加快了算法收敛速度并增强了全局寻优能力。仿真结果表明，在不改变原算法复杂度的情况下，相较于基本PSO算法和其他对比算法，改进PSO算法可以耗费更少的资源达到更好的寻优效果，可以有效地解决网络覆盖盲区和覆盖冗余问题，提高网络覆盖率。

In order to improve the coverage of Wireless Sensor Network(WSN), a Modified Particle Swarm Optimization(MPSO) algorithm based on mutual learning ability and dynamic learning factors is proposed. First, the Latin Hypercube Sampling(LHS) sequence is introduced to initialize the population to increase the diversity of the population and lay the foundation for the subsequent optimization. Secondly, a mutual learning method is introduced, in which target particles are randomly selected to enhance their own learning ability and improve local optimization performance. Finally, a dynamic learning factor strategy is used to accelerate the convergence speed and enhance the global optimization ability by changing the learning ability of the particles. The simulation results show that compared with the basic PSO algorithm and other algorithms, the MPSO algorithm can consume less resources to achieve better optimization results, effectively solve the problems of network coverage blind area and coverage redundancy, and improve the network coverage.