基于爬山粒子群优化的移动传感网络定位算法  

Novel mobile wireless sensor network localization algorithm based on hill climbing particle swarm optimization

作　　者：吉小洪 魏开平[1] 胡文杰[2] 

机构地区：[1]华中师范大学计算机学院,武汉430079 [2]咸宁职业技术学院信息工程学院,湖北咸宁437100

出　　处：《计算机工程与应用》2016年第5期84-88,共5页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61170305)

摘　　要：针对蒙特卡洛定位(Monte Carlo Localization,MCL)采样效率不高,定位精度较低的问题,提出一种新的基于爬山法优化策略的移动无线传感网络定位算法HCPSO-MCL(Hill Climbing Particle Swarm Optimization-MCL),将节点定位问题转化为全局优化问题。HCPSO-MCL算法采用基于爬山策略的混合粒子群优化算法对MCL的估计值进行修正,从而实现节点快速准确定位。实验仿真结果表明,HCPSO-MCL较之于MCL算法在定位精度上有很大改进,而且比PSO-MCL(Particle Swarm Optimization-MCL)算法有更快的收敛性。The Monte Carlo localization algorithm suffers the problem of low sampling efficiency and low localization accuracy. Hence a novel mobile wireless sensor network localization algorithm, HCPSO-MCL, based on hybrid particle swarm optimization combined with hill climbing is proposed. It converts the localization problem to global optimization problem. The HCPSO-MCL algorithm is used to correct the estimated localization of MCL, it can determine the location quickly. The simulation results show that, compared with MCL algorithm, HCPSO-MCL has higher localization accuracy and is faster than PSO-MCL in convergence speed.

关 键 词：无线传感网络 蒙特卡洛定位 爬山粒子群算法 

分 类 号：TP273.5[自动化与计算机技术—检测技术与自动化装置]