基于GA-BP神经网络的超宽带室内定位  

Ultra-Wideband Indoor Positioning Based on GA-BP Neural Network

作　　者：薛阳阳 沈重[1] Xue Yangyang;Shen Chong(College of Information and Communication Engineering,Hainan University,Haikou 570228,China)

机构地区：[1]海南大学信息与通信工程学院,海南海口570228

出　　处：《海南大学学报（自然科学版）》2021年第3期235-241,共7页Natural Science Journal of Hainan University

摘　　要：超宽带(UWB)室内定位的精度主要取决于复杂环境产生的非视距(NLOS)误差以及多径干扰,NLOS误差的降低是进行研究的重点之一.将遗传算法(GA)和BP神经网络结合应用于UWB室内定位中,使用BP神经网络提升UWB室内定位系统的抗NLOS能力,利用遗传算法可以将全体解搜索出来的特性,使得网络的输出结果为权值和阈值全局最优解,提高了定位精度.由实验结果可知,在三基站构成的定位场景中,GA-BPChan算法预测轨迹和实际轨迹基本吻合;通过对比算法在不同TDOA标准差下的均方根误差(RMSE)可知,GA-BP-Chan算法的抗NLOS能力最强,且相比于BP-Chan算法定位精度提升了16.12%.The accuracy of ultra-wideband(UWB)indoor positioning mainly depends on the non-line-of-sight(NLOS)error and multipath interference generated by the complex environment.The reduction of the error is one of the focuses of this research.In the report,Genetic algorithm-back propagation(GA-BP)neural network was used for UWB indoor positioning,BP neural network was used to improve the anti-NLOS ability of UWB positioning system,the genetic algorithm was used to search out all solutions to avoid the result of BP neural network as a local optimal solution,which can improve the positioning accuracy.The experimental results indicated that the predict trajectory with GA-BP-Chan algorithm conform to the actual trajectory in the three-base station positioning scenario,the capacity of antiNLOS of GA-BP-Chan algorithm is the strongest,with a higher positioning accuracy of 16.12%than BP-Chan algorithm.

关 键 词：超宽带 室内定位 GA-BP神经网络 权值和阈值 NLOS 

分 类 号：TN919.72[电子电信—通信与信息系统]