OFDM系统中基于BP神经网络的定位算法  

Location algorithm based on BP neural network in OFDM system

作　　者：毛永毅[1] 李成[1] 张宏君[2] 

机构地区：[1]西安邮电学院电子工程学院,西安710061 [2]西安邮电学院通信与信息工程学院,西安710061

出　　处：《计算机应用》2012年第9期2426-2428,共3页journal of Computer Applications

基　　金：陕西省自然科学基金资助项目(2009JM8015);陕西省教育厅专项科研基金资助项目(2010JK815)

摘　　要：为了减小正交频分复用(OFDM)系统中多径干扰对定位精度的影响,提出一种基于后向传播(BP)神经网络的定位算法。该算法采用多重信号分类(MUSIC)算法估计OFDM信号首径的到达时间(TOA),再计算出到达时间差(TDOA),然后利用BP神经网络对其进行修正,最后使用Chan算法确定移动台的位置。在多径环境下对算法进行仿真,仿真结果表明该算法能够有效地降低多径干扰的影响,性能优于最小二乘(LS)算法、Chan算法和泰勒算法。For the purpose of reducing multi-path interference for positioning accuracy in Orthogonal Frequency Division Multiplexing （0FDM） systems, a location algorithm based on Back Propagation （BP） neural networks was proposed. MUltiple Signal Classification （MUSIC） algorithm was used to estimate the Time Of Arrival （TOA） of the first arrival path and calculate the Time Difference Of Arrival （TDOA）. Then BP neural network was used to correct the TDOA. Finally Chan algorithm was used to determine the location of the mobile station. The location algorithm was simulated in multi-path environment. The simulation results show that this algorithm can effectively reduce the effect of the multi-path interference and the performance is better than Least Square （LS） algorithm, Chan algorithm and Taylor algorithm.

关 键 词：正交频分复用 到达时间差 神经网络 多重信号分类算法 多径干扰 

分 类 号：TN929.53[电子电信—通信与信息系统]