针对现有正交时序复用(Orthogonal Time Sequence Multiplexing,OTSM)通信均衡方法在高速移动环境下难以有效处理符号间干扰的问题,提出一种基于最小二乘QR分解(Least Square QR,LSQR)的双迭代判决反馈均衡方法.该方法利用基于零填充的OTSM系统时域传输矩阵的分块特性,使用LSQR迭代算法对时域观测数据进行逐块均衡以实现对待检测符号的快速粗估计;随后将其转换至延迟-序列域中通过数次迭代的方式进行动态门限量化判决,进而利用判决反馈结果对延迟-序列域观测数据进行分块干扰消除,最后通过LSQR迭代均衡实现符号粗估计的迭代优化.仿真结果表明,在500 km/h移动速度下该均衡方法与目前广泛使用的基于高斯赛德尔(Gauss-Seidel,GS)的均衡方法和基于最大比合并(Maximum Ratio Combining,MRC)的均衡方法相比,采用16-QAM调制且误码率为10-4时分别带来了2.61dB和1.76dB的性能增益.
Abstract
Aiming at the problem that the existing orthogonal time sequence multiplexing (OTSM) communication equalization method is difficult to effectively deal with inter-symbol interference in high-speed mobile environment,a dou-ble iterative decision feedback equalization method based on least squares QR (LSQR) decomposition is proposed. The method utilizes the block characteristics of the time-domain transmission matrix of the OTSM system based on zero pad-ding,and uses the LSQR iterative algorithm to equalize the time domain observation data in blocks to realize the rapid pre-liminary estimation of the signal symbol to be detected. Then,it is converted to the delay-sequence domain for dynamic threshold quantization decision by several iterations,and then the decision feedback results are used to eliminate block inter-ference on the delay-sequence domain observation data. Finally,the iterative optimization of symbol initial estimation is re-alized by LSQR iterative equalization. The simulation results show that at the moving speed of 500 km/h and compared with Gauss-Seidel equalization method and maximum ratio combining equalization method,the performance gains of 2.61dB and 1.76dB are obtained respectively by using 16-QAM modulation with the bit error rate at 10-4.
关键词
正交时序复用/延迟-序列域/动态门限判决/干扰消除/LSQR迭代均衡
Key words
orthogonal time sequence multiplexing/delay-sequence domain/dynamic threshold decision/interference elimination/LSQR iterative equalization
维普
万方数据