基于线性预测的盲多用户检测技术的研究与应用

STUDY AND APPLICATION OF BLIND MULTI-USER DETECTION TECHNOLOGY BASED ON LINEAR PREDICTION

CDMA系统多用户检测技术(MUD)是第三代移动通信系统的关键技术之一。常规CDMA用户检测方法是使用期望用户的扩频序列对接收信号直接进行相关运算,再经过符号硬判决得到检测符号,这样的处理方式将解扩后的多址干扰成分看作噪声而未加任何处理,导致误码率增加。围绕自适应信号处理及盲多用户检测两方面技术,对CDMA系统的最佳盲多用户检测、次佳盲多用户检测进行研究,提出了一种具有工程实用价值的基于线性预测的盲多用户检测方法。仿真结果表明,在低SNR情况下,其运算复杂度大为降低、BER性能明显优于其他两种盲多用户算法。

MUD（Multi-User Detection） for CDMA system is the key technology in the third generation （3G） mobile communication system. The conventional way is that the received signal directly does correlated operation with the spreading code of the desired user and then hard-decision is used to calculate the check symbol. The multiple access interference components incurred from de-spreading is regarded as noise with no treatment being imposed on in that way, so it will increase bit error rate. A blind multi-user detection method with practical engineering value is given based on linear prediction from two aspects of the adaptive signal processing technology and the blind Multi-user detection technology through the investigation of the optimal and suboptimal blind multi-user detection technology in CDMA system, By computer simulation, the BER performance of the proposed detection method obviously surpasses other two blind multi-user algorithms in low SNR （ signal-noise ratio） condition with significant reduction of computation complexity.