一种基于子空间的多径CDMA信道估计方法

Subspace-Based Multipath Channel Estimation for CDMA Communication Systems

提出了一种单径或多径 CDMA信道参数的估计方法 .该方法利用了接收信号协方差矩阵的特征结构 ,把观测子空间分解成为信号子空间和噪声子空间 .利用信号向量在噪声子空间的投影为零的性质 ,可以估计出信道的参数 .设计了一个代价函数 ,满足此代价函数就可使信号向量在噪声子空间的投影为零 .与 S.E.Bensley等人 1 996年提出的估计方法相比 ,本文提出方法可以得到信道估计的闭式表达 ,有利于进一步的分析 .仿真结果表明 ,该方法有很强的抗远近效应及多址干扰的能力 .

An estimation method of channel parameters for CDMA systems operating over channel with single or multiple propagation paths is proposed.This method exploits eigenstructure of thd received signal's sample correlation matrix in which the observation space can be partitioned into a signal subspace and a noise subspace.The channel estimation is formed by projecting a give user's spreading waveform into the estimated noise subspace and nulling the projection.Furthermore,a cost function which can null the noise compontnt of signal vector is presented.Compared with the method presented by S.E.Bensel et al (1996),this method can get the close form solution which can be used for further analysis.Simulations state this estimation has strong ability against near far effect and MAI.