采用时域测量矩阵的压缩感知稀疏信道估计方法

针对传统最小二乘和伪随机序列相关信道估计方法在稀疏信道应用时估计精度差的问题,提出一种采用时域测量矩阵的压缩感知稀疏信道估计方法.新方法首先将循环前缀单载波分块传输系统中的稀疏信道估计建模为一个典型的压缩感知问题,然后利用具有最优循环相关特性的伪随机序列优化构造确定性压缩感知测量矩阵,避免了使用随机测量矩阵造成的存储不便及估计性能差的问题,且提高了信道估计性能.基于准静态COST 207典型城市信道模型的仿真结果表明:该估计方法能够有效地降低稀疏信道的估计均方误差,在16 dB处的误码率可达2×10-5,而相同情况下最小二乘信道估计方法的误码率只能达到3×10-3.

An Compressed Sensing Estimation Method for Sparse Channels Using Time Domain Measurement Matrix

Since the estimation accuracy of traditional least square and pseudorandom binary sequence correlation based channel estimation methods are not satisfactory when applied in sparse wireless channels,a compressed sensing sparse channel estimation method is proposed for cyclic prefixed single carrier block transmission(CP-SCBT) system by using a time domain measurement matrix.The new method first formulates the sparse channel estimation problem in CP-SCBT system as a typical compressed sensing one,then utilizes a deterministic Pseudorandom binary sequence with the optimal cyclic autocorrelation to minimize the mutual incoherence property(MIP) of the measurement matrix,so that the storage inconvenience of the random measurement matrix is avoided,and the recovery performance is improved.Computer simulations based on quasi-static COST 207 typical urban channel model show that the proposed compressed sensing channel estimation method can greatly reduce the mean square error of the estimated channel,and achieve a bit error rate of 2×10-5 when the signal to noise ratio is 16 dB,while the traditional least square estimation method only achieves a bit error rate of 3×10-3 in the same scenario.