基于奇异值和奇异向量的盲自适应多用户检测方法

Singular value and singular vector based blind adaptive multiuser detector

提出了一种基于奇异值和奇异向量的盲自适应多用户检测方法。传统的基于子空间的多用户检测方法需要估计信号子空间的特征值和特征向量 ,收敛速度较慢。这种新方法通过信号子空间的奇异值和奇异向量得到 CDMA系统的线性最小均方误差 (MMSE)多用户检测器。论文采用FST算法跟踪信号子空间的奇异值和奇异向量 ,并通过信号能量和噪声能量比值来确定信号子空间的阶数 ,使得多用户检测器能够很快地收敛 ,多用户检测器的输出信噪比很高 ,达到很好的多用户检测效果。

A singular value and singular vector blind multiuser detector is proposed in this paper. The conventional multiuser detector is based on subspace requires eigenvectors and eigenvalues of the signal subspace so it converges slowly. The proposed detector uses singular values and singular vectors of the signal subspace in a linear MMSE multiuser detector for the DS CDMA system. The method uses the FST algorithm to estimate singular values and singular vectors and uses the ratio between the energy subspace and the signal subspace to estimate the rank of the signal subspace. The multiuser detector proposed converges quickly and the multiuser detector output can achieve a high SNR.