基于信号子空间的盲自适应多用户检测

多用户检测技术是DS—CDMA中解决多址干扰问题的一种重要技术。针对基于PASTd多用户检测方法在子空间追踪收敛速度慢和收敛的不稳定问题,提出一种快速子空间追踪的盲自适应多用户检测改进算法。该算法通过空间分解得到信号子空间,并利用信号子空间的特征值和特征向量得到解相关多用户检测器;最后,应用归一化正交Oja快速自适应地追踪信号子空间。仿真结果表明,相对PASTd多用户检测方法,降低计算复杂度,加快算法的收敛速度,系统稳定性好,拟制多址干扰具有良好的性能,达到很好的多用户检测效果。

Signal Subspace Based Blind Adaptive Multiuser Detection

Multiuser detection is the key technology to suppress the multiple access interference in direct sequence CDMA system. As the procession of tracking subspace of the PASTd multiuser detection is slow convergence and oscillates in most cases, an improved algorithm for the fast estimating channel and the exact tracking principal subspace multiuser detection is proposed. The algorithm gets the signal sunspace by the space decomposition, and then the decorrelation multiuser detection is given by the eigenvectors and the eigenvalues of the signal subspace. Using the normalized version of orthogonal Oja method tracks the signal subspace fast and adaptively. The computer simulation illustrates that the algorithm can suppress the multiple access interference and the inter-symbol interference, decreases the computational complexity and increases the speed of convergence, the system stability and the better effect of the detection.