论文部分内容阅读
在对传统LMS算法、变步长SVSLMS算法及归一化LMS算法分析的基础上,提出了一种改进的归一化变步长LMS算法即N-SVSLMS(Normalized-SVSLMS)算法。该算法结合了参考文献中两种算法的思想,得到了改进的归一化LMS自适应算法。该算法在信道环境多变的情况下,收敛速度和稳定性能有了进一步的提高。理论分析及计算机仿真结果表明,N-SVSLMS算法明显优于传统LMS算法、变步长SVSLMS算法及归一化的LMS算法。
Based on the analysis of traditional LMS algorithm, variable-step SVSLMS algorithm and normalized LMS algorithm, an improved normalized variable-step LMS algorithm called N-SVSLMS (Normalized-SVSLMS) algorithm is proposed. The algorithm combines the ideas of two algorithms in reference, and obtains an improved normalized LMS adaptive algorithm. In the case of changing channel environment, the convergence speed and stability of the algorithm have been further improved. The theoretical analysis and computer simulation results show that the N-SVSLMS algorithm is obviously superior to the traditional LMS algorithm, the variable step size SVSLMS algorithm and the normalized LMS algorithm.