论文部分内容阅读
该文详细讨论了ＬＭＳ算法中输入信号相关性、加权数目和稳态均方误差的关系。通过公式推导，从理论上证明了增加加权数目并不能保证减小稳态均方误差。对于具体的输入信号，存在一个最佳（或准最佳）的加权数目，使稳态均方误差最小，再增加权数目，稳态均方误差有增大的可能。该文以强相关的直流信号和弱相关的正弦信号分别作为自适应滤波器的输入信号进行了计算机仿真实验，实验结果与公式推导结果一致。该理论为自适应滤波器设计时阶数的选择提供了理论指导，有实际意义
This paper discusses in detail the relationship between input signal correlation, weight number and steady-state mean square error in LMS algorithm. Through formula derivation, it is theoretically proved that adding weight numbers does not guarantee to reduce the steady state mean square error. For a particular input signal, there is an optimal (or quasi-optimal) weighting number that minimizes the steady-state mean-square error, increases the number of weights, and increases the steady-state mean-square error. In this paper, a computer simulation is carried out using the strong correlated DC signal and the weakly correlated sinusoidal signal respectively as the input signals of the adaptive filter. The experimental results are consistent with the results of the formula. The theory provides a theoretical guide for the choice of adaptive filter design order, and has practical significance