论文部分内容阅读
本文绘出了一个自适应滤波器的修正的时域正交性算法.计算机模拟结果表明,该算法有两个特性:一、相同的收敛因子c可以在比较宽范围的输入信号信噪比内使用。二、应用这种方法的噪声消除器的一个实用特性是,对于相同的期待信号,输入的参考信号的功率变化,不影响收敛因子c值的选择,这就使这种噪声消除器的应用更灵活。文中给出了这种算法用于谱估值时的特性,并与LMS算法进行了比较。
In this paper, we propose a modified time-domain orthogonality algorithm of adaptive filter. Computer simulation results show that the algorithm has two characteristics: one, the same convergence factor c can be within a relatively wide range of input signal SNR use. A practical feature of noise cancellers applying this method is that the power variation of the incoming reference signal for the same expected signal does not affect the choice of the value of the convergence factor c, which makes this noise canceler more useful flexible. In this paper, the characteristics of this algorithm for spectral estimation are given and compared with LMS algorithm.