一种基于正交多小波的自适应均衡算法

A NEW ADAPTIVE EQUALIZATION ALGORITHM BASED ON ORTHOGONAL MULTIWAVELETS

该文提出了用正交多小波来表示均衡器,由于多小波可同时具有正交性、紧支性和线性相位等特点,因此经多小波变换后所得到的信号相关阵的稀疏化估计与单小波变换相比非零元素较少,边界效应减小,基于此,文中给出了正交多小波变换域的一种Newton-LMS类自适应均衡算法,其计算复杂性可通过有预处理的共轭梯度法进一步降低为O(N log N),仿真结果表明了该算法收敛速度较快,且易于实时实现。

t A new equalizer represented by a set of orthogonal multiwavelets is presented. Since multiwavelets can be orthogonal, compactly supported and linear phase, the multiwavelets transformed correlation matrices have less non-zero elements and smaller boundary effects than that of wavelet. So, a new multiwavelet transform domain Newton-LMS adaptive equalization algorithm is described, and its complexity is O(NlogN) by using the preconditioned conjugate gradient algorithm. Simulation shows its convergence speed is faster and its realization is easier.