基于小波包变换的非线性信道均衡器

Nonlinear Channel Equalizer Based on Wavelet Packet Transform

针对严重线性失真和轻度非线性失真的数字信道,为了提高基于最小均方误差算法的判决反馈均衡器的收敛速度,首先提出用一族正交小波包基函数来表示非线性信道判决反馈均衡器及其输出,然后给出基于小波包变换的非线性信道自适应均衡算法。该算法实现了小波包与非线性信道模型的结合,在计算量增加不多的前提下,利用小波包对小波空间的进一步划分以及比小波变换更强的去相关能力来减小输入信号相关阵的条件数。对典型非线性信道模型的仿真结果表明,该算法可有效提高均衡器的收敛速度。

To improve the convergence speed of the decision feedback equalizer for the nonlinear channel with the severe linear distortion and the mild nonlinear distortion, a equalizer and its output are expressed by a family of orthogonal wavelet packet functions. Then, a new decision feedback adaptive equalization algorithm based on the wavelet packet transform is presented. Since the wavelet packet transform has a better decorrelation ability than the wavelet transform in the combined nonlinear channel model, the convergence speed of the new equalization algo- rithm is improved by the normalization least mean square algorithm. Computational complexity increases a little for the fast wavelet packet transform. Simulation result on the typical nonlinear channel model shows that its convergence speed is faster.