研究了二阶Volterra滤波器的一种乘积耦合近似实现结构及其非线性NLMS自适应算法 ,并用这种少参数二阶Volterra滤波器 (RPSOVF)研究了一些混沌信号的非线性自适应预测性能 .仿真研究结果表明 :所给出的非线性NLMS自适应算法能够保证这种RPSOVF的稳定性和收敛性 ,且RPSOVF用这种非线性NLMS自适应算法能够自适应预测一些混沌时间序列 . A product coupling approximation structure of second order Volterra filter and its nonlinear NLMS adaptive algorithm are studied. And the nonlinear adaptive predictive performance of some chaotic signals is studied by using the second - order Volterra filter with fewer parameters (RPSOVF) The simulation results show that the proposed non-linear NLMS adaptive algorithm can guarantee the stability and convergence of this RPSOVF, and RPSOVF can adaptively predict some chaotic time series with this non-linear NLMS adaptive algorithm.