一种新颖的双线性自适应滤波器

A Novel Bilinear Adaptive Filter

在非线性系统领域,Volterra滤波器作为一种经典的工具得到了广泛的应用。但是这种自适应滤波器的数学计算量非常大,尤其是当滤波器的阶数较大时,其应用往往受到很大限制。为了克服这个缺点,介绍了一种新颖的双线性自适应滤波器。详细分析了该滤波器的收敛性。在系统数学模型、算法和性能方面对该滤波器与Volterra自适应滤波器进行比较。仿真结果表明,这种新颖的滤波器在稳态误差和收敛速率等方面都比Volterra滤波器更优异。

Volterra filter is known to be an efficient adaptive filter in a variety of applications for non-linear systems. However, the huge amount of mathematical computation limits its further applications, especially as orders of the filter increase. To overcome this disadvantage of the Volterra filter, a novel adaptive filter based on layered bilinear architecture was proposed, which exhibited better convergence performance in terms of convergence speed and steady-state error, compared with the conventional 2hal-order Volterra filter.