Nonnegative adaptive filter

Abstract

The invention discloses a nonnegative adaptive filter, and belongs to the field of digital filter design. The filter is obtained by a cost function composed of a 10-norm and a logarithm absolute error function which minimize a coefficient vector. Minimization of the 10-norm speeds up convergence of a nonnegative adaptive filter estimation unknown system, while minimization of the logarithm absolute error function improves the robustness of the nonnegative adaptive filter. The nonnegative adaptive filter can be applied to an occasion in which an electronic or communication system is interfered by high pulse noise.

Description

technical field

[0001] The invention discloses a non-negative adaptive filter, which belongs to the field of digital filter design. Background technique

[0002] System identification is an important branch of adaptive signal processing. Traditional adaptive channel equalization, adaptive noise cancellation, adaptive echo cancellation, active noise control and many other problems can be attributed to system identification. In some applications, due to the limitation of the inherent physical characteristics of the system, it is necessary to impose non-negativity constraints on the optimal values ​​of the coefficients of the system to be estimated. The system identification problem under the non-negativity constraint is a specific form of the optimization problem under the non-negativity constraint. The optimization problem under non-negativity constraints is often involved in theory and engineering practice, and is one of the current research hotspots, such as non-negative l...

Images