Mixed norm non-negative adaptive filter

Abstract

The invention discloses a mixed norm non-negative adaptive filter, and belongs to the field of digital filter design. The filter is established by minimizing a non-negative constraint condition and apiecewise cost function. The non-negative constraint is adopted, so that the adaptive filter can better approach a system with a non-negative coefficient vector, and the algorithm not only has high robustness, but also can obtain a high convergence rate by minimizing a segmentation cost function. The non-negative adaptive filter disclosed by the invention can be applied to electronic and communication systems subjected to large pulse noise interference.

Description

technical field [0001] The invention discloses an adaptive filter, in particular a robust mixed norm non-negative adaptive filter, which belongs to the field of digital filter design. Background technique [0002] In many applications, in the process of modeling and analysis, there may be negative elements in the derivation results. From a mathematical point of view, the existence of these negative elements is understandable, but from the perspective of engineering applications, these negative results It violates the basic physical facts, which is not advisable. For example, non-negativity is an inevitable physical fact in demographics, in the analysis of concentration fields, and in the calculation of image gray levels. Therefore, signal processing in situations such as the above requires the system to meet non-negativity constraints. In order to better simulate the non-negative characteristics of the system, Jie Chen et al. proposed a non-negative least mean square algor...

AI Technical Summary

Problems solved by technology

However, when this algorithm is used for non-negative sparse system identification, the effect is not ideal, and it is not even applicable.

Images