Combined Regularization Factor for Affine Projection Algorithm Using Variable Mixing Factor

Abstract

The affine projection algorithm with a fixed regularization parameter is subject to a compromise concerning the convergence speed and steady-state misalignment. To address this problem, we propose to employ a variable mixing factor to adaptively combine two different regularization factors in an attempt to put together the best properties of them. The selection of the mixing factor is derived by minimizing the energy of the noise-free a posteriori error, and for the sake of suppressing large fluctuations, a movingaverage method is designed for updating the mixing factor. Based on a random walk model, we also prove that the proposed mixing factor is as well available for the non-stationary system. The mathematical analysis including the stability performance, steady-state mean square error, and computational complexity are performed. In practice, we compare with the existing related algorithms in system identification and echo cancellation scenarios, the results illustrate that the proposed algorithm outperforms them with notable margins.

Dynamic Limited Variable Step-Size Algorithm Based on the MSD Variation Cost Function

“…where 𝛽 is the smooth factor and its value is close to 1. Besides, 𝐸[𝑒 𝑎,𝑓 (𝑛)] also appears in the equation (10). It is often calculated by the following method [9]:…”

“…As the system noise information is difficult to obtain in some unknown environments, the proposed algorithm adopts some relevant statistical values of input and error signals to calculate and estimate the system noise. The full calculation process is as follows [10]:…”

“…At the initial stage, as the distance between the actual weight and the optimum weight is relatively large, 𝑒 𝑎,𝑓 (𝑛) will be also large and close to the error signal. According to the equation (10), this situation will result in a relatively large step-size value at the beginning of iterations. When the proposed algorithm approaches to a steady-state, the weight vector closes to the optimum value.…”

“…To better balance the convergence and steady-state performances, many scholars have proposed some variable step-size LMS algorithms [3][4][5][6][7][8][9][10][11]. The functional controlled variable step-size (FCVSS) algorithm was proposed in [4].…”

A Family of Swish Diffusion Strategy Based Adaptive Algorithms for Distributed Active Noise Control

“…Recently, robust techniques pertaining to maintaining the performance of active noise control in the wake of impulsive noise interference have been introduced such as: error reused filtered-x LMS (ErF-xLMS) [32], modified filtered-x robust normalized least mean absolute third (MF-xRNLMAT) [33] and modified filtered-x affine-projectionlike MCC (MFxAPLMCC) [34]. In addition, a combination of regularization factors by virtue of a mixing parameter for affine projection algorithm is discussed in [35]. A robust arctangent framework is presented for system identification in [36] .…”