北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2022, Vol. 45 ›› Issue (1): 102-107.doi: 10.13190/j.jbupt.2021-118

• 研究报告 • 上一篇    下一篇

一种改进的凸组合仿射投影符号算法

火元莲1, 连培君1, 齐永锋2, 王丹凤1   

  1. 1. 西北师范大学 物理与电子工程学院, 兰州 730070;
    2. 西北师范大学 计算机科学与工程学院, 兰州 730070
  • 收稿日期:2021-06-08 出版日期:2022-02-28 发布日期:2021-12-16
  • 作者简介:火元莲(1973—),女,副教授,硕士生导师,邮箱:hylqqq@nwnu.edu.cn
  • 基金资助:
    国家自然科学基金项目(61561044);西北师范大学研究生培养与课程改革项目(2019KGLX01018)

A New Fast Convex Combination Affine Projection Sign Algorithm for S-type Functions

HUO Yuanlian1, LIAN Peijun1, QI Yongfeng2, WANG Danfeng1   

  1. 1. School of Physics and Electronic Engineering, Northwest Normal University, Lanzhou 730070, China;
    2. School of Computer Science and Engineering, Northwest Normal University, Lanzhou 730070, China
  • Received:2021-06-08 Online:2022-02-28 Published:2021-12-16

摘要: 针对传统的凸组合仿射投影符号算法收敛速度慢和计算量大的问题,提出了一种基于S型函数的快速凸组合仿射投影符号算法。该算法用新的S型函数代替传统算法中的sigmoid函数,改进了联合参数的迭代公式,避免了复杂的指数运算,减少了计算量。同时,采用瞬时转移方案,使凸组合算法在初始自适应过程或在突然发生变化之后能加快小步长慢速算法的自适应过程,进而提升凸组合算法的收敛速度。在系统辨识和声学回声消除环境下的仿真实验表明,该算法具有良好的均方性能和跟踪性能,与传统的凸组合仿射投影算法相比具有更快的收敛速度和更低的计算复杂度。

关键词: 自适应滤波, 仿射投影符号算法, 凸组合, S型函数

Abstract: Aiming at the problems of slow convergence speed and heavy computation of traditional convex combinatorial affine projection sign algorithm, a fast convex combinatorial affine projection sign algorithm based on S-type function is proposed. The algorithm uses a new S-type function to replace the sigmoidfunction in the traditional algorithm. It also, improves the iterative formula of joint parameters, avoids complex exponential operation and reduces the amount of calculation. Meanwhile, the instantaneous transfer scheme is adopted to make the convex combination algorithm accelerate the adaptation process of the small-step slow algorithm after the initial adaptation process or a sudden change, so as to improve the speed of the convex combination algorithm. Simulation results in the system identification and acoustic echo cancellation environment show that the algorithm has good mean square performance and tracking performance, and has faster convergence speed and lower computational complexity compared with the traditional convex combination affine projection sign algorithm.

Key words: adaptive filtering, affine projection sign algorithm, convex combination, S-type function

中图分类号: