摘要
自适应回声消除算法是回声消除领域中的主流技术,但由于回声信道的脉冲响应具有阶数高、稀疏以及时变的特点,使得传统的自适应回声消除算法的性能出现了明显的下降。为了改善传统算法的性能,提出了记忆成比例仿射投影算法。但对于高稀疏系统,记忆成比例仿射投影算法仍然表现出收敛速度慢和跟踪性能差的问题。针对这些问题,文中提出了一种改进的记忆成比例仿射投影算法,大大改善了收敛速度和跟踪性能。仿真结果验证了提出算法的有效性。
Adaptive echo cancellation algorithm is mainstream technology of echo cancellation field,but the performance of the traditional adaptive echo cancellation algorithms have shown a significant decline because the impulse response of the echo channel has a high order, sparse and time-varying characteristics. Recently,to improve the performance of traditional algorithms,a memory proportional affine projection algorithm( MPAPA) was proposed. However,the MPAPA exhibits slow convergence rate and poor tracking performance for high sparse system. To address these issues,this paper presents an improved memory proportional affine projection algorithm,which greatly improves the convergence rate and tracking performance. The proposed algorithm is verified in the simulation experiments.
出处
《信息技术》
2016年第8期163-166,共4页
Information Technology
关键词
自适应滤波
记忆成比例仿射投影算法
稀疏系统
回声消除
adaptive filter
memory proportional affine projection algorithm
sparse system
echo cancellation
