A Delayless Frequency-Domain Kalman Filter with Improved Tracking Capability for Acoustic Feedback Cancellation [一种可快速再跟踪的无延时频域卡尔曼滤波啸叫抑制算法]

作者：Lu C.-X.; Yang F.-R.; Yang J.^*

来源：Acta Electronica Sinica, 2018, 46(8): 1954-1959.

DOI：10.3969/j.issn.0372-2112.2018.08.021

• Acoustic feedback cancellation
• Frequency-domain adaptive algorithm
• Kalman filtering
• Tracking ability

The partitioned block frequency-domain Kalman filter (PBFDKF) is appealing due to its fast convergence speed, low steady-state misalignment and small inherent delay. However, the tracking ability of the PBFDKF is poor. In this paper, we firstly analyze the tracking performance of the PBFDKF and then we propose an improved PBFDKF-based adaptive feedback cancellation algorithm. Moreover, we present a new delayless implementation with lower complexities. Simulation results demonstrate the performance advantage of the proposed algorithm.

• 单位
  中国科学院大学