摘要
针对传统Fx LMS算法前馈自适应主动降噪耳机系统因果性条件不足时在宽带噪声环境中产生的高频噪声抬升问题,该文引入权重滤波误差信号Fx LMS算法用于抑制高频噪声的抬升,但该算法带来了低频降噪量不足问题。因此,进一步提出将固定系数混合控制器与权重滤波误差信号Fx LMS算法结合,在解决高频噪声抬升问题的同时,保证了良好的低频降噪量。基于DSP平台实现了提出的主动降噪耳机方案。实验证明,该方案针对宽带和单频等噪声都取得了较好的降噪效果。
It is commonly observed that conventional feedforward FxLMS algorithm in headphone ANCsystems will boost the high-frequency noise in the wide-band noise environment when the causality condition isnot satisfied. To deal with this problem, we present a weighted-error-FxLMS algorithm to attenuate the highfrequencynoise. However, the new algorithm achieves a degraded low-frequency noise reduction. To achieve abetter noise reduction performance, we combine the weighted-error-FxLMS feedforward adaptive filter with ahybrid fixed-coefficient controller. The proposed ANC headphone system is implemented in a DSP platform.Experiments show that the system achieves improved noise reduction performance for both the broad-bandnoise and single-frequency noise environment.
作者
李楠
安峰岩
杨飞然
杨军
LI Nan;AN Fengyan;YANG Feiran;YANG Jun(University of Chinese Academy of Sciences, Beijing 100049, China;Key Laboratory of Noise and Vibration Research, Institute of Acoustic, Chinese Academy of Sciences,Beijing 100190, China)
出处
《应用声学》
CSCD
北大核心
2018年第3期391-399,共9页
Journal of Applied Acoustics
基金
国家自然科学基金项目(11404367
61501449
11474306)
中国青年科学基金项目(11504405)
中国科学院声学研究所青年英才计划项目(QNYC201722)
2016年湖北省省院合作专项
