自适应剪枝的多参考有源控制算法

Multiple references active noise control with adaptive pruning

多参考信号有源控制算法是多噪声源环境噪声控制的研究热点之一。该文设计了Sigmoid函数加权的滤波最小均方算法,该算法不需要事先计算参考信号与误差信号的相关性以选取参考信号,而是在参考信号中引入自调整参数,通过调整该参数减小无关信号对系统的影响。仿真实验对比验证了算法的两个优点:一是算法可以看作对每个参考信号采用了变步长的策略,二是算法可以依据参考信号与误差信号的相关性高低调整参考信号的权重,实现自适应剪枝的功能以降低系统的开销。

The active control algorithm for multiple reference signals is one of the research hotspots in the control of environmental noise with multiple noise sources.This paper designs a Sigmoid weighted filtered-x least mean square algorithm,which does not need to calculate the correlation between the reference signal and the error signal in advance for the selection of the reference signals.Self adjusting parameters are introduced into the reference signal to reduce the impact of irrelevant signals on the system.The comparison of simulation experiments has verified two advantages of the algorithm:firstly,the algorithm can be seen as adopting a variable step size strategy for each reference signal,and secondly,the algorithm can adjust the weights of the reference signals based on the correlation between the reference signal and the error signal,thus achieves adaptive pruning of the reference signals to reduce system overhead.