调节水床效应的双梯度有源噪声控制自适应算法*

为了使自适应反馈有源噪声控制系统能够兼顾水床效应,通过约束次级信号的能量以调节噪声放大,设计了一种双梯度算法。当次级信号满足约束,算法沿着最小化误差信号能量的梯度方向迭代,反之,则沿着最小化次级信号能量的方向迭代。在有源降噪耳机实例中的对比结果表明,该算法能调节噪声放大并保持较大的降噪带宽,且运算量没有显著增加。

Adaptive double-gradient active noise control algorithms for tuning waterbed effect

To make the adaptive feedback active noise control system take into account the waterbed effect, a double gradient algorithm is designed by limiting the secondary signal energy to adjust the noise amplification. If the secondary signal satisfies the constraint, the algorithm iterates along the gradient direction of minimizing the error signal energy, otherwise, it iterates along the direction of minimizing the secondary signal energy. The comparison results in an active noise reduction headset show that the designed algorithm can adjust the noise amplification while maintaining a large noise reduction bandwidth and the computation load does not increase significantly.