降采样稀疏FIR次级通道有源降噪研究

针对有源噪声控制中线性长时延次级通道系数多导致计算量大的问题，提出了基于降采样稀疏FIR次级通道模型的有源降噪方法。该方法采用降采样技术解决了一定硬件能力下长时延次级通道辨识时的系数不收敛问题；采用稀疏FIR次级通道系数表征完整次级通道系数，有效减少了单次迭代的算法计算量，同时给出了基于和表征函数的稀疏阈值选取方法。仿真和管道有源降噪系统实验验证了文中方法可有效减少计算量，提高收敛速度，该方法可解决一定硬件资源下长时延次级通道系统降噪问题，具有较好的实用性。

Down-sampled sparse FIR secondary path for active noise control

Aiming at reducing the heavy computational load caused by longtime delay secondary path in active noise control (ANC) system, a down-sampled sparse finite impulse response (FIR) secondary path processing method is proposed. The down-sampled process is used to solve the non-convergence problem for the longtime delay secondary path. The sparse FIR secondary coefficients selected by the threshold value, which is determined by the sum cost function, can reduce the computational load in comparison with the complete secondary coefficients. The simulations and actual tests of the duct ANC system are conducted to verify the effectiveness of the method. The proposed method has favorable practicability to solve the noise reduction problem for the long time delay secondary path ANC system under certain hardware resources.