基于震荡权重粒子群杂交的过2采样DFT滤波器组优化设计

Optimum design of 2-oversampling DFT filter bank based on shocking weight PSO hybridization

基于离散傅里叶变换(discrete Fourier transform,DFT)滤波器组的数字射频存储器具备宽频带覆盖和干扰宽带雷达的特点。针对临界采样时DFT滤波器组信道间混叠严重的问题,推导了过2采样DFT滤波器组的高效多相实现结构,将带间混叠减少了一半。为进一步提高重构信号的精度,提出了基于震荡粒子群杂交算法的滤波器频域设计方法,通过优化目标函数,提高了输出峰值信噪比,降低了最大输出误差,缩减了优化求解时间。仿真结果验证了过2采样DFT滤波器组结构的正确性,提出的震荡权重粒子群杂交滤波器设计可以精确地重构跨多个信道的宽频信号,具备干扰宽频信号的能力。

Digital radio frequency memory based on discrete Fourier transform(DFT)filter banks has features of wide frequency coverage and interference with broadband radars.The efficient polyphase implementation structure of the 2-oversampling DFT filter bank is deduced,which reduces the half of alias between the bands.In order to further improve the accuracy of the reconstructed signal,a filter frequency domain design method based on the particle swarm hybridization algorithm is proposed.By optimizing the objective function,the output peak signal to noise ratio is improved,the maximum output error is reduced,and the optimal solution time is reduced.The simulation results verify the correctness of the structure of the 2-oversampling DFT filter bank.The proposed filter design method can accurately reconstruct the wideband signal across multiple channels,with the ability to interfere with the wideband signal.