基于参数提取约束的快速滤波器设计方法

Fast Filter Design Based on Constrained Parameter Extraction

隐式空间映射算法应用于微波滤波器设计能够显著地提高设计效率,但也会因提取参数的不理想造成弱收敛的现象。本文通过对参数提取过程施加限制条件,从而对迭代过程加以约束,较好地避免了弱收敛的问题,改善了算法效率。以差分带通滤波器为例,中心频率为7.85GHz,经过三次仿真迭代获得一组较好的参数指标,进行微调后得到理想的仿真结果,加工实物并进行测量,实测结果与仿真结果较吻合,由此验证了算法的可行性和准确性。

The design efficiency of microwave filters can be significantly improved by utilizing implicit space mapping(ISM)algorithm.However,the weak convergence phenomenon may be caused by the unsatisfactory parameter extraction.Optimized algorithm was proposed to avoid the weak convergence problem through appropriate restrictions of the auxiliary parameters.The algorithm could improve the speed of convergence and the calculation efficiency.As an example,a differential band-pass filter with center frequency of 7.85 GHz was simulated by using the improved algorithm.A set of objective parameters were obtained by performing iterative calculation three times,and then some fine-tuning would further lead to ideal simulation results.The filter was fabricated according to the simulation.Measurement results of the filter are in good agreement with the simulated results,which verifies the feasibility and accuracy of the algorithm.