Random Structures Based Design Method For Dual-Passband IIR Digital Filters

The digital filter can process the transmitted signal effectively.It can transmit the signal of effective frequency band and filter out the signal of useless frequency band.Therefore,the digital filter has the characteristics of filtering and noise reduction.With the rapid development of wireless communication technology,digital filter has become a key component of many modern digital systems.In digital filters,dual-band digital filter is a very important module,which attracts wide attention.In order to design a dual-band digital filter with good signal processing ability,this thesis designs a second-order subsystem,which can be used as the basic unit of dual-band digital filter.Then,according to the requirements,a series of instructions are generated,and the second-order subsystems are connected one by one according to the instructions,thus forming a complete and effective dual-band digital filter system.Finally,a new evolutionary algorithm is proposed to design dual-band digital filters.This algorithm is based on the structure evolution of the adaptive multi-elites-guided composite differential evolution algorithm with a shift mechanism(DEAMECo).The proposed method can directly design the dual-band digital filter without using the known circuit structure,and then optimize the dual-band digital filter structure with DEAMECo algorithm,which can effectively search and prevent premature convergence of individuals.Compared with the existing methods,the dual-band digital filter designed in this thesis has a variety of structures and can achieve the design requirements.The research work of this thesiscan be divided into the following five parts:(1)Introduce the implementation method of IIR digital filter.(2)A second-order subsystem structure and a series of instructions are designed,and then the second-order subsystem is added to the filter system in turn according to the instructions to form a dual-band digital filter structure.(3)The structure of dual-band digital filter is coded,and then the structure of the generated dual-band digital filter is optimized by using DEAMECo algorithm.(4)The simulation experiments of dual-band digital filters are carried out,and the experiments of dual-band digital filters with different parameters and evolutionary algorithms are carried out.The results are compared and analyzed,so as to explore the optimal design parameters and optimization algorithm of dual-band digital filters.(5)Summarize the design of dual-band digital filter,and then put forward the future research prospects.Through the analysis and comparison of the experimental results in this thesis,it can be proved that the dual-band digital filter based on random structure and the design method optimized by evolutionary algorithm can basically meet the experimental requirements and have better performance.According to the characteristics of the target,the structure of the dual-band filter is designed directly by using the designed second-order subsystem without prior knowledge of the order and transfer function of the filter.Then the structure of the dual-band digital filter is optimized by using DEAMECo algorithm,which achieves good results.In this thesis,a new design method of dual-band digital filter is proposed,and a new idea for related research is put forward.At the same time,the method is universal and can be applied to other similar optimization problems.