Cooperative Coevolutionary Genetic Algorithm for Digital IIR Filter Design

Abstract

A novel algorithm for digital infinite-impulse response (IIR) filter design is proposed in this paper. The suggested algorithm is a kind of cooperative coevolutionary genetic algorithm. It considers the magnitude response and the phase response simultaneously and also tries to find the lowest filter order. The structure and the coefficients of the digital IIR filter are coded separately, and they evolve coordinately as two different species, i.e., the control species and the coefficient species. The nondominated sorting genetic algorithm-II is used for the control species to guide the algorithms toward three objectives simultaneously. The simulated annealing is used for the coefficient species to keep the diversity. These two strategies make the cooperative coevolutionary process work effectively. Comparisons with another genetic algorithm-based digital IIR filter design method by numerical experiments show that the suggested algorithm is effective and robust in digital IIR filter design.

Real Coded Genetic Algorithm for Design of IIR Digital Filter with Conflicting Objectives

“…The order of digital IIR filter is represented by control gene (Figure 1). The coding method followed has been inherited from [7][8][9]. The control genes determine activation/deactivation of corresponding blocks of filter coefficients by setting 1/0, respectively.…”

“…Conventional gradient-based design may easily stick in the local minima due to non-linear and multimodal nature of error surface of IIR filters. Therefore, researchers have developed design methods [7][8][9][10][11][12][13][14][15][16][17][18], based on modern heuristics optimization algorithms.…”

“…The IIR filter with linear phase response is important in most of practical applications because non-linear phase response alter the frequency components of the signal, which causes distortion. The cooperative co-evolutionary genetic algorithm (CCGA) [8] considers the magnitude response error, phase response error, and lowest order simultaneously and uses NSGA-II [10] to maintain the diversity in the three objectives. A new local search operator enhanced multi-objective evolutionary algorithm (LS-MOEA) [9] simultaneously optimizes the structure and coefficients of digital IIR filter to obtain relatively better linear phase response and lower order, besides the good magnitude response.…”

“…The real-coded genetic algorithm for designing the optimal digital IIR filters is described in Section 3. In Section 4, the performance of the proposed method has been evaluated and achieved results are compared with the design obtained in [7][8][9] for the low-pass (LP), high-pass (HP), band-pass (BP), and band-stop (BS) filters. Finally, the conclusions and discussions are outlined in Section 5.…”

Improved Differential Evolution Based on Stochastic Ranking for Robust Layout Synthesis of MEMS Components

“…ROBUST OPTIMIZATION USING IDE Many types of evolutionary approaches have been developed and implemented as design optimization tools [19], [20]. Genetic algorithms with a robust solution searching scheme were first presented by Tsutsui and Ghosh [21] and later discussed by Deb and Gupta [22], [23].…”