Coefficient estimation of IIR filter by a multiple crossover genetic algorithm

✍ Scribed by Wei-Der Chang

• Publisher: Elsevier Science
• Year: 2006
• Tongue: English
• Weight: 422 KB
• Volume: 51
• Category: Article
• ISSN: 0898-1221
• DOI: 10.1016/j.camwa.2006.01.003

✦ Synopsis

paper proposes an improved genetic algorithm (GA) with multiple crossovers to estimate the system coeffÉcients for the infinite impulse response (IIR) digital filter. In the traditional crossover operation, it needs two parent chromosomes to achieve the crossover work, whereas in this paper the proposed algorithm selects three chromosomes for crossover in order to generate more promising offspring toward the problem solution. Each of unknown IIR coefficients is called a gene and the collection of genes forms a chromosome. A population of chromosomes is evolved by the genetic operations of reproduction, multiple crossover, and mutation. Finally, two illustrative examples including 5he band pass and band stop IIR filters are demonstrated to verify the proposed method.