The applications of evolutionary computation to the design of digital filter

• Main Authors: Chien-Tao Wang, 王建道
• Other Authors: Hung-Chih Wu
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/34405488218469634016

碩士 === 樹德科技大學 === 資訊工程學系 === 94 === In this thesis, two evolutionary computation algorithm, genetic algorithm (GA) and differential evolution (DE), are applied to the design of IIE filter. First, we use GA to design a Canonical Signed-Digit (CSD) code based IIR filter. It can be seen that, are to the special configuration of CSD code, GA is a better algorithm to evolve the CSD coded parameters. Second, an criterion for checking the stability of an IIR filter will be derived. The stability criterion is a base of designing stable IIR filter in this thesis. Finally, the two algorithms, GA and DE, are used to design the stable IIR filter according to the stability criterion derived in previous step. We will compare the efficiency of the two algorithms in de signing the stable IIR filter.