Design of discrete coefficient FIR filters using fast simulated evolutionary optimization

In this paper we present a new algorithm, fast simulated evolutionary optimization (FSEO), a multi-agent stochastic search method for optimizing the coefficients of digital filters in the signed power-of-two (SPT) space by minimizing the weighted squared error between discrete filter frequency response and desired frequency response. Unlike conventional methods where each coefficient is allocated a fixed number of SPT terms, our method allows the number of SPT terms for each coefficient to vary subject to the number of SPT terms for the entire filter. Examples on low pass FIR filter design show that FSEO achieves up to 10.76 dB improvement over simulated annealing (SA) and SEO