# Genetic or evolutionary algorithms for filter design?

For what types of filter design problems might genetic or evolutionary algorithms be useful?

What kinds of genetic or evolutionary algorithms are used for DSP problems?

Edit: I expanded the question to include the larger set of evolutionary algorithms, such as differential evolution.

• What types of problems are genetic algorithms used for in general? Seems like they're used for optimizing systems with so many variables that they can't be optimized analytically? – endolith Dec 2 '11 at 21:44

## 1 Answer

Application of a breeder genetic algorithm for system identification in an adaptive finite impulse response filter: Used a GA to determine the parameters of an adaptive FIR.

Optimizing the design of IIR filter via genetic algorithm: Used a GA to determine the parameters for an IIR based on $H_2$ and $H_{\infty}$ criteria.

A genetic algorithm for optimisation of linear phase FIR filter coefficients: (from the abstract) "The algorithm generates a population of genomes that represents the filter coefficients and compares the amplitude response of each genome to that of the desired amplitude response."

Based on this, it seems like it was a popular way to generate the coefficients for filters in the late 90's early 2000's, but there doesn't seem to be a lot of recent activity on it in the IEEE pubs.