Skip to content

Using machine learning algorithms for IIR filter optimization

License

Notifications You must be signed in to change notification settings

igorinov/optimal-filter

Repository files navigation

optimal-filter

Using machine learning algorithms for recursive digital filter optimization

This is an example of using automatic differentiation and optimization methods of TensorFlow to design a 6-order gigital band-pass IIR filter with frequency response as close as possible to the analog filter of the same order.

Initial digital filter is computed using bi-linear transform: Initial filter, frequency response

Adam algorithm is used to make the digital filter's frequency response closer to analog filter. Optimized filter, frequency response

About

Using machine learning algorithms for IIR filter optimization

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages