L1 penalized least squares with SALSA

This folder contains MATLAB functions to implement:

• basis pursuit (BP),
• basis pursuit denoising (BPD),
• dual basis pursuit (dual BP),
• dual basis pursuit denoising (dual BPD).

MATLAB functions

• BP.m
• BPD.m
• dualBP.m
• dualBPD.m

Examples

Illustrative examples with real and complex data.

• BP with DFT (complex data)
• BPD with DFT
• BPD with STFT
• dual BP with DFT and STFT
• dual BPD with DFT and STFT
• BPD with DFT (complex data)
• BPD with STFT (complex data)
• dual BPD with DFT and STFT (complex data)

The implementation here assumes that A1 and A2 are Parseval frames; i.e., A A^H = I

Web

http://eeweb.poly.edu/iselesni/lecture_notes/SALSA/

Contact

Ivan Selesnick
Email: selesi@poly.edu
Electrical and Computer Engineering Dept.
NYU Polytechnic School of Engineering
Brooklyn, New York, USA