Subspace Pursuit Algorithm Implementation for Compressive Sensing

This documentation demonstrates the implementation of Wei Dai and Olgica Milenkovic's Subspace Pursuit algorithm, highlighting its exceptional performance in compressive sensing applications while maintaining manageable computational complexity. The algorithm's theoretical foundation and performance characteristics are detailed in their paper "Subspace Pursuit for Compressive Sensing: Closing the Gap Between Performance and Complexity", which we strongly recommend reading for comprehensive understanding. To execute this example, you must install all accompanying .m files in the same directory and run the trialSP function from the MATLAB command prompt. The trialSP function serves as the main entry point that demonstrates the algorithm's capability in reconstructing sparse signals from limited measurements using subspace projection techniques. This implementation was originally developed by Igor Carron, a contributor to http://nuit-blanche.blogspot.com. Users are expected to comply with Creative Commons licensing requirements and provide proper attribution when utilizing this code. For technical assistance or clarification regarding the implementation details, including the core subspace projection mechanism and iteration control parameters, please contact the maintainers.