This is the MATLAB code supporting the paper "SPECTRAL COMPRESSIVE SENSING WITH POLAR INTERPOLATION", submitted for publication at ICASSP 2013.

The two main files are comp_ite10_n100_k4_c5.m and comp_noisy_ite10_n100_k4_c5.m, which runs the two numerical experiments from the paper.
The output from those two functions is saved in the files comp_ite10_n100_k4_c5_RESULTS.mat and comp_noisy_ite10_n100_k4_c5_RESULTS.mat.

To generate Figure 1 and 2 in the paper, use the scripts plot_results.m and plot_noisy_results.m, respectively.
The two *.mat files containing the results shown in the paper is also included in this folder.

Some of the implemented functions and algorithms are not our own, namely:
- The SIHT algorithm is from the SCS toolbox, http://dsp.rice.edu/scs
- The Hungarian algorithm is by Alexander Melin, http://www.mathworks.com/matlabcentral/fileexchange/11609
- The Root MUSIC algorithm is from the "Statistical and Adaptive Signal Processing" Toolbox, http://www.ee.uwa.edu.au/~roberto/teach/as413/course/sasptoolbox/

The SDP algorithm was kindly provided by the Gongguo Tang and since it has not been released for public use, we have also excluded it from our released code.

If you have any questions or suggestions for improvements, feel free to contact us.

Karsten Fyhn and Hamid Dadkhahi
kfn@es.aau.dk / hdadkhahi@ecs.umass.edu
