MSE Analysis Based on Nearly-Oracle Estimation for SCoSaMP Algorithm

For the reconstruction of signals acquired with Sub-Nyquist sampling system based on redundant Gabor frames, SCoSaMP (Signal space-based CoSaMP) algorithm has excellent performance. However, there’s still no analysis about the MSE analysis under Gaussian noise and it is hard to estimate SCoSaMP algorithm reconstruction performance from a theoretical point of view. This paper presents an MSE analysis method based on nearly-oracle estimation to assess the error generated by Gaussain noise. With the proposed method, the upper bound of the MSE (Mean Square Estimate) is calculated, which shows how to improve the algorithm more quickly.

[1]  Miguel R. D. Rodrigues,et al.  On the Use of Unit-Norm Tight Frames to Improve the Average MSE Performance in Compressive Sensing Applications , 2012, IEEE Signal Processing Letters.

[2]  Deanna Needell,et al.  Signal Space CoSaMP for Sparse Recovery With Redundant Dictionaries , 2012, IEEE Transactions on Information Theory.

[3]  G. Pisier.  Probabilistic methods in the geometry of Banach spaces , 1986 .

[4]  Yonina C. Eldar,et al.  Sub-Nyquist Radar via Doppler Focusing , 2012, IEEE Transactions on Signal Processing.

[5]  Yonina C. Eldar,et al.  Xampling: Analog to digital at sub-Nyquist rates , 2009, IET Circuits Devices Syst..

[6]  Michael Elad,et al.  Can we allow linear dependencies in the dictionary in the sparse synthesis framework? , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.