CS: Sparse Reconstruction of Complex Signals in Compressed Sensing Terahertz Imaging, Compressive Sensing for Sparsely Excited Speech Signals

Laurent Jacques just released a presentation he made recently entitled: A Compressed Introduction to Compressed Sensing: Combining Sparsity and Sampling. It's a good compressed introduction to the subject. Thanks Laurent for the mention at the very end. Today, we have two new papers: one concerns the use of the compressive sensing single pixel camera concept to get away from the raster mode, while the second paper applies CS to speech. Enjoy!

Sparse Reconstruction of Complex Signals in Compressed Sensing Terahertz Imaging by Zhimin Xu, Wai Lam Chan, Daniel M. Mittleman and Edmund Y. Lam. The Abstract reads:

In reconstructing complex signals, many existing methods apply regularization on the magnitude only. We show that by adding control on the phase, the quality of the reconstruction can be improved. This is demonstrated in a compressed sensing terahertz imaging system.

Compressive Sensing for Sparsely Excited Speech Signals by Thippur V. Sreenivas and W. Bastiaan Kleijn. The abstract reads:

Compressive sensing (CS) has been proposed for signals with sparsity in a linear transform domain. We explore a signal dependent unknown linear transform, namely the impulse response matrix operating on a sparse excitation, as in the linear model of speech production, for recovering compressive sensed speech. Since the linear transform is signal dependent and unknown, unlike the standard CS formulation, a codebook of transfer functions is proposed in a matching pursuit (MP) framework for CS recovery. It is found that MP is efficient and effective to recover CS encoded speech as well as jointly estimate the linear model. Moderate number of CS measurements and low order sparsity estimate will result in MP converge to the same linear transform as direct VQ of the LP vector derived from the original signal. There is also high positive correlation between signal domain approximation and CS measurement domain approximation for a large variety of speech spectra.