Compressed Sensing Investigation in an End-to-End Rayleigh Communication System: Speech Compression

In this paper, the compressed sensing (CS) technique is investigated in an end-to-end mobile communication system. The purpose is to reduce the transmission chain by performing simultaneously the acquisition and the compression and to reduce the amount of data for transmission through a Rayleigh channel. Orthogonal frequency division multiplexing, receive antenna diversity, and maximum ratio combining technique are used to mitigate the multipath fading effects due to the channel. In addition, l(1) minimization method is used to perform the recovery. Perceptual evaluation of speech quality and root mean square error measures are used to evaluate the performances of the proposed system. CS allows improving the communication system by saving resources while reducing processing and transmission rate. It also ensures reliability and robustness against noise, besides securing the communications. These results can be investigated in many applications where the rapidity is the most important constraint such as military mobile communication.