Compressive Sensing for Autoregressive Hidden Markov Model Signal

Cornpressive sensing(CS) is an emerging filed based on the revelation that a small collection of linear projections of a sparse signal contains enough information for stable, sub-Nyquist signal acquisition. One challenging problem in compressive sensing is that it is difficult to represent signal in sparse basis, which makes this algorithm sometimes impractical. In this paper, we can setup a new standard compressive sensing problem for autoregressive hidden markov signal by utilizing the original observation vector and the autoregressive coefficients.