Bayesian Sparse Blind Deconvolution Using MCMC Methods Based on Normal-Inverse-Gamma Prior

Bayesian estimation methods for sparse blind deconvolution problems conventionally employ Bernoulli-Gaussian (BG) prior modeling sequences and utilize Markov Chain Monte Carlo (MCMC) the of unknowns. However, discrete nature BG model creates computational bottlenecks, preventing efficient exploration probability space even with recently proposed enhanced sampler schemes. To address this issue, we propose an alternative MCMC method by using Normal-Inverse-Gamma (NIG) prior. We derive effective Gibbs samplers illustrate that burden associated can be eliminated transferring problem into a completely continuous-valued framework. In addition to sparsity, also incorporate time frequency domain constraints on convolving sequences. demonstrate effectiveness via extensive simulations characterize gains relative existing modeling.