Unitary Approximate Message Passing for Sparse Bayesian Learning

Sparse Bayesian learning (SBL) can be implemented with low complexity based on the approximate message passing (AMP) algorithm. However, it does not work well for a generic measurement matrix, which may cause AMP to diverge. Damped has been used SBL alleviate problem at cost of reducing convergence speed. In this work, we propose new algorithm structured variational inference, leveraging unitary transformation (UAMP). Both single vector and multiple problems are investigated. It is shown that, compared stateof- the-art AMP-based algorithms, proposed UAMPSBL more robust efficient, leading remarkably better performance.