Abstract
Approximate message passing (AMP) has been successfully applied to sparse regression codes. Furthermore, the presence of an outer code has been shown to improve the finite-length performance of sparse regression codes. Recently, a new paradigm was introduced in the form of a dynamic denoiser that integrates the structure of the outer code into the estimation of the state. This algorithmic opportunity can yield further performance improvements. This methodology can be pushed in several directions, e.g., approximations to the state evolution for outer code design, applying this paradigm to multi-user scenarios, extending the concept to multiple measurement vectors. The presentation will discuss some of these notions.