Massive Unsourced Random Access Based on Uncoupled Compressive Sensing: Another Blessing of Massive MIMO

We put forward a new algorithmic solution to the massive unsourced random access (URA) problem, by leveraging rich spatial dimensionality offered large-scale antenna arrays. This paper makes an observation that signature is key URA in connectivity setups. The proposed scheme relies on slotted transmission framework but eliminates need for concatenated coding was introduced context of coupled compressive sensing (CCS) paradigm. Indeed, all existing works CCS-based rely inner/outer tree-based encoder/decoder stitch slot-wise recovered sequences. takes different path harnessing nature-provided correlations between reconstructed channels each user order together its decoded required channel estimates and sequences are first obtained through hybrid generalized approximate message passing (HyGAMP) algorithm which systematically accommodates multiantenna-induced group sparsity. Then, correlation-aware clustering based expectation-maximization (EM) concept used with Hungarian find optimal assignment matrices enforcing two constraints very specific problem at hand. Stitching then accomplished associating their respective users according ensuing matrices. Exhaustive computer simulations reveal can bring performance improvements, high spectral efficiencies, as compared state-of-the-art technique investigates use arrays URA.