Numerical analysis for joint PHY and MAC perspective of Compressive Sensing Multi-User Detection with coded random access

Abstract

Massive Machine Communication (MMC) in the next generation of mobile communication (5G) systems requires new Medium Access Control (MAC) and physical (PHY) layer concepts to handle massive access with low overhead. The recently developed concept coded random access is capable of resolving collisions in massive access at MAC layer. Furthermore, Compressive Sensing Multi-User Detection (CS-MUD) achieves joint activity and data detection from the PHY layer perspective by exploiting sparsity in sporadic multi-user detection. In [1], a joint design of these two concepts was semi-analytically evaluated combining CS-MUD with coded random access under certain assumptions, which showed a promising performance for supporting MMC. In this work, a full link-level analysis of the joint protocol will be presented to verify the previous results and provide deep insights on the link-level performance. Although the joint approach on the link-level shows a small throughput loss compared to the semi-analytical evaluation, high system flexibility can be achieved by tuning MAC and PHY resources dynamically based on the numerical analysis which can be referred to in the joint MAC-/PHY-layer design.