Dynamic multi-user detection based on structured compressive sensing for IoT-oriented 5G systems

Abstract

In Internet of things (IoT)-oriented 5G systems, more efficient multiple access is essential to handle the massive number of sporadic traffic generating IoT users/devices, which are inactive most of the time but regularly or irregularly access or leave the wireless network without human interaction. Non-orthogonal multiple access (NOMA) is a promising solution to support massive connectivity in future IoT-oriented 5G systems, where the dynamic multi-user detection (MUD) is required. In this paper, by exploiting the structured user activity sparsity, we propose the structured matching pursuit (SMP)-based dynamic MUD to jointly realize dynamic multi-user signal detection in several continuous time slots based on structured compressive sensing (SCS). Specifically, in several continuous time slots, the set of active users usually changes slowly due to continuous data transmission. Therefore, we can divide the active user sets into two different parts, i.e., the common active user set and the dynamic active user sets. Accordingly, the proposed SMP-based dynamic MUD simultaneously detects the common active user set in all time slots at first, and then the dynamic active user sets are detected in each single time slot individually. Simulation results show that the proposed SMP-based dynamic MUD can achieve much better performance than that of the conventional compressive sensing (CS)-based MUD, while they share very similar computational complexity.