Design and Analysis of Asynchronous Massive Multiple Access Schemes

Next generation massive multiple access protocols have the aim of achieving high scalability (number of simultaneously active users the system may support) in presence of reliability and latency constraints. This has encouraged research towards grant-free random access protocols, as opposed to grant-based ones. The combined use of time diversity, according to which more replicas of the same packet are sent in different randomly chosen slots, and techniques of collision resolution based on successive interference cancellation has led to a substantial improvement in performance with respect to standard ALOHA schemes and to a radical expansion of the designed space for random access protocols. Some analogies have been discovered with iterative decoding of codes based on sparse graphs, thus opening the possibility to use coding theory and tools for designing more efficient schemes. Nevertheless, most of the current ones require a global slot synchronization between transmitters, which is one of the major drawbacks for large networks as those that will be present in the 6G, since the signalling overhead scales up with the number of users independently from the fact that only a small subset is active at a certain time instant. This thesis studies a novel approach dubbed asynchronous contention resolution diversity slotted ALOHA (ACRDA) that performs a truly asynchronous random access, in which slot boundaries are not defined globally within the wireless network, but they are local to each transmitter and completely asynchronous among transmitters. This approach considerably reduces control signalling for connection establishment, which makes ACRDA one of the candidate schemes for 6G.