Statistical Learning-Based Dynamic Retransmission Mechanism for Mission Critical Communication: An Edge-Computing Approach

Mission-critical machine type communication (MC-MTC) systems in which machines communicate to perform various tasks such as coordination, sensing, and actuation, require stringent requirements of ultra-reliable and low latency communications (URLLC). Edge computing being an integral part of future wireless networks, provides services that support URLLC applications. In this paper, we use the edge computing approach and present a statistical learning-based dynamic retransmission mechanism. The proposed approach meets the desired latency-reliability criterion in MC-MTC networks employing framed ALOHA. The maximum number of retransmissions Nr under a given latency-reliability constraint is learned statistically by the devices from the history of their previous transmissions and shared with the base station. Simulations are performed in MATLAB to evaluate a framed-ALOHA system’s performance in which an active device can have only one successful transmission in one round composed of (Nr + 1) frames, and the performance is compared with the diversity transmission-based framed-ALOHA.