Abstract
The fifth-generation wireless communication networks (5G) facilitate a wide range of newly-emerging applications alongside existing cellular mobile broadband services. One of the key service classes of 5G is Ultra-Reliable and Low-Latency Communications (URLLC), which guarantees the rapid delivery of short packets (up to 1 ms) with a success probability rate of 99.999%. The challenging reliability and latency requirements of URLLC cannot be delivered by existing cellular networks, resulting in the need for significant air interface modifications. This study aims to satisfy the link latency requirements of URLLC applications, and specifically reduce the latency associated with the presence of the Hybrid Automatic Repeat reQuest (HARQ) feedback scheme. To this end, we investigate a supervised learning method to provide early HARQ (E-HARQ) feedback on the decodability status of the coded-received signal, ahead of the decoding processing. This strategy allows the transmitter to react faster and minimize the signal round-trip time (RTT). The simulation results demonstrate the capability of the proposed mechanism to speed up the feedback releasing and enhance the prediction accuracy by 12% with the introduction of a new feature derived by the channel state estimation.