LSTM-Aided Hybrid Random Access Scheme for Machine-Type Communication in 5G-Enabled Industry 4.0

5G enables Industry 4.0 through machine-type communication (MTC). Massive MTC (mMTC) and ultra-reliable and low-latency communication (URLLC) are two key 5G MTC scenarios to support diverse quality of service (QoS) requirements for Industry 4.0. In this paper, an LSTM-aided hybrid random access scheme (LSTMH-RA) is proposed to support 5G-enabled Industry 4.0 where mMTC and URLLC devices coexist. In the proposed LSTMH-RA scheme, mMTC devices access the network via a timing advance (TA)-aided four-step procedure to meet massive access requirement, while the access procedure of the URLLC devices is completed in two steps coupled with the mMTC devices' access procedure to reduce latency. Furthermore, we propose an attention-based LSTM prediction model to predict the number of active URLLC devices, thereby determining the parameters of the multi-user detection algorithm to guarantee the latency and reliability access requirements of URLLC devices. We analyze the successful access probability of the LSTMH-RA scheme. Numerical results show that, compared with the benchmark schemes, the proposed LSTMH-RA scheme can significantly improve the successful access probability, and thus satisfy the diverse QoS requirements of URLLC and mMTC devices.