NOMA Assisted Semi-Grant-Free Scheme for Scheduling Multiple Grant-Free Users

Nonorthogonal multiple access (NOMA) assisted semi-grant-free (SGF) transmission is a powerful solution to significantly reduce the end-to-end latency of data transmission and improve spectrum utilization efficiency facing the future Internet of Everything (IoE). Moreover, the recently proposed hybrid successive interference cancellation (HSIC) scheme can avoid the user's outage probability error floor, which is unavoidable in traditional uplink NOMA. However, the existing NOMA assisted SGF transmission schemes using this HSIC scheme admit only one grant-free (GF) user into the grant-based (GB) user's channel at a time. In order to further improve the throughput and reduce the waiting delay of GF users, in this paper, we propose a NOMA assisted SGF scheme with a new HSIC scheme allowing a certain number of GF users to share GB user's channel. First, we propose a new HSIC scheme to decode the signals of multiple GF users and one GB user. Then based on the signal's decoding order at the BS and stochastic geometry, we provide a general mathematical framework to analyze the association success probability of GF users. Besides, in order to intuitively show the throughput improvement brought by multiple admitted GF users, we derive the analytical expressions of admitted GF users' association success probability, outage probability, and throughput when two GF users are allowed to share simultaneously GB user's channel. Finally, the simulation results validate the throughput improvement brought by two GF users being served simultaneously and the accuracy of the theoretical analyses.