Throughput optimization for IRS-assisted multi-user NOMA URLLC systems

In this paper, we consider an intelligent reflecting surface (IRS)-assisted multi-user non-orthogonal multiple access (NOMA) ultra-reliable low-latency communication network, where an access point (AP) transmits data to multiple users with the short-packet communication (SPC) by applying the NOMA protocol. To maximize the sum throughput of all users, the transmit power of the AP and the IRS reflecting beamforming are jointly optimized. Since the data rate expression of the SPC is in a complicated form and the optimization variables are coupled with each other, the considered optimization problem is highly non-convex and challenging to solve. Based on the techniques of block coordinate ascent, semidefinite relaxation, Gaussian randomization, and successive convex approximation, we propose an efficient algorithm to obtain a high-quality solution. Simulation results show that the proposed algorithm can achieve significant throughput performance gain over some existing benchmark schemes.