Heterogenous CSI based Coexistence of URLLC and eMBB in RIS-assisted mMIMO system

As two main services in 5G communication network, Ultra-Reliable Low-Latency Communications (URLLC) and enhanced Mobile Broadband (eMBB) are inevitably coexist in a same network. The higher spatial degrees of freedom that a large number of antennas bring to massive Multiple-input Multiple-output (mMIMO) enables it to provide reliable service for more devices within the same time-frequency resource, hence, it’s obvious that mMIMO can promote the coexistence of URLLC and eMBB greatly. However, mMIMO brings a greater cost for CSI accurating while proving extra efficiency. Thus we use statistical CSI(S-CSI) based beamforming for URLLC to reduce the delay. It was proposed in previous study that the performence of S-CSI based beamforming is poor when there was a weak line-of-sight(LoS) channel. Therefore, deploying a RIS to creating a LoS transmission dominated channel can enhance the performance of S-CSI based beamforming system. In this paper, we consider a RIS-assisted mMIMO system to enable the coexistence of URLLC and eMBB. In order to reduce the channel estimation time, we design a heterogenous CSI based precoding scheme. Then, by taking into account the data rate constraint of eMBB and the outage probability constraint of URLLC, the optimization problem that minimize transmit power at the base station (BS) is formed. To tackle the rate outage constraint, we transform it into a data rate constraint through inequalities, and then we have designed an iteratively algorithm to solve the optimization problem. The simulation results show the effectiveness of the algorithm and the benefit of mMIMO and RIS.