[en] This paper addresses the radio access network (RAN) resource slicing problem in the context of the joint allocation of transmit powers and time-frequency resource blocks (RBs) in the 5G system consisting of ultra-reliable and low-latency communication (URLLC) and enhanced mobile broadband (eMBB) users. Specifically, we formulate a modulation and coding scheme (MCS) based optimization problem to maximize the sum goodput of eMBB users while satisfying URLLC and eMBB users' QoS requirements. The proposed scheme considers the impact of imperfect channel state information (CSI) and the active user's queue status for the dynamic assignment of radio resources to the heterogeneous users according to its demand. The resulting mixed-integer non-convex problem is first transformed into a tractable form by exploiting the probabilistic to non-probabilistic conversion, Big-M theory, and difference-of-convex (DC) programming. Later, the transformed problem is solved using the successive convex approximation (SCA) based iterative algorithm. Our simulation results illustrate the superiority of the proposed algorithm compared to the baseline methods in terms of eMBB rate, latency in delivering the URLLC packets, and total power consumption.
