Allocating Three-dimensional Resources for URLLC Services with Different Latency Requirements

In this paper, the problem of three-dimensional joint resource allocation in ultra reliable low latency communication (URLLC) scenario for fifth generation (5G) cellular systems is studied. To extend the existing time-frequency two-dimensional resource allocation, the spatial domain of the multiantenna system is introduced to realize joint three-dimensional resource allocation. In the considering model, a base station needs to allocate radio resources to URLLC services and non-URLLC services, where URLLC services have unique latency and throughput requirements based on the service characteristic. Our goal is to maximize the resource utilization, i.e., the total throughput of non-URLLC services, while meeting URLLC service requirements. To effectively allocate resources among time, frequency, and spatial domains, we first discretize the available resources into resource elements (RE). Then, adjacent REs can be aggregated into resource block (RB), which is the smallest resource allocation unit. We formulate the resource allocation problem as an integer linear programming problem. To solve this problem, a fast resource allocation (FRA) algorithm is proposed. Simulation results show that the proposed algorithm can achieve up to 14% gain, in terms of the throughput of non-URLLC services, compared to the existing two-dimensional resource allocation. FRA can approach the optimal solution based on Gurobi optimizer with a reasonable performance gap while using only 0.21% runtime of Gurobi.