A linear programming approach to joint scheduling of real-time and non real-time services in OFDMA-based systems

As we move forward toward the fifth generation (5G) mobile communication systems, an increasing interest is devoted to heterogeneous broadband access techniques and application requirements. In this paper, we propose a linear programming formulation for the inter-class scheduling problem in OFDMA systems. Our objective is to efficiently and fairly allocate the limited system resources to both real-time and non real-time applications. Taking advantage of the medium size of this problem and the rapid growth of computational power, we find that it will be possible to use linear programming algorithms to solve it. However, because of the capacity constraint, an infeasibility may occur, which is unacceptable for this online problem. To overcome this issue, we extend the system with dummy subchannels that will accept the unsatisfied demands. Computational results show that it is possible to solve this joint scheduling problem within reasonable time for moderately and heavy loaded system.