A resource allocation algorithm for users with multiple applications in 4G-LTE

In this paper, we consider resource allocation optimization problem in fourth generation long term evolution (4G-LTE) with users running multiple applications. Each mobile user can run both delay-tolerant and real-time applications. In every user equipment (UE), each application has a application-status differentiation from other applications depending on its instantaneous usage percentage. In addition, the network operators provide subscriber differentiation by assigning each UE a subscription weight relative to its subscription. The objective is to optimally allocate the resources with a utility proportional fairness policy. We propose an algorithm to allocate the resources in two-stages. In the first-stage, the UEs collaborate with the evolved node B (eNodeB) that allocates the optimal rates to users according to that policy. In the second-stage, each user allocates its assigned rate internally to its applications according to their usage percentage. We prove that the two-stage resource allocation algorithm allocates the optimal rates without eNodeB knowledge of the UEs utilities. Finally, numerical results on the performance of the proposed algorithm are presented.