Dynamic Resource Management for LTE-Based Hybrid Access Femtocell Systems

Abstract

Hybrid access femtocells for long term evolution (LTE)-based cellular networks provide a tradeoff between closed and open access femtocells whereby all subscribers are granted access albeit with priority given to closed access subscribers. Due to the need to accommodate both closed and open access subscribers, quality of service (QoS) provisioning for LTE-based hybrid access femtocells has become more challenging. This paper addresses this issue and proposes a new dynamic resource management scheme for such hybrid architectures. In particular, the proposed scheme first classifies and performs lexicographic admission control on the incoming traffic data flows using an optimal greedy algorithm. A suboptimal delay-bounded packet scheduling algorithm and a dual decomposition-based power allocation algorithm are developed to solve the non-convex maximization problem such that the weighted sum rate of each femtocell is maximized, subject to bounded packet delays and power constraints. Simulation results show that the proposed scheme can significantly outperform existing schemes in terms of QoS, throughput and fairness.