High system capacity pre-scheduler for multi-cell wireless networks

Abstract

With the constant increase of throughput demands, maximizing the wireless network capacity has always been a crucial issue. Network densification becomes one of the adopted strategies to guarantee a high Quality of Experience (QoE). These new access points are included in the traditional 5G coverage scheme and offload the network in dense areas where the demand is high. A mobile may be located in the coverage area of several base stations which implies the need to extend traditional scheduling to a multicell context. This leads to the emergence of new issues concerning resource allocation strategies. When one user is suitable to receive resources from different cells, a well designed users' distribution is essential to limit unbalanced and overloaded cells. The main contribution of this paper is to propose a pre-scheduler called Multi-Cell Pre-Scheduler (MCPS) that adequately allocates user traffic between different access points, intervening before the scheduling application. Compatible with any type of simple cell existing schedulers, MCPS adapts schedulers to multi-cell context and improves their performance by delaying the system congestion and making the system more robust to unexpected traffic peaks. Performance evaluations show that the use of MCPS solution allows to increase the throughput capacity and energy efficiency while improving QoE fairness between users.