Delay-Based Weighted Proportional Fair Algorithm for LTE Downlink Packet Scheduling

This paper proposes an algorithm of delay-based weighted proportional fair (DBWPF) for downlink packet scheduling in the long term evolution cellular networks. Besides the tradeoff between throughput and throughput fairness, which is the target of the traditional PF scheduling, DBWPF also considers the weighted average delay of each user, where the user with larger delay is more likely to obtain resources. With this treatment, the DBWPF scheduling is conducive to achieve delay fairness and implementation rate fairness between different users, which is significant in the case that users in a cell have different traffic volumes. Simulation results indicate that the proposed method has an approximate throughput and throughput fairness compared with the PF scheduling. Moreover, in the aspects of delay fairness and implementation rate fairness, DBWPF is competitively superior to the traditional algorithms.