Latency-Rate Downlink Packet Scheduler for LTE Networks

Abstract

The Long-Term Evolution standard is currently the leading technology used in mobile 4G networks. Under LTE user devices may request services available on the Internet such as voice/video streaming and Web pages. The access to such services is managed by a base station, which does resource scheduling through a multiple access network technology, ensuring Quality of Service. Hence, one of the main challenges regards responding to requests for services that require low latency and high bandwidth. Considering that resource scheduling in LTE networks is a difficult problem this work proposes the Latency-Rate Downlink Packet Scheduler (LR-DPS) for the scheduling of resources of downlink traffic, aiming to meet the maximum delay requirements for the input traffic. The proposal is divided into three hierarchical stages. In the first stage, traffic is bounded by a token bucket. In the second, time allocation and source rates are determined in order to meet the restrictions. In the third stage, the data is allocated in resource blocks in a balanced way, ensuring fairness. The results of the simulations considering different input traffic show that the LR-DPS met the requirements when other known schedulers exceeded the maximum delay requested by up to 90%.

Keywords

• wireless networks
• channel allocation
• 4g mobile communication
• heuristic algorithms
• streaming media