A Hierarchical Scheduler Design for Multiple Traffics with Reduced Feedback

Abstract

In this paper, a novel scheduler is designed for multiple traffic classes with reduced feedback. The novel scheduler consists of two stages, i.e., at the first scheduling stage, the inter-traffic scheduler dynamically allocates the clustered bandwidth resources to each traffic class by a periodic and triggered way; at the second scheduling stage, and the intra-traffic scheduler schedules different users’ packets within each traffic class simultaneously. In order to reduce the feedback, each user only reports the CQI of n subchannels that have the best normalized SNR. The proposed algorithm not only enhances the system throughput but also reduces the feedback load under the condition that the QoS requirements of RT traffic are satisfied. Simulations validate the effectiveness and good performance of the proposed scheme.