QoS-aware bit scheduling in multi-user OFDM systems

Radio resource management (RRM) is widely regarded as a critical component in improving spectral efficiency of beyond third generation (3G) cellular air interfaces. Much of the published work in Orthogonal Frequency Division Multiplexing (OFDM) RRM focuses on exploiting multi-user, multi-channel and multi-application diversity. In this paper, we propose a Quality of Service (QoS)-aware bit scheduling framework to increase the flexibility and granularity of the resource allocation algorithms by adaptively matching the QoS requirements of the application bits of users to the characteristics of the OFDM subcarriers in a mixed-traffic environment. We show that with the finesse control of QoS-aware bit scheduling, it is possible to simultaneously achieve both an increase in user throughput and a reduction in user packet drop probability by accepting a within packet drop threshold increase in user latency. The performance gains obtainable are quantified in terms of user throughput, user latency, user jitter and user packet drop probability.