A QoS-aware downlink packet scheduler using token bucket algorithm for LTE systems

Abstract

LTE is one of the leading technologies to meet the requirements of 4G telecommunication systems. Important design parameters intended to achieve this goal include high data rate, low latency and efficient network bandwidth requirements. Scheduling in LTE is one of the most important functions of the eNodeB due to the existence of different subcarriers in OFDMA modulation and the frequency diversity of the channel. However, the scheduling algorithms are not defined by the Third Generation Partnership Project (3GPP) and are left open for designers and manufacturers. In this paper, we propose a novel QoS-Aware scheduling algorithm in which a Token Bucket algorithm is used in order to limit the access to the resources used by the real time traffic. Simulation results show that the Token Bucket algorithm, by limiting the amount of resources allocated to real time and high priority traffic streams, causes other type of traffic not to be starved. This feature of the scheduling algorithm not only increases the efficiency of the frequency utilization, but also causes the QoS requirements of various traffic classes to be satisfied.