Packet Scheduling in LTE with Imperfect CQI

In this paper, a Kalman filter based packet scheduling scheme with low complexity is proposed for LTE downlink system to adapt to various imperfect channel states. A channel quality predictor based on Kalman filter is developed to estimate the current channel quality information (CQI) from previous CQI value and observed CQI value, and a grouping method is employed to cover the gaps caused by Kalman filter. Based on the simulation results, the estimated CQI values from the channel quality predictor are rather close to the realistic CQI values. A packet scheduling scheme that uses the channel quality predictor together with a grouping method in packet scheduling decisions is introduced. Simulation results show the proposed scheme offers a simple yet more effective way to overcome multiple imperfect CQIs .and the performance benefits of the proposed scheme in maximizing system throughput and decreasing packet loss ratio. Keywords— LTE, packet scheduling, imperfect CQI, prediction, Kalman filter