The Evaluation of CQI Delay Compensation Schemes Based on Jakes' Model and ITU Scenarios

This paper addresses the evaluation of channel quality indicator (CQI) delay compensation schemes considering the impact of signaling interaction delay between evolved node B (eNodeB) and user equipments (UEs). The Jakes' model and International Telecommunication Union (ITU) channel models are introduced, which are the representatives of simple and complex simulation environments, respectively. The channel characteristics and the variation of CQIs of four typical scenarios are analyzed in detail. We then propose two types of CQI delay compensation schemes, namely weighted mean scheme (WMS) and prediction scheme based on normalized least mean square (NLMS-PS). Evaluation results show that both of these schemes do not work well on ITU channel models because system performance of complex simulation environment depends on not only the accuracy of CQI, but also the combined effects of the scheduling and other factors. However, these proposed schemes can boost the system performance effectively based on Jakes' model when the CQI delay is less than 5 ms.

I. Introduction

The Long Term Evolution (LTE) is one of the systems beyond the third generation(3G). It has the ability to provide up to 100 Mbps in the downlink and 50 Mbps in the uplink [1]. As one of the key link adaptive technologies, adaptive modulation and coding (AMC) technology can improve the system throughput by adjusting coding rate to fit the instantaneous channel state information. Channel quality indicator (CQI) is derived by user equipment (UE) based on an unrestricted observation interval in time and frequency. The CQI index is restricted between 0 to 15 and each of them indicates a combination of modulation and coding schemes [2]. The accuracy of CQI plays an important role on the effectiveness of AMC schemes. However, CQI is not always reliable due to channel estimation error, CQI quantization error and CQI feedback delay.