Interference Measurement And Signal To Interference Plus Noise Power Ratio Prediction In LTE Uplink

The data rate and spectral efficiency of LTE systems can be effectively improved by link adaptation. Especially in the uplink data transmission, link adaptation is regarded as an important method to enhance throughput. There are two key points in adaptive technology research, measurement and prediction. The purpose is to dynamically change transmission parameters with channel and interference environment, and to achieve the best balance between spectrum utilization and block error rate.Aiming at signal to interference plus noise ratio(SINR) which directly associates with the feedback modulation and coding schemes, this paper discusses the interference measurement and SINR prediction problems in time division duplex(TDD) uplink, and gives theoretical explanations and computer simulations.Firstly, this paper introduces TDD frame structure configuration and sounding reference signal(SRS) in LTE protocols, providing a reference standard for the following studies. SRS sounding mode can be divided into periodic sounding and aperiodic sounding. Periodic sounding is chosen for system building. Huawei soft frequency reuse scheme served as a way to reduce co-channel interference between adjacent cell edge users in the cell modeling. For SRS transmission, a flexible combination of TDM, FDM and CDM forming an interference measurement program under a specific frame structure. Interference information can be obtained in a short period of time and updated in time. Then select a best algorithm from the commonly used SRS channel estimation algorithms. This algorithm is called cyclic shift time domain channel estimation method, and is less sensitive to multiuser scene than conjugate cancel method. Combine this algorithm with MMSE channel estimation method to form a new algorithm which can guarantee better performance in the whole band. The normalized MSE reaches 0.1, which means we can effectively obtain the desired interference power information and accurate SINR.According to the latest SINR information, BS sends feedback MCS to UE. There will be some delay in the transmission process. During the period of delay, the transmission channel and interference background have changed, which cannot accurately reflect the MCS used in the next transmission, so we need SINR prediction. Prediction methods are mainly divided into linear and nonlinear prediction methods. Nonlinear prediction methods have highly complexity and need to adjust parameters in practical experience. Linear methods are simple and effective, but not favorable as nonlinear methods in data jitter processing. Finally, test various types of prediction methods under the simulation system and observe the throughput enhancement.