Signal To Noise Ratio Estimation And Timing Estimation Based On Sounding Reference Signal In LTE

With the rapid development of data communications and multimedia services, the existing mobile communication systems have become increasingly unable to meet people’s needs. Therefore 3GPP organization proposed the Long Term Evolution(LTE), intended to provide higher transmission rates and greater system bandwidth.As the direction of the new generation of mobile communications, LTE has been widespread concern of global researchers and communications industry. LTE system takes MIMO and OFDM as its key physical technologies, while using a lot of new communication technologies. One of the important technologies is the definition of two important reference signals in LTE uplink: Demodulation Reference Signal( DMRS) and Sounding Reference Signal( SRS). This paper focuses on the SRS.Firstly, the paper makes an overview about the uplink physical layer in LTE system, including the radio frame structure, the time-frequency resource and the subcarriers mapping methods. The paper briefly introduces SC-FDMA: the multiple access way in LTE uplink. Then according to the definition of 3GPP agreement, analyze the SRS sequences’ generation method in detail, the good cyclic autocorrelation characteristic, bandwidth allocation of SRS and time-frequency resource configuration.This paper mainly studies SNR estimation algorithm and timing synchronization algorithm based on the SRS. In terms of SNR estimation: the article first gives the transmission mode of SRS and SNR estimation system model; then briefly describes three common estimation algorithm, and according to their respective shortcomings, propose two improved algorithm: improved DFT algorithm and improved DASS algorithm. By modifying the noise estimate range in the time domain, the improved DFT algorithm can reduce the influence of the useful signal energy leakage in the high SNR region,so this algorithm can obtain a more accurate SNR estimation. Simulation results show that the performance of the new algorithm is improved by about 6dB than Boumard and DFT algorithms in the high SNR region; By redefining the calculation method of subcarrier, the improved DASS algorithm reduces the noise estimation error, making the performance in high SNR be significantly improved, and the complexity of the method is only 1/3 of the original DASS method.In terms of timing estimation: this paper firstly makes a detailed analysis on the impact of timing errors, and based on the two existing timing estimation algorithm, propose an improved backward search algorithm which is applicable to SRS. By redefining the threshold and the new search decision technology we can determine the location of the 1st Path in multipath environments, further get the correct timing position. Simulation results show that the algorithm can effectively eliminate the impact of sidelobes in the decision process, and the probability of correct timing are significantly improved.Finally, this paper study SRS under the multi-user condition, and focus on SNR estimation algorithm under the CDM multiplexing circumstances. The multi-user channel estimation and noise estimation are studied respectively, and a method to eliminate the interference of other users is present. At last, the two improved SNR estimation algorithms mentioned above are generalized to a multi-user environment, and the effectiveness of these algorithms is verified by computer simulation.