The Research Of Channel Detection Technique Based On Sounding Reference Signal In LTE System

With the development of science and technology progress and social informatization, the public has put forward higher requirements for broadband access technology. Although3G mobile communication system can provide more capacity, better quality communications services, and support multimedia features, it still exists some shortcoming that high-speed data transfer performance is not ideal enough. Therefore,3GPP has developed3G Long Term Evolution (LTE) protocol in order to increase large data rate, reduce the system’s transmission delay, increase coverage and system capacity, while reducing operating costs and improving service quality.LTE system defines the sounding reference signal (SRS). Different from demodulation reference signal in the shared link and control link, SRS is mainly used for the base station (eNodeB) to estimates the uplink channel quality, and does not participate in the demodulation of the data processes. eNodeB gets the uplink channel information through the signal-to-noise ratio, which is estimated by SRS, then input the information to scheduler as the main basis for the uplink scheduling. Accurately reflecting the channel information plays a key role to ensure LTE system transmission quality and its superiority, therefore this article focuses around the SRS channel detection technology. This article conducts the study from channel estimation and noise estimation two aspects, improves the original algorithm, structures SRS simulation link and compares simulation, focusing on the completion of the following work:1) Study the SRS subframe configuration and frequency domain calculation on the basis of in-depth understanding of the SRS physical layer structure. In particular, provide a time-frequency domain parameters design method which can effectively avoid collisions between the multi-user.2) Deduce multi-user channel estimation algorithm based on the study of single-user channel estimation algorithm and study the truncated window in detail. Analyze the shortcomings of the current algorithm, that is, the window is likely to reveal some energy in the case of larger timing error or bad channel conditions. When the number of users multiplexed is small, discuss how to adjust the length of the time domain truncation window. And then according to the timing error existing in the actual system, the position of the time domain truncation window is searched to avoid the influence of changing. At the same time, the original algorithm is improved by increasing the noise suppression process in the time-domain window.3) Build a simulation link, the sending and receiving ends of the SRS program is designed, analyze factors which affect the channel estimation algorithm performance to set the test case, simulate the system performance before and after the change.4) Study the existing SRS noise estimation algorithm, and focus on analysis of the minimum mean square error method. Combining the characteristics of the SRS, improve two noise estimate algorithm, i.e., by calculating the energy of the noise region in time domain and the idle subcarriers to estimate the noise.Finally, simulate conventional algorithm and the improved algorithm, and compare with theoretical values to verify the effectiveness of the algorithm.