Abstract: The Signal-to-Noise Ratio (SNR) is an important parameter to measure the quality of the channel, this paper studies SNR estimation method based on Sounding Reference Signal (SRS) in the Long Term Evolution (LTE) system. Since the noise estimation error of Difference of Adjacent Subcarrier Signal (DASS) algorithm is larger in high SNR region, this paper presents an improved DASS method applicable to SRS. By redefining the differential mode of the subcarriers, the estimation error of the noise in this method is reduced, on the other hand, since the three consecutive SRS frequency points only need to estimate noise once, the complexity of the method is only 1/3 of the original DASS method. Simulation results show that the estimated performance of the proposed method is superior to the rest of the method, especially for the low-latency and medium-latency channel, estimation accuracy of the proposed method is improved by about 10 times in high SNR region.