Noise in pipeline pressure signal is a small noise of which the amplitude is much smaller than that of signal. But it affects recognition of pressure drop and results in inaccurate leakage localization or false alarm. Thus, a small noise reduction method based on empirical mode decomposition (SNR-EMD) is proposed to reduce the noise in pipeline pressure signal. SNR-EMD removes the noise considered as small fluctuations in signal around the mean line calculated by signal's upper envelope and lower envelope. Meanwhile, end effect of SNR-EMD is restrained by extrema mirror extension (EME). Then tested by pressure signal in field, SNR-EMD can reduce small noise well through making the denoised signal smooth and conserving the mutation characteristics of the original signal. Finally, negative pressure wave (NPW) combined with SNR-EMD is used to locate pipeline leakage. The case study indicates that pressure drop can be well recognized and leakage can be accurately located. (C) 2016 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.