基于瑞利散射的布里渊光时域分析系统(BOTDA)因为单端入射的特性而有广泛的应用前景。研究了应用提升小波变换对传感信号进行降噪,与传统小波变换相比,提升小波变换结构简单,运算速度快,易于硬件实现。分别利用累加平均法、传统小波变换和提升小波变换对瑞利BOTDA系统传感信号进行降噪,通过信噪比、均方误差和计算时间对三种方法的降噪效果进行比较。结果表明,累加平均法将信噪比提高了约14dB,传统小波变换将信噪比提高了约18dB,计算速度是累加平均法的3.26倍。提升小波变换将信噪比提高了约19dB,计算时间仅为传统小波变换的约60%。结果表明,提升小波变换适用于瑞利BOTDA系统实时降噪处理。 The Brillouin optical time-domain analysis system based on Rayleigh scattering (BOTDA) has a wide range of applications due to the single-ended incidence characteristics. Compared with the traditional wavelet transform, lifting wavelet transform is simpler in structure, faster in calculation speed and easier in hardware implementation. The signals of Rayleigh BOTDA system are denoised respectively by the means of averaging, traditional wavelet transform and lifting wavelet transform. The noise reduction effects of the three methods are compared by using signal-to-noise ratio, mean square error and calculation time. The results show that the accumulative average method improves the SNR by about 14dB, the SNR by about 18dB with the traditional wavelet transform, and the computation speed is 3.26 times of the accumulative averaging method. Ascending wavelet transform improves the signal-to-noise ratio by about 19dB, and the calculation time is only about 60% of the traditional wavelet transform. The results show that lifting wavelet transform is suitable for real-time noise reduction of Rayleigh BOTDA system.