Phase-sensitive optical time-domain reflectometry (φ-OTDR) offers advantages such as high sensitivity, resistance to electromagnetic interference, and long transmission distance, and is widely used in fields such as rail transit, pipeline monitoring, marine acoustic-signal detection, and perimeter security. However, the performance of φ-OTDR systems is affected by laser, balanced photodetector, erbium-doped fiber amplifier, and environmental noises. Based on existing wavelet denoising algorithms, this study proposes a new adaptive threshold-calculation method and a new continuous low-error wavelet-threshold function using the wavelet coefficients in each layer wavelet transform as parameters for threshold calculation, as well as analyzes the performance of the new wavelet-threshold function. Experimental results show that compared with wavelet soft- and hard-threshold denoising algorithm, the improved adaptive wavelet-threshold denoising algorithm proposed herein offers better denoising effect.