Distributed fiber optic sensing technology based on Brillouin scattering is gradually being applied in engineering monitoring in fields such as infrastructure, rail transit, and submarine cables. The attenuation and noise effects of the photoelectric signal in the Brillouin demodulation system lead to a decrease in the signal-to-noise ratio of long-distance sensing, which limits measurement accuracy. Researchers improve signal-to-noise ratio through methods such as optical pulse coding and Raman amplification, achieving good results while also increasing system complexity and equipment costs. Improving system performance through backend signal processing could reduce hardware costs and facilitate iterative upgrades. This paper reviews the methods of using signal processing methods to improve signal-to-noise ratio in optical fiber sensing systems based on Brillouin scattering both domestically and internationally in the past decade, including one-dimensional signal denoising, image and video denoising, and deep learning denoising technologies. These technologies are discussed and analyzed, and the performance improvement effects of each technology on the system are compared. Finally, the development prospects of signal-to-noise ratio improvement technology in distributed fiber optic sensing system based on Brillouin scattering are discussed.