<p>Distributed acoustic sensing (DAS) turns a fiber optic cable into a kilometers-long sensing array. These DAS systems offer extended coverage, good sensitivity, and fine spatial resolution. However, DAS signals suffer from low signal-to-noise ratios. Additionally, the sensing signals are negatively impacted by the complexity and intensity of the varying noise sources affecting DAS systems. This paper reviews the recent advances in distributed acoustic sensing denoising (DAS), which are critically needed to improve the reliability and feasibility of DAS systems. The review first outlines the challenges facing distributed acoustic sensing denoising, as well as the various noise sources that affect it. The review then categorizes denoising efforts into hardware redesigns, conventional signal denoising, and deep learning-based techniques. Afterwards, the review highlights the advantages, limitations, and applications of these methods. Lastly, the review concludes by discussing the current open questions and recommendations for the direction of DAS denoising research, which are urgently needed to develop more robust, adaptable DAS systems.</p>