Abstract: With the large-scale deployment of 5G and the rapid progress toward 6G， unmanned aerial vehicles （UAVs） have drawn increasing attention for their flexible aerial access capabilities and excellent environmental adaptability， making them well-suited for providing communication and sensing services to terrestrial Internet of Things （IoT） systems. Compared with single-UAV systems， UAV swarms offer significant advantages in target coverage， task robustness， and operational efficiency， thereby enhancing overall resource utilization and situational awareness. These capabilities make UAV swarms as critical enablers in a range of application scenarios， including military reconnaissance， disaster response， environmental monitoring， and smart cities， and help establish a solid foundation next-generation wireless networks featuring enhanced sensing and intelligence. Distributed signal processing， as a core enabling technology for achieving efficient collaborative sensing and communication within UAV swarms， has attracted extensive attention from both academia and industry. It refers to a class of methods in which， within a network composed of multiple sensors， communication nodes， or computing units， signal acquisition， processing， and decision-making do not rely on a single central entity. Instead， each node performs partial computations locally and achieves global coordination through limited information exchange. This paper provides a comprehensive survey of distributed signal processing for UAV systems. We first review the technical background， developmental trajectory， and key challenges of the field， and then summarize and analyze research progress across multiple application domains， including sensing， communications， computation， and control. In the sensing domain， we introduce collaborative parameter estimation methods based on traditional models and assess their advantages and limitations in target detection and localization. We also discuss data-driven multi-view information fusion techniques， elaborating on the role of deep learning and graph neural networks in enhancing the accuracy and robustness of multi-UAV collaborative sensing. For communication applications， we survey distributed channel estimation methods and resource allocation strategies， emphasizing their adaptability under bandwidth and energy constraints， and analyzes the implementation of distributed beam management in complex low-altitude propagation environments. With respect to computation， we focus on computation offloading， resource scheduling， collaborative intelligence， and federated learning， highlighting their critical roles in enhancing the real-time performance， energy efficiency， and privacy protection of low-altitude networks. In the area of UAV control， we review distributed formation control， path planning， and pose control， demonstrating how distributed signal processing supports intelligent cooperation and autonomous decision-making in UAV swarms. Finally， considering current research status and technical bottlenecks， we outline promising future directions for distributed UAV signal processing， such as ultra-massive antenna systems， near-field communications， and AI-driven adaptive cooperation.