RASPNet: A Benchmark Dataset for Radar Adaptive Signal Processing Applications

RASPNet is a newly introduced benchmark dataset tailored for radar adaptive signal processing applications. It encompasses over 16 terabytes of data, making it a substantial resource for researchers in this field. The dataset includes 100 realistic scenarios drawn from diverse terrains across the contiguous United States, providing a broad range of environmental conditions. Each scenario is accompanied by 10,000 clutter realizations, which enhances the dataset's utility for developing and testing data-driven radar models. By offering extensive and varied data, RASPNet aims to support advancements in adaptive signal processing techniques. Its scale and diversity position it as a valuable tool for the radar community seeking to improve detection and classification performance. The dataset's release on arXiv further facilitates accessibility and collaboration among researchers.