An Indoor Moving Personnel Detection Algorithm Based on Adaptive DBSCAN

This paper aims to enhance the performance of existing 60-GHz millimeter-wave radar-based target detection systems in addressing the challenges of identifying indoor highspeed moving targets. To achieve this issue, an optimized algorithmic framework that innovatively incorporates a Doppler velocity verification mechanism into the conventional signal processing-target tracking-target identification workflow is presented. This mechanism analyses target motion states to effectively discriminate and suppress spurious target interference. Furthermore, adaptive adjustment of two critical Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering parameters - neighborhood radius and minimum points - is achieved through parameter optimization strategies. Experimental results demonstrate that the enhanced system significantly improves target detection accuracy while effectively reduces false alarm rates in indoor environments, thereby providing more reliable technical support for complex indoor target monitoring and recognition scenarios.