Understanding the mobility patterns of park users is crucial for optimizing urban planning and park management. However, existing methods often rely on intrusive or labor-intensive data collection techniques, which may not be scalable for large public spaces. Our proposed framework includes a temporal analysis of the daily device count patterns across different day types, as well as investigating popular trajectories in a large park. We identified distinct usage patterns across different areas of the parks, which varied according to the type and time of day. Our findings demonstrate the potential of using WiFi signal data as a scalable and non-intrusive method for mobility data collection.