Abstract: As the main natural space of a city, parks play an irreplaceable role in the life of community residents. For a long time, the rationality of a park layout is measured according to the park size and its service radius. However, traditional methods are unable to accurately verify and analyze whether the actual service scope of a park is consistent with its set service radius, because the actual behavior rules of residents are hard to obtain. New data sources, such as mobile phone signaling data, contain massive spatial information of residents’ living activities and park use data, which can break through the constraints of traditional data. This research selects 42 community parks in downtown Shanghai, and collects the mobile phone signaling data of people who visited the parks in nine days in May 2019 (including four weekends and five working days). It calculates the actual service radius of each park by tracking the residents’ moving behaviors. It has found that the average service radius of community parks in Shanghai is generally large (5,879.9 m). By analyzing the influencing factors of community park service radius, it was found that the community park service radius has a significant positive correlation with park use satisfaction and spatial vitality, a negative correlation with park locations, and no significant correlation with the park’s area and traffic convenience. Community park construction should encourage public participation, transform from “space construction” to “scene construction”, and improve the quality of parks, and incorporate the surrounding public space, integrate the functions and spaces to improve the vitality of the space around the park. Elaborate identification of the service area of a park with mobile phone signaling data can promote the rationality of park planning.