Research on urban functional zoning using mobile phone signalling data

Traditional methods for identifying urban functional areas primarily rely on data sources such as remote sensing imagery and points of interest (POIs). However, these approaches suffer from limitations including slow data update cycles and low classification accuracy, making it challenging to accurately identify mixed-use functional zones. With the advent of the big data era, mobile phone signaling data offers massive, continuous, dynamic, and wide-area spatial activity information. This presents significant potential for tracking individual travel behavior and opens up new opportunities for the identification of urban functional areas. Therefore, this study proposes a novel method for identifying urban functional areas using mobile phone signaling data. It comprehensively integrates administrative boundaries, transportation networks, natural barriers, and communication base station coverage cells to delineate the fundamental units of urban space. Building on this foundation, a Gaussian mixture clustering algorithm is proposed to classify the functional types of urban spatial units. By incorporating factors such as the scale of transportation infrastructure, land use characteristics through points of interest, and patterns of individual travel behavior, the method enables accurate identification of seven major types of functional areas, including commercial, residential, mixed residential-public-commercial districts and so on.