Accurate people counting model for smart mobility using WiFi channel state information

In recent years, the advent of wireless communication technologies has revolutionized the transportation industry, enabling the development of smart transportation systems. One such technology that has gained significant attention is Channel State Information (CSI). Effective monitoring of users in these transportation systems is essential for optimizing operations and ensuring passenger safety and satisfaction. Therefore, this paper presents a novel approach to estimating urban mobility through a People Counting Long Short-Term Memory (PC-LSTM) model. Utilizing WiFi CSI, the PC-LSTM method accurately counts the number of passengers in public transport vehicles without compromising privacy. The model outperforms existing methods with an impressive mean accuracy rate of 99.44% across various scenarios, offering a reliable and privacy-preserving solution for smart city infrastructure. This research contributes significantly to the advancement of sustainable urban mobility systems.