Human behavioral information is known to reflect individual characteristics and can be used for user authentication, similar to biometric data like face or fingerprints. Specifically, location information, which details a person’s whereabouts and movement history, strongly reflects personal characteristics and can achieve high accuracy in user authentication within behavioral authentication methods. However, because location information can be easily inferred by others, there is a risk that such inferred information could be used for impersonation. Therefore, it is crucial to enhance resistance to impersonation in user authentication methods that rely on location information, even when the location data is inferred. In this paper, we aim to improve forgery resistance by utilizing not only the location information collected by smartphones but also Wi-Fi data and the correlation between location and Wi-Fi information in the context of user authentication methods. We applied the average fusion and the neural network fusion to combine the three modalities. As a result, we successfully improved resistance to impersonation without compromising authentication performance.