Assessment of Android Network Positioning as an Alternate Source for Robust PNT.

Android devices employ several methods to calculate their position. This paper's focus is the Network Location Provider (NLP), which leverages Wi-Fi and cell tower signals via the fingerprinting/database approach. Unlike GNSS-based positioning, the NLP should be able to compute positions even when the device is indoors or experiencing GNSS radio frequency interference (RFI), making it an enticing candidate for ensuring robust PNT solutions. However, the inner workings of NLP are largely undisclosed, remaining as a 'black-box' system. Using the Samsung S24 and Xiaomi Redmi K80 Ultra, we explored the NLP's response to GNSS spoofing and offline operation (no network connection), as well as attempting NLP spoofing. The GNSS spoofing test confirmed that when satellite signals are spoofed, the NLP solution is maintained at the truth location. This reinforces the robustness of the NLP in RFI environments. In offline mode, NLP continued to operate without a Google server connection, indicating that it can compute positions locally using internally stored cache data. This behavior deviates from the conventional understanding of NLP and offers valuable insights into the latest NLP mechanism. These findings build upon previous work to uncover the inner workings of the NLP and ultimately contribute to robust smartphone PNT.