MAC address randomization is a privacy-preserving mechanism implemented in modern mobile devices whereby the hardware-level MAC address broadcast in Wi-Fi probe requests is periodically replaced with a randomly generated address rather than the device's true, globally unique identifier. This poses a significant challenge for Wi-Fi sensing and people-counting research because systems that rely on tracking unique MAC addresses to count or monitor individuals will overcount or miscount, as a single device may appear as multiple distinct devices over time. Addressing this problem has become a central concern in passive Wi-Fi sensing, with researchers developing approaches such as deep learning-based fingerprinting and privacy-compliant statistical estimation frameworks to infer crowd size without depending on consistent MAC address identity.
Source Papers
- A low-cost automatic people-counting system at bus stops using Wi-Fi probe requests and deep learning ↗ — A low-cost automatic people-counting system at bus stops usi
- Privacy-preserving WiFi fingerprint-based people counting for crowd management ↗ — Privacy-preserving WiFi fingerprint-based people counting fo