Exploiting transmitter I/Q imbalance for estimating the number of active users

The number of active users in a network is crucial for understanding the security level of wireless operating environments, since any node in a network could perform malicious attacks and be a potential threat. In this paper, we propose a novel estimation technique for the number of active users by exploiting a typical device fingerprint - I/Q imbalance, which has been identified as a device-specific hardware impairment and can be utilized to distinguish different wireless devices. In the design, I/Q imbalance of a transmitter is first estimated from its transmitting signals. The estimate is then compared with the observed I/Q imbalances of previously identified users through a hypothesis testing, where the distances between the new estimate and previous observations are adopted as the test metric. If all the distances are larger than a properly selected threshold, a new active user is claimed. Finally, the number of active users is determined by counting all the distinct I/Q imbalances. Simulation results are provided to validate the proposed estimation scheme.