Device Association Through Passive Wi-Fi Monitoring

Abstract

This research initiative is focused on identifying associated devices using a network of wireless sensors. These sensors collect network traffic meta data in monitor mode. Monitor mode allows a receiver to see certain aspects of all traffic within a network, including the source of data packets, regardless of their destination; leveraging this information with the radiation patterns of the wireless antennas, devices can be identified as geographically close at a specific instant. Over time, this data shows correlations between certain devices, and can be processed to create a network of relationships. Furthermore, the effect of antenna radiation pattern on the sensors' ability to associate devices is examined by using various antenna designs in the experiments. It is seen that antenna patterns which have smaller footprints are better able to distinguish between related and unrelated devices, at the expense of viewing fewer possible interactions. This relationship is analogous to a quality versus quantity decision.