Location-based information has become an attractive attribute for use in many services including localization, tracking, positioning, and authentication. An additional layer of security can be obtained
by verifying the identity of users who wish to access confidential resources only within restricted,
small, indoor trusted zones. The objective of this paper is to construct highly secure indoor areas
primarily by detecting only legitimate users within their work cubicles. In this paper, we present a
fine-grained location-based authentication system (LocAuth) which ensures the physical presence of
the user within his/her small trusted zone (2 meters area). To do this, LocAuth exploits the ambient
wireless network characteristics (e.g., BSSID, SSID, and RSSI) of nearby Wi-Fi and Bluetooth devices observed from each trusted zone. We propose a novel technique called Top-Ranked Network
Nodes (TRNNs) to accurately overcome the fluctuations in wireless signals and enhance the ability
to distinguish targeted trusted zone from neighboring areas. In addition, we developed an application to implement LocAuth on Android-based smartphones and tested it in a real indoor environment.
The tested area is composed of seven adjacent and closely spaced work cubicles located in our lab.
We evaluated LocAuth in two ways: through RSSI-based nearest neighbors (RSSI-based NN) and
through supervised machine learning algorithm (Support Vector Machines). The results of the experiment show the effectiveness of LocAuth by achieving a high classification accuracy (above 98%).
This demonstrates its feasibility in terms of both accuracy as well as fine-grained classification.