BloodHound: Early Detection and Identification of Jamming at the PHY-layer

Traditional jamming detection techniques, adopted in static networks, require the receiver (under jamming) to infer the presence of the jammer by measuring the effects of the jamming activity (packet loss and received signal strength), thus resulting only in a-posteriori analysis. However, in mobile scenarios, receivers (e.g., drones, vehicles, etc.) typically experience an increasing jamming effect while moving toward the jamming source. This phenomenon allows, in principle, an early detection of the jamming activity - being the communication not yet affected by the jamming (no packet loss). Under such an assumption, the mobile receiver can take an informed decision before losing the radio connection with the other party. To the best of our knowledge, this paper represents the first attempt toward the detection of a jammer before the radio link is fully affected by its activity. The proposed solution, namely, BloodHound, can early detect the approach to a jammer in a mobile scenario, i.e., before losing the capability of communicating, thus enhancing situational awareness and robustness. We performed an extensive measurement campaign, and we proved our solution to be able to detect the presence of a jammer with an accuracy higher than 0.99 even when the bit error rate is less than 0.01 (early detection), by varying several configuration parameters of the scenario.