Detection of Signal Jammers Targeting Automotive RKE Systems

Sammanfattning

Modern cars use radio signals for Remote keyless entry (RKE) systems. This creates a vulnerability, where thieves can use readily available radio signal jammers to interrupt the communication between the car key and the vehicle. When a car owner presses the lock button, the jammer prevents the signal from reaching the vehicle, leaving it unlocked and vulnerable to theft without the owner's knowledge. This thesis investigates the feasibility of detecting such jamming signals using a low-cost Software Defined Radio (SDR) platform, specifically the KrakenSDR, combined with a GNU Radio signal processing pipeline. A bimodal demodulation scorer is developed to estimate how well a received signal conforms to the two-symbol modulation structure of legitimate RKE transmissions, serving as the primary analytical metric for evaluating the discriminability between jamming signals and legitimate RKE traffic. Three jamming profiles (spot, barrage, and sweep) are evaluated across four simulated distances derived from Friis transmission equations. These simulated distances cover attack scenarios up to approximately 158 m. Results show that the demodulation scorer cleanly separates legitimate keyfob transmissions from spot and barrage jamming signals. Sweep jamming introduces a non-negligible false-positive rate of approximately 8% due to pseudo-modulation induced by the sweeping waveform. The findings demonstrate that SDR-based demodulation analysis is a technically viable and cost-effective basis for civilian jammer detection. It provides a more diagnostically rich alternative to existing binary-alarm hardware detectors, while distinguishing malicious interference from legally operating transmitters such as amateur radio equipment remains an open challenge for any RF-based approach.