Reactive Jamming Detection for LoRaWAN Based on Meta-Data Differencing

Reactive jamming in LoRaWAN networks is a stealthy way to implement Denial-of-Service attacks against selected devices because the cause of the interference remains hidden from a network operator. In order to make such attacks more detectable this paper proposes a novel algorithm, which is able to expose a jamming attempt from a single LoRaWAN packet. By monitoring deviations in meta-data of LoRaWAN frames, our method can distinguish a jamming attempt from a normal packet collision with up to 99% accuracy. Furthermore, the presented algorithm is also suitable for light-weight implementations on LoRaWAN devices due to its low complexity.