Enhancing the resilience of UAV networks against GPS spoofing attacks via byzantine distributed detection algorithm

Unmanned Aerial Vehicles (UAVs) have been widely adopted for diverse tasks in recent years due to their high flexibility, versatility, and rapid deployment. However, the navigation and coordination of UAV swarm heavily rely on global navigation satellite systems, rendering them susceptible to GPS spoofing attacks. Such attacks not only disrupt individual UAV movement, impair obstacle avoidance, and compromise the resilience of the networked UAV system, ultimately jeopardizing both the system’s safety and the completion of its mission. Most spoofing detection methods require substantial computational resources due to high complexity, limiting their applicability in resource-constrained environments like UAVs, where limited onboard processing ability exacerbates the problem. Here, we propose a novel and computationally efficient method, Byzantine Distributed GPS Detection (BDGD) algorithm, that fuses received signal strength indicator-based distance checks with an adaptive reputation scoring mechanism to counter GPS spoofing attacks in UAV networks to achieve consensus on genuine positions without resorting to global synchronization. By promptly identifying and isolating the malicious UAVs affected by spoofing attacks, our BDGD algorithm can effectively restore the stability and connectivity of the UAV network, and safeguard mission performance. Experimental results demonstrate that BDGD significantly improves the packet delivery ratio and reduces end-to-end delay in GPS spoofing scenarios. Moreover, BDGD achieves faster recovery and higher stability compared to baseline methods, highlighting its effectiveness in enhancing the resilience of UAV networks against GPS spoofing attacks.