GPU-Accelerated WPA/WPA2 Credential Recovery as an Enabler for Non-Destructive Drone Neutralization

Drones have become a significant threat in both modern warfare and civilian environments, presenting unprecedented security challenges. While hard kill (physical destruction) defense methods exist to counter threat drones, they remain prohibitively expensive and often result in collateral damage. This study focuses on software-based peaceful neutralization as an anti-drone countermeasure, with particular emphasis on optimizing Wi-Fi credential recovery by leveraging exploitable vulnerabilities in threat drone wireless communications. The methodology included extensive performance evaluation experiments to assess vulnerabilities in drone Wi-Fi networks through GPU-accelerated credential recovery. Across the tested GPU platforms, the proposed implementation achieved an average throughput improvement of 8.78% over the Hashcat baseline, with platform-specific gains of 5.0% on H100, 8.77% on A100, 6.53% on RTX PRO 6000, and 14.8% on RTX 4060. The proposed approach represents a preliminary step toward a cost-effective and damage-minimizing alternative, applicable to Wi-Fi-dependent drone platforms under specific operational conditions. Our experimental results suggest the potential feasibility of credential-recovery-based approaches as an enabling step toward software-based neutralization of Wi-Fi-dependent drone systems. This paper contributes to the emerging field of counter-unmanned aerial systems (C-UAS) by establishing a methodology for credential recovery against Wi-Fi-based drone communications, which may contribute to reducing reliance on physical intervention in applicable scenarios.