C2B-DroneNet: cyber clone-driven blockchain process for secure and efficient drone network operations
摘要
The Internet of Drones (IoD) is becoming foundational for collaborative aerial services in agriculture, inspection, disaster response, and logistics, where fleets must coordinate sensing, computing, and decision-making at scale. However, resource constraints on UAVs (limited energy/CPU), dynamic topologies, and latency sensitivity, as well as security threats (identity spoofing, rogue task injection, block forgery, and resource exhaustion), continue to impede reliable and large-scale deployment. We present C2B-DroneNet, a Cyber Clone-Driven Blockchain framework that unifies: (i) operation externalization, which offloads compute-intensive tasks from resource-constrained drones to fog nodes or capable peers; (ii) cyber clones, cloud-hosted predictive orchestrators that mirror drone states and proactively coordinate scheduling, failure avoidance, and routing; and (iii) a lightweight blockchain with L-PBFT consensus and elliptic-curve signatures to provide tamper-evident, low-overhead trust. Unlike prior IoD solutions that address offloading, security, or orchestration independently, this unified integration enables simultaneous resource optimization, Byzantine fault tolerance, and predictive coordination; a combination not achieved in existing frameworks. A hybrid reinforcement learning (RL) + Graph Neural Networks (GNN) decision engine exploits topology-aware embeddings and real-time telemetry to optimize offloading under mobility and deadline constraints. In simulation, C2B-DroneNet improves battery life by up to 18%, reduces latency by 68%, achieves 14% faster task completion, 12% higher throughput, and 6% higher decision accuracy relative to strong baselines. A security evaluation shows that over 99% of spoofing attempts are thwarted, malicious block forging is rejected by consensus, and deposit-reward economics suppress spam/denial-of-service (DoS). Overall, C2B-DroneNet contributes to building resilient infrastructure by unifying adaptability, computational offloading, and robust defense mechanisms for next-generation IoD environments.