SDN-Blockchain Based Security Routing for UAV Communication via Reinforcement Learning

TL;DR

The paper tackles the challenge of securing low-latency, energy-efficient routing in dynamic UAV emergency networks. It builds an SDN-enabled architecture augmented with blockchain-based trust management and introduces a security degree $SD_i=\alpha RE_i+(1-\alpha)A_i$ to quantify UAV trustworthiness, then presents BSPPO, which uses beam search to pre-screen high-security paths and proximal policy optimization for hop-by-hop routing. The authors formulate a constrained objective that minimizes a weighted sum of delay $\mathcal{T}$ and energy $E_{tot}$ under $\text{SNR}_{ij}\ge\text{SNR}_{min}$, $SD_i\ge\theta_{SD}$, and range constraints, and solve it via the BSPPO algorithm that combines path screening with reinforcement learning. Simulation results show BSPPO outperforms PPO, BS-Q learning, and BS-actor critic in delay, energy, and transmission success rate across varying attack densities and rerouting events, demonstrating robustness and adaptability for secure UAV-based emergency communications. The work offers a practical routing paradigm for mission-critical UAV networks with real-time security supervision and adaptive decision-making, potentially enhancing disaster response and public-safety communications.

Abstract

The unmanned aerial vehicle (UAV) network plays important roles in emergency communications. However, it is challenging to design reliable routing strategies that ensure low latency, energy efficiency, and security in the dynamic and attack-prone environments. To this end, we design a secure routing architecture integrating software-defined networking (SDN) for centralized control and blockchain for tamper-proof trust management. In particular, a novel security degree metric is introduced to quantify the UAV trustworthiness. Based on this architecture, we propose a beam search-proximal policy optimization (BSPPO) algorithm, where beam search (BS) pre-screens the high-security candidate paths, and proximal policy optimization (PPO) performs hop-by-hop routing decisions to support dynamic rerouting upon attack detections. Finally, extensive simulations under varying attack densities, packet sizes, and rerouting events demonstrate that BSPPO outperforms PPO, BS-Q learning, and BS-actor critic in terms of delay, energy consumption, and transmission success rate, showing the outstanding robustness and adaptability.

Figures (5)

• Figure 1: Routing scenario for emergency rescue.
• Figure 2: Rewards and convergence performance under the different algorithms.
• Figure 3: Performances of different algorithms with or without attacks.
• Figure 4: Transmission success rate of different algorithms under the varying numbers of attack nodes.
• Figure 5: Delay and energy consumption performances with attacks of different algorithms under different number of reroutes.