Abstract
This paper investigates a secure communication framework for Unmanned Aerial Vehicle (UAV) networks where a swarm of energy-constrained drones harvests power from a source signal to serve as relays and "friendly jammers." We analyze a Collaborative Time-Switching Relaying (TSR) protocol and derive closed-form expressions for the Secrecy Outage Probability (SOP) under Shadowed-Rician fading. The results highlight a critical trade-off in energy harvesting timing and demonstrate that cooperative jamming effectively breaks the "secrecy floor" inherent in conventional relaying systems, providing robust protection against sophisticated eavesdropping.
Introduction
UAVs are increasingly deployed as "flying base stations" to provide connectivity in disaster zones or complex urban environments where direct line-of-sight is obstructed. However, these systems face two primary bottlenecks: limited onboard battery life and vulnerability to wireless eavesdropping. While "friendly jamming"—the use of interference to degrade an eavesdropper's channel—is a potent defense, its implementation in energy-starved UAV swarms is challenging. This paper addresses these hurdles by coordinating a swarm to harvest energy from the source signal and then split their roles into data relaying and cooperative jamming.
System Model and Protocol
The proposed system consists of a source ($S$), a destination ($D$), an eavesdropper ($E$), and a swarm of $U$ UAVs. The UAVs are positioned to overcome obstacles that block the direct $S \to D$ link. To model realistic urban environments, the authors employ a Shadowed-Rician fading model, which accounts for both line-of-sight components and shadowing from buildings.
The communication is governed by a Collaborative Time-Switching Relaying (TSR) protocol, which divides the transmission block $T$ into three distinct phases:
- Energy Harvesting (EH): For a duration of $\alpha T$, all UAVs scavenge Radio Frequency (RF) energy from the source's initial broadcast.
- Source-to-UAV: In the next $(1-\alpha)T/2$ interval, the source transmits data. All UAVs receive the signal, though the eavesdropper may also intercept it.
- Relaying & Jamming: In the final $(1-\alpha)T/2$ interval, the UAV with the best link quality acts as an Amplify-and-Forward (AF) relay. Simultaneously, the remaining $U-1$ UAVs utilize their harvested energy to transmit jamming signals toward the eavesdropper.
Secrecy Analysis and Key Insights
The performance of the system is evaluated using the Secrecy Outage Probability (SOP), which measures the likelihood that the secrecy rate falls below a target threshold. The study identifies several critical design factors:
- The Secrecy Floor: In standard relaying, increasing transmit power eventually fails to improve security because the eavesdropper benefits from the stronger signal as much as the destination. Cooperative jamming breaks this "floor" by specifically targeting the eavesdropper's reception.
- The $\alpha$ Trade-off: The time-switching factor $\alpha$ is a double-edged sword. A low $\alpha$ provides insufficient energy for effective jamming, while a high $\alpha$ leaves too little time for data transmission. The research indicates that with jamming, the optimal $\alpha$ is significantly higher (approximately 0.85) than in non-jamming systems.
- Strategic Hovering: To maximize secrecy, the swarm should ideally hover at low altitudes and near the source. This positioning maximizes energy harvesting efficiency and the impact of the jamming signal on the eavesdropper.
Results and Discussion
Monte Carlo simulations validated the derived analytical formulas for SOP, confirming their accuracy across various scenarios, including different eavesdropper combining techniques (Selection Combining and Maximum-Ratio Combining). Key results include:
- Jamming Effectiveness: At high Signal-to-Noise Ratios (SNR), the proposed cooperative jamming framework reduced the SOP by orders of magnitude compared to conventional relaying.
- Swarm Scaling: Increasing the number of UAVs in the swarm consistently improved the secrecy rate, as more "jammers" create a more hostile environment for the eavesdropper without requiring more power from the source.
Limitations and Future Work
While the model provides a robust theoretical foundation, it assumes perfect cancellation of the jamming signal at the destination. Future research should consider residual interference from imperfect hardware. Additionally, moving from static hovering to trajectory optimization would allow UAVs to dynamically adapt to mobile eavesdroppers. Finally, incorporating non-linear energy harvesting models would better reflect the physical limitations of actual RF-to-DC conversion hardware.
Summary updated