In terrestrial wireless networks, buildings, terrain, and walls act as natural shields. As radio signals bounce off obstacles, they experience severe attenuation and multi-path fading. While this degrades signal quality, it provides a form of "environmental masking"—an eavesdropper sitting a few blocks away often receives nothing but un-decodable noise.
Next-generation wireless architectures, however, are stripping this natural shield away. High-altitude platforms, such as Unmanned Aerial Vehicles (UAVs) acting as flying base stations or satellites broadcasting from orbit, rely on pristine, unobstructed Line-of-Sight (LoS) links to maximize throughput and coverage.
But this visibility is a double-edged sword. As noted by Jiahui Li et al. in Two-Way Aerial Secure Communications via Distributed Collaborative Beamforming under Eavesdropper Collusion, the elevated position of an aerial transmitter creates a massive broadcast footprint with severe "geometric exposure." A signal beamed down to a legitimate ground station is frequently just as visible to any unauthorized observer or malicious warden in the vicinity. Because legacy protocols often prioritize transmission efficiency over cryptographic security, these open channels are highly vulnerable to interception and signal injection.
To secure these inherently exposed channels, researchers are turning to Physical Layer Security (PLS). Instead of relying solely on upper-layer cryptographic encryption—which introduces computational overhead and latency—PLS exploits the physical characteristics of the wireless channel to protect data.
The Core Mechanism: Engineering Channel Asymmetry
The fundamental goal of physical layer security is to maximize the secrecy capacity, which is mathematically defined as the difference between the capacity of the legitimate communication channel and the capacity of the eavesdropper’s channel.
$$\text{Secrecy Capacity} = C_{\text{legitimate}} - C_{\text{eavesdropper}}$$
If the system can force the eavesdropper's channel to be fundamentally worse than the legitimate user's channel, secure communication is theoretically guaranteed without a pre-shared cryptographic key.
In a standard terrestrial network, path loss and obstacles naturally create this asymmetry. In an open LoS aerial network, however, the eavesdropper often enjoys an equally clear line of sight to the transmitter, meaning $C_{\text{eavesdropper}}$ can match or exceed $C_{\text{legitimate}}$. To break this symmetry, PLS frameworks dynamically manipulate three main degrees of freedom: mobility, spatial phase alignment, and surface reflection.
Terrestrial (Environmental Masking):
Transmitter ----[Obstacle]----> Eavesdropper (Weak Signal)
Transmitter ------------------> Legitimate User (Strong Signal)
Aerial Line-of-Sight (Geometric Exposure):
Transmitter ------------------> Eavesdropper (Strong Signal)
Transmitter ------------------> Legitimate User (Strong Signal)
Exploiting Mobility via Trajectory Optimization
Unlike fixed terrestrial infrastructure, aerial nodes possess a unique advantage: controlled mobility. By dynamically planning flight paths, an aerial transmitter can physically alter the geometric distances of the channels in real time.
In Trajectory and power design for aerial CRNs with colluding eavesdroppers, Hongjiang Lei et al. demonstrated a joint trajectory and power optimization framework. By utilizing successive convex approximation, the UAV continuously repositions itself to maximize its distance from potential ground-based eavesdroppers while closing the gap to the legitimate receiver.
When transmissions must be completely hidden from a warden (covert communication), the authors in Aerial Relay to Achieve Covertness and Security showed that combining trajectory design with a "phase-switching" factor allows a drone to mask its transmissions within the background noise of the environment, ensuring security even when the exact locations of malicious nodes are unknown.
Spatial Engineering: Collaborative Beamforming and Friendly Noise
When a single aerial node cannot escape an adversary's line of sight, multiple nodes can coordinate to manipulate the spatial distribution of the signal. This is achieved through two primary techniques: distributed beamforming and cooperative jamming.
Distributed Collaborative Beamforming
By treating a swarm of independent drones as a single virtual antenna array, transmitters can control the phase of their signals. Jiahui Li et al. leveraged this to implement distributed collaborative beamforming. By calculating precise phase offsets, the swarm ensures that the individual waveforms interfere constructively at the legitimate receiver (boosting signal strength) but interfere destructively at the eavesdropper's coordinates, effectively creating a signal "null" in the direction of the threat.
Cooperative Friendly Jamming
Instead of trying to hide the signal, the network can choose to actively blind the adversary. Meng Hua et al. proposed a framework in Energy-Efficient Cooperative Secure Transmission in Multi-UAV Enabled Wireless Networks that divides a UAV swarm into "Source" and "Jamming" nodes. While the source drones transmit data, the jamming drones emit artificial noise targeted directly at the eavesdropper.
This decentralized approach shifts the burden of interference management away from the primary transmitter. As Lun Dong demonstrated in Secure Wireless Communications via Cooperation, cooperative relays can successfully null out signals at an eavesdropper's location even if the adversary initially possesses a superior direct link to the source. To overcome energy constraints, Hanh Dang-Ngoc et al. showed that these friendly jammers can even harvest wireless energy from the primary source signal before re-radiating it as defensive interference.
[Source UAV] -------- Data Signal --------> [Legitimate User] (Decodes Data)
\ ^
\--- Data --- | (Interference)
\ |
[Jamming UAV] ---- Artificial Noise ----> [Eavesdropper] (Blinded)
Reshaping the Channel with Reconfigurable Surfaces
The newest frontier in PLS bypasses the need to move the transmitter entirely, focusing instead on dynamically altering the propagation environment using Reconfigurable Intelligent Surfaces (RIS) and Movable Antennas (MAs).
An RIS consists of an array of passive reflecting elements that can independently shift the phase of incoming electromagnetic waves. In UAV-IRS Assisted Covert Communication: Introducing Uncertainty via Phase Shifting, Xinying Chen et al. utilized an IRS to introduce deliberate, rapid phase perturbations into the environment. This creates artificial channel fading for any unauthorized observer, preventing them from cleanly tracking or decoding the transmission.
Taking this concept further, researchers are mounting antennas on mechanical tracks, allowing them to slide continuously within a localized space.
- In Secure Transmission for Movable Antennas Empowered Cell-Free Symbiotic Radio Communications, Jiayu Guan et al. demonstrated that adjusting the physical positions of antennas at an access point optimizes the local spatial channel, significantly widening the secrecy gap between users and eavesdroppers.
- Jingjing Zhao et al. expanded this to full-space coverage in Movable-Element STARS-Aided Secure Communications, using movable elements that simultaneously transmit and reflect signals to eliminate blind spots and protect data across a full 360-degree radius.
Defending Against Passive Threats
A persistent challenge in physical layer security is the "silent" or passive eavesdropper—an adversary that intercepts signals without transmitting anything that would reveal their location.
To counter this, modern architectures are merging security with radar-like capabilities. In Near-Field Integrated Sensing and Communications for Secure UAV Networks, Jingjing Zhao et al. proposed an Integrated Sensing and Communications (ISAC) framework operating in the antenna's near-field. By exploiting the spherical properties of near-field wave propagation, the aerial base station actively senses and tracks the physical 3D coordinates of passive intruders. The system then uses this real-time tracking data to dynamically steer communication beams away from the detected coordinates, achieving robust physical layer isolation without requiring any cooperation from the eavesdropper.