Physical-Layer Security (PLS) establishes information-theoretic security by leveraging the physical properties of a communication channel, rather than relying on encryption. The key idea is to leverage an asymmetry between the legitimate channel (sender → receiver) and the eavesdropper’s channel (sender → eavesdropper).
When the legitimate channel is better, meaning it has a higher signal-to-noise ratio or less loss, the sender can encode messages in such a way that the receiver can reliably decode the message, while the eavesdropper’s observation is (in the limit) statistically independent of the message, revealing no information.
Achieving this sacrifices some capacity, as part of the legitimate channel’s advantage is spent on “confusing” the eavesdropper. When the eavesdropper’s channel improves, the secure throughput decreases.
PLS can be realized with classical signals or quantum states. With appropriate channel conditions and coding, one can obtain security against an adversary who has access to unbounded computation and optimal quantum measurements.
PLS is particularly attractive when one can make strong, realistic assumptions about the eavesdropper’s channel. Examples include Satellite links, where geometry, path loss, antenna patterns, and pointing constraints create inherent asymmetries, and directional free space links, where highly focused beams limit what an off-axis eavesdropper can capture.
Depending on the circumstances, PLS can be used in conjunction with QKD, or replace it entirely in some scenarios, particularly when one can credibly and durably bound the eavesdropper’s channel. Otherwise, PLS can still harden the physical channel for QKD and provide the initial authentication key, which is necessary to prevent man-in-the-middle attacks.
The assumptions on the channel are what make or break PLS; the channels must be sufficiently asymmetric, and the asymmetry must be guaranteed. If the eavesdropper can improve their channel (e.g., better antenna, closer proximity), the secrecy rate can shrink or vanish.
An additional advantage of PLS is integration. It works well in conjunction with higher-layer security methods and can often be integrated into existing systems with relative ease, reducing reliance on computational hardness assumptions.
back: Security proofs in QKD next: Additional readings