Distortion Based Metrics for Physical Layer Security

You are reading version 1 of this direction, frozen 2026-10-04. The latest version is version 2.

Version 1 of 226 papers50 conceptsfrozen 2026-10-04

Traditional cryptography is often an all-or-nothing game: either an eavesdropper recovers the exact message, or they recover nothing. However, in many real-world applications—like streaming video, sensor telemetry, or industrial control—we don't actually need the data to be bit-perfect. We only need it to be "good enough" to be useful.

This realization has sparked a research direction that moves away from "perfect secrecy" toward distortion-based security. Instead of trying to hide every single bit, researchers are designing systems that ensure any data intercepted by an adversary is so garbled (distorted) that it becomes functionally useless.

The Intuition: Security Through Fuzziness

In a standard secure communication setup, a sender (Alice) wants to send a message to a receiver (Bob) while keeping it from an eavesdropper (Eve). In a "lossy" system, Bob is willing to accept a slightly "fuzzy" version of the original data.

Joffrey Villard and Pablo Piantanida in Secure lossy source coding with side information at the decoders demonstrate a powerful trade-off: by allowing Bob to have a tiny amount of reconstruction error, Alice can disproportionately increase the uncertainty experienced by Eve. Essentially, Alice can hide the most sensitive parts of the signal within the "noise floor" of the allowed distortion. A marginal sacrifice in Bob's accuracy can lead to a massive, multi-fold increase in the difficulty Eve faces when trying to recover the signal.

Moving Beyond Information Leakage

Why focus on distortion instead of just counting leaked bits? Standard information-theoretic metrics like "equivocation" measure how much uncertainty an attacker has, but they don't tell us if the attacker's best guess is actually helpful.

As Eva C. Song, Emina Soljanin, and Paul Cuff argue in Rate-Distortion-Based Physical Layer Secrecy in Multimode Fiber, if an attacker's reconstructed message exceeds a certain error threshold, the data is useless regardless of how much "information" was technically leaked. This is particularly relevant in physical layer security, where the goal is to make the intercepted signal too garbled to interpret.

To formalize this, Houqiang Li and Lei Yu introduced the concept of security rate in Secrecy Communication with Security Rate Measure. They define security as the minimum amount of additional information an attacker would need to obtain from an outside source to finally reach a specific level of reconstruction accuracy. This treats security as a cost-benefit problem: how many "bits of help" does Eve need before her guess becomes dangerous?

Practical Mechanisms for Distortion

How do we actually force this distortion on an attacker? Several mechanisms have emerged across different domains:

  1. Geometric Distortion in Control Systems: In cyber-physical systems (CPS), the "distance" of an error matters more than the number of wrong bits. Gaurav Kumar Agarwal and colleagues in Distortion-Based Lightweight Security for Cyber-Physical Systems show that using just 1 to 3 bits of a shared key can maximize the geometric error in an eavesdropper's estimate of a system's state, making it impossible for them to track or maliciously control the system.
  2. Filter Design and MSE Thresholds: In radio communications (MIMO), Hugo Reboredo and João M. F. Xavier propose designing transmit filters that specifically target the Mean-Squared Error (MSE) of the eavesdropper. By forcing Eve’s estimation error above a specific threshold, they ensure her bit error rate is high enough to prevent decryption (Filter Design With Secrecy Constraints: The MIMO Gaussian Wiretap Channel).
  3. Exploiting System Instability: In wireless networked control, Anastasia Impicciatore and George J. Pappas demonstrate a clever trick: by strategically withholding sensor data, one can exploit the inherent instability of a physical plant. This ensures that while Bob (who has the full data) keeps his error bounded, Eve’s estimation error grows exponentially to infinity (Secure state estimation over Markov wireless communication channels).

The Challenge of Measurement

While the intuition is clear, measuring distortion-based secrecy is mathematically difficult. Ersen Ekrem and Sennur Ulukus note in Secure lossy source coding with side information that calculating the physical distortion of an adversary's best possible guess remains an open challenge.

To bridge this gap, researchers like Manjesh Kumar Hanawal and Rajesh Sundaresan have looked at "distortion-tolerant guessing," which analyzes how many attempts a wiretapper needs to find a message that is "approximately correct" (The Shannon Cipher System with a Guessing Wiretapper: General Sources). Similarly, Neri Merhav uses a large deviations framework to measure the probability that an eavesdropper's guess will ever fall below a dangerous distortion threshold (A Large Deviations Approach to Secure Lossy Compression).

By shifting the goalpost from "zero information leakage" to "maximum reconstruction error," this field provides a more practical and flexible framework for securing the next generation of communication and control systems.

Papers in this version

History & citationVersion 2 · 2026-10-04

Current · Permanent link

  1. Version 2 (latest) · 2026-10-04 · 14 papers
  2. Version 1 · 2026-10-04 · 26 papers
Cite this direction

Latest version (a living page):

@misc{lacuna_list_secrecy_e2ba969e,
  title = {list secrecy},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/list-secrecy/rdr_e2ba969e0ba54b6291236262ce57aa5e},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_distortion_based_metrics_for_physical_la_e2ba969e_v1,
  title = {Distortion Based Metrics for Physical Layer Security},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/distortion-based-metrics-for-physical-layer-security/rdr_e2ba969e0ba54b6291236262ce57aa5e/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}