Beyond the Passive Eavesdropper Assumption

42 concepts43 papers1 proposal

Physical Layer Security (PLS) offers an elegant promise: achieve mathematically provable confidentiality by exploiting the inherent physics of the wireless channel, entirely independent of computational complexity. In the classic PLS narrative, Alice wants to send a message to Bob while an eavesdropper, Eve, listens from the shadows.

To make this problem mathematically tractable, decades of research have relied heavily on the passive eavesdropper assumption. Under this assumption, Eve is modeled as a silent, isolated observer who merely captures transmissions without interacting with the network or revealing her presence. While this abstraction has yielded foundational theoretical insights, a growing body of research reveals that designing systems exclusively around a single, passive adversary creates a dangerous illusion of security. Real-world threats are scaling up in number, coordinating spatially, and actively fighting back.


The Tractable Ideal of the Passive Observer

To understand why the passive eavesdropper assumption is so pervasive, we must first look at the elegant engineering solutions it enables. When an adversary is completely passive, security reduces to a clean channel optimization problem: Alice and Bob must maximize the secrecy capacity—the difference between the capacity of the legitimate channel (Alice-to-Bob) and the wiretap channel (Alice-to-Eve).

Because Eve is silent, Alice and Bob can exploit their known Channel State Information (CSI) to engineer a deliberate statistical disadvantage for her. This is typically achieved through two primary mechanisms:

  1. Spatial Nulling and Beamforming: Multi-antenna transmitters can shape their signals to maximize energy in Bob's direction while steering a radiation null directly toward Eve's suspected location.
  2. Artificial Noise Injection: As demonstrated by Wing-Kin Ma et al. in A robust artificial noise aided transmit design for MISO secrecy, a transmitter can deliberately inject artificial noise (AN) into the null space of Bob's channel. Because the noise is orthogonal to Bob's channel, it degrades Eve's signal-to-noise ratio (SNR) without affecting Bob at all.
   [ Alice ] 
    /     \
   /       \  (Artificial Noise injected into Eve's direction)
  v         v
[ Bob ]   [ Eve (Passive) ]
 (Safe)    (Blinded by Noise)

This paradigm extends naturally to emerging hardware like Reconfigurable Intelligent Surfaces (RIS). For instance, Wei Shi et al. in Secure Outage Analysis of RIS-Assisted Communications with Discrete Phase Control show how dynamically adjusting the phase shifts of reflecting elements can constructively reinforce signals at Bob while destructively interfering at a passive Eve. Even in cryptographic domains like steganography, foundational models by Nicholas J. Hopper et al. in Provably Secure Steganography rely on a "passive warden" who merely monitors communication for hidden patterns without altering the traffic.

In all these cases, the defense works because the environment is static and predictable. The security architecture exploits a fundamental asymmetry: Alice and Bob know the rules of the channel, and a passive Eve cannot alter them.


The Reality of Spatial and Numerical Scale

The passive, single-eavesdropper model breaks down the moment the adversarial threat scales in number or coordinates spatially. When multiple eavesdroppers enter the field, the spatial advantages engineered by Alice and Bob begin to evaporate.

If multiple eavesdroppers operate independently, the probability that at least one of them bypasses defensive nulls increases dramatically. Wei Shi et al. analyze this exact vulnerability in Secure Outage Analysis for RIS-Aided MISO Systems with Randomly Located Eavesdroppers, demonstrating how a spatial field of randomly distributed adversaries alters the secure outage probability landscape.

The threat becomes exponentially worse if these passive eavesdroppers collude. As highlighted by Hao Xu et al. in Latency-aware secure transmission in MEC-assisted HSR communication systems, defenses optimized for an isolated attacker fail against coordinated interception attempts.

            [ Alice ]
             /     \
            /       \
           v         v
        [ Bob ]    [ Eve 1 ] <--- Collusion ---> [ Eve 2 ]
                    \                           /
                     \---> [ Joint Processing ]<---/

Why does collusion break physical layer defenses? When eavesdroppers act independently, they are limited by their local SNR. But when they collude, they can jointly process their intercepted signals. By combining their observations, they effectively synthesize a massive virtual antenna array. This allows them to mathematically cancel out the artificial noise injected by Alice or reconstruct the signal components that were nullified in any single direction.

Furthermore, internal network dynamics can introduce unexpected adversaries. Suili Feng demonstrates in Joint Power Splitting and Secure Beamforming Design in the Multiple Non-regenerative Wireless-powered Relay Networks that "idle" or untrusted relays within a wireless-powered network can easily turn into internal eavesdroppers. Because these nodes are legitimate participants in the network infrastructure, simple spatial exclusion zones or directional beamforming are fundamentally incapable of isolating them.


When Eve Fights Back: The Active Adversary

The most severe limitation of the passive assumption is its complete vulnerability to active manipulation. An active adversary does not just listen; they inject energy, jam signals, or alter data to destroy the channel asymmetries that Alice and Bob rely on.

Consider distributed storage systems (DSS), where data is split across multiple nodes to ensure reliability. Secure cooperative regenerating codes can easily protect data confidentiality against a passive observer who merely reads the stored data (Onur Ozan Koyluoglu et al. in Secure Cooperative Regenerating Codes for Distributed Storage Systems.). However, as Ravi Tandon et al. observe in Toward Optimal Secure Distributed Storage Systems With Exact Repair, an active attacker can manipulate data or introduce malicious errors during the node repair process. This transforms a passive confidentiality problem into an active integrity crisis, rendering passive-only defenses obsolete.

In wireless communications, active adversaries break the core mechanism of physical-layer defenses by manipulating the channel estimation process itself:

  • Blinding Randomization: Physical-layer defenses like channel and modulation randomization—used by Haitham Al Hassanieh et al. to blind passive RFID eavesdroppers (Securing RFIDs by Randomizing the Modulation and Channel)—offer no inherent protection if an active adversary injects high-power jamming signals to completely disrupt the legitimate reader's receiver.
  • Disrupting Key Generation: In physical-layer secret key generation, Alice and Bob extract shared cryptographic keys from the reciprocal, random properties of their shared wireless channel. While a passive eavesdropper merely slows down the rate at which keys can be securely extracted (Ningya Xu et al. in Passive Eavesdropping Can Significantly Slow Down RIS-Assisted Secret Key Generation), an active attacker can execute pilot-spoofing or synchronous jamming attacks. By actively injecting signals during the channel probing phase, the attacker forces the channel states at Alice and Bob to correlate with the attacker's own signal, effectively dictating the generated key.

Proactive Eavesdropping and the Game-Theoretic Shift

Perhaps the most compelling evidence that the passive assumption is insufficient is the emergence of proactive eavesdropping. In this scenario, the traditional boundary between an eavesdropper and an active jammer blurs entirely.

Lingjie Duan explores this mechanism in Proactive Eavesdropping via Cognitive Jamming in Fading Channels. Instead of passively waiting for a strong signal to intercept, a legitimate monitor (or a malicious adversary) actively transmits cognitive jamming signals toward a target communication link.

[ Target Transmitter ] ------------ (High Data Rate) -----------> [ Target Receiver ]
                                                                        ^
                                                                        |
                                                                 (Active Jamming)
                                                                        |
[ Eavesdropper ] <------ (Forced Low Data Rate: Easy to Decode) --------+

By carefully calibrating the jamming power, the eavesdropper deliberately degrades the quality of the target link. To maintain connectivity over this degraded channel, the target transmitter is forced to adapt: it drops its data rate, alters its modulation scheme, or simplifies its precoding. This intentional degradation makes the transmission significantly easier for the eavesdropper to decode.

Proactive eavesdropping demonstrates why passive models are fundamentally incomplete. Adversaries are not static entities bound by the physics of the environment; they can actively reshape the environment's physics to maximize their interception capabilities. Securing future networks requires moving away from static optimization toward game-theoretic frameworks where defenses anticipate an adversary who can actively manipulate channels, coordinate multi-node attacks, and dynamically adapt to defensive countermeasures.

Key Foundational Work

A Formalization of Anonymity and Onion Routing (2004) — contributed 1 concept elements to this direction.
Provably Secure Steganography (2002) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_passive_eavesdropper_assumption_f03a2c25,
  title = {passive eavesdropper assumption},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/passive-eavesdropper-assumption/rdr_f03a2c250cdd41e8b4954ee44e29135a},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-21},
}

This exact version (version 1, frozen text):

@misc{lacuna_passive_eavesdropper_assumption_f03a2c25_v1,
  title = {passive eavesdropper assumption},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/passive-eavesdropper-assumption/rdr_f03a2c250cdd41e8b4954ee44e29135a/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-21},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake