Case · DiggingBeagle record

AiR-ViBeR moved data through desk vibrations into a smartphone accelerometer

AiR-ViBeR research modulated computer-fan speed to create low-frequency surface vibrations that could be sensed by a smartphone accelerometer on the same or an adjacent table.

Laboratory vibrational covert channel requiring compromised fan control and a suitably placed smartphone receiver.

First seen
Apr 13, 2020
Case kind
vulnerability
Claims
1

Reconstruction

AiR-ViBeR moves the carrier from airborne sound to the surface supporting the computer. Malware changes fan speed in controlled patterns. The spinning fan creates mechanical vibration, and those vibrations propagate through a desk or nearby furniture even when the resulting sound is not a useful microphone channel.

A smartphone placed on the same or an adjacent table uses its accelerometer to measure the low-frequency movement and decode the pattern. The underlying cause is an actuator-sensor pairing: software influences fan rotation, fan imbalance and mounting transfer energy into the surface, and commodity accelerometers are sensitive enough to observe some of that movement.

The channel requires prior compromise, controllable fans and physical coupling between the computer and receiver. It is consequently sensitive to furniture construction, fan hardware, placement and background vibration.

Mechanism & boundary

  1. 01

    Run code that can influence fan behavior

    The transmitter must already execute on the isolated system and control or provoke repeatable fan-speed changes.

    Boundary: protected host / cooling actuator

  2. 02

    Encode bits as fan-speed changes

    Secret data is mapped to rotation-speed states or timing patterns.

    Boundary: secret data / fan-control pattern

  3. 03

    Transfer energy into the supporting surface

    Rotating hardware produces low-frequency mechanical vibration that propagates through a desk or adjacent structure.

    Boundary: computer chassis / shared surface

  4. 04

    Sense movement with a smartphone accelerometer

    A suitably placed phone measures the vibration without requiring a microphone channel.

    Boundary: shared surface / receiver sensor

  5. 05

    Decode the vibration pattern

    Receiver software converts the accelerometer signal back into the transmitted bitstream.

    Boundary: accelerometer readings / recovered data

Claims & evidence

Implications

AiR-ViBeR demonstrates that physical coupling through furniture can cross a containment boundary that is neither network nor acoustic in the usual sense. The result is relevant only when a receiver shares enough mechanical coupling with the protected machine, but it broadens the inventory of nearby sensors that a high-assurance environment may need to control.

Controls & mitigations

  • Mechanically isolate protected equipment from surfaces accessible to untrusted devices.
  • Keep smartphones and other sensitive accelerometer-equipped devices outside the immediate physical zone.
  • Constrain software fan-control interfaces and investigate patterned fan behavior where it is distinguishable from normal thermal management.

What remains unknown

  • Performance varies substantially with furniture, mounting, fan design, distance and environmental vibration.
  • The result does not imply useful transmission through arbitrary floors, walls or structurally isolated surfaces.
  • No cited evidence shows an AI system autonomously discovering or operating the channel.

Cite this record

DiggingBeagle. “AiR-ViBeR moved data through desk vibrations into a smartphone accelerometer.” First seen Apr 13, 2020. https://diggingbeagle.com/cases/air-viber-moved-data-through-desk-vibrations-into-a-smartphone-accelerometer/

Citation guidance

Why this archive exists

The source matters after the headline fades.

DiggingBeagle is a non profit research project documenting AI security incidents, agent failures, vulnerabilities and AI-assisted operations. A case keeps its claims beside the sources that support, contest or limit them. Later updates stay visible, so a reader can see when the account changed.

We publish case reconstructions, dated reporting and analysis across records. Each has a different evidentiary role. About the project and our methodology explain how the work is reviewed.