SenSig:IoT Sensor Fingerprinting Using Calibration Data
A novel, simple and lightweight means of fingerprinting motion sensors.
The Challenge
Sensing technologies are being used in more places: smart homes, hospitals, factories, transport, and other Internet of Things (IoT) systems are just some examples.
As technology is deployed to more situations, new solutions are required to ensure the security and safety of sensing environments.
Traditional security solutions do not consider some of the dependencies and limitations to IoT e.g. power, processing, and memory.
Fingerprinting sensors and devices via novel methods which overcome the current limitations is a challenge.
The Solution
We have identified and developed a novel, simple and lightweight means of fingerprinting motion sensors. By extension, it can also fingerprint the devices a motion sensor resides in.
This solution employs a novel form of quantisation to generate sensor fingerprints from their calibration data.
These fingerprints are able to uniquely identify a given sensor and host device with a very high accuracy.
We believe this technique can be used in various practical applications such as:
- dentification of a sensor in a device or environment
- detection of any faults in known sensors
- detection of external sensors to smart and IoT environments
The Opportunity
Application Description
A simple and lightweight sensor fingerprinting solution based on their calibration data with multiple applications.
The technique could support fingerprinting sensors in sensing and smart environments e.g.:
- homes
- hospital
- factories
- cities
- industrial IoT
- smart devices