What is it about?

For our everyday life, we are dependent on critical infrastructures, for example, water treatment systems, waste management plants, energy systems, autonomous transport, and communication systems. These systems are much more connected with each other, with people, and with the devices on the internet, than they used to be in the past. If not enough attention is given these critical services could fall prey in the hands of bad actors and it is even easier now since these systems are connected to the internet. We venture to ensure these systems work securely and safely and citizens can enjoy these critical services without any disruption.

Featured Image

Why is it important?

The critical systems like industrial control systems or industrial internet of things have few major components, those include sensors to measure the physical surroundings, controllers to compute and keep the system in the desired state based on the sensor measurements, and the actuators to drive the physical system, e.g., an electric motor to run an autonomous train. The challenge is to ensure all the devices are securely operating as desired and have not been modified with malicious intent. Imagine if a sensor responsible to detect humans in an autonomous vehicle fails or gets compromised/hacked then the consequences can be catastrophic. Our work is in the direction to ensure that we can detect such a malicious or accidental failure of a sensor. Surprisingly we achieve our goal by using an undesirable feature called the "noise" of the sensor. Noise is defined as the inaccuracies in measurements due to sensors manufacturing imperfections. Therefore, it is shown that the noise for each sensor is different and makes a unique identifier for each sensor.


This article for me is an amazing contribution because it is surprising to see how an undesirable feature of a device is converted into a useful tool to secure the devices. This work is built on top of lessons learned from so many great works in the field of device identification or device fingerprinting.

Chuadhry Mujeeb Ahmed
Singapore University of Technology and Design

Read the Original

This page is a summary of: NoiSense Print, ACM Transactions on Privacy and Security, January 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3410447.
You can read the full text:




The following have contributed to this page