The objective of the Signal Insights Lab is to perform research in exploiting all types of signals coming from various types of sources for surveillance. The domain of the Signal Insights Lab spans from electromagnetic side-channel analysis to bio-acoustics.
Currently, two ongoing research projects require interested students of all levels — undergraduates and postgraduates — to contribute. If you are a student interested in working with me on any of these projects, please drop me an email (asa[at]ucsc[dot]cmb[dot]ac[dot]lk).
- EMvidence: Electromagnetic Side-Channel Analysis for Security and Digital Forensics
- Eloc: Locating and Tracking Elephants through Infrasonic Bio-acoustics
EMvidence: Electromagnetic Side-Channel Analysis for Cyber Security and Digital Forensics
Computing devices generate electromagnetic radiation (EM) due to their internal operations. Interestingly, the EM radiation from the CPU happens to correlate with the internal software behaviour of the device and the data being processed. Electromagnetic side-channel analysis (EM-SCA) is the domain of exploiting such EM radiation to eavesdrop on computers. EM-SCA has applications in both cyber security and digital forensics.
This project is a joint work between the Signal Insight group of University of Colombo, Sri Lanka and the Artificial Intelligence Solution for Cyber Security, Digital Forensics and Smart Systems (ASEADOS) group in the University College Dublin, Ireland.
If you are a student with the following skills, or willing to work hard and gain these skills along the way, this can be the project for you.
- Comfortable with Unix-like environments.
- Programming skills in Python and C languages.
- Experienced with LaTeX document processing environment.
- Good reading and writing skills.
- Hands-on experience on experimental IoT/embedded systems, such as Arduino and Raspberry Pi.
- Experience with software defined radio (SDR) hardware (HackRF, RTL-SDR) and software (GNURadio).
- Knowledge in digital signal processing, machine learning, and relevant software libraries.
Eloc: Locating and Tracking Elephants through Infrasonic Bio-acoustics
Elephants are known to make low frequency sounds, i.e., infrasonic, for their communication over longer distances. Such elephant infrasonic calls have been detected from several kilometres away from them. It has been shown that elephant infrasonic calls can be used for various purposes such as tracking them in their natural habitats for conservation purposes, detecting their approach near human habitats to prevent human-elephant conflicts, etc.
Previously, the SCoRe lab of University of Colombo, Sri Lanka and the Uppsala University, Sweden conducted research in this area, building low-cost hardware/software systems, and carried out preliminary field experiments in Sri Lanka. The next step of the project is currently being planned in collaboration with the International Elephant Project based in Indonesia. The objective is to improve the low-cost infrasonic detector built in the previous project and test it in real-world conditions in Indonesia with the assistance of the International Elephant Project.
Currently, the following two aspects of the project are expected to be conducted by Sri Lankan group to support the project.
Development of Eloc Infrasonic Detectors
The objective of this part of the project is to reproduce the Eloc infrasonic detectors created in the previous project for field experiments. Eloc is a low-cost infrasonic detector built at University of Colombo using off-the-shelf electronics components. Although the device has been tested for its capability, further improvements are necessary to make it usable in real-world scenarios as expected by the International Elephant Project in Indonesia.
If you are a student with the following skills, or willing to work hard and gain these skills along the way, this can be the project for you.
- Experience in designing, building and testing electronic circuits.
- Comfortable with Unix-like environments.
- Programming skills in Python and C languages.
- Experienced with LaTeX document processing environment.
- Good reading and writing skills.
Analysis of Elephant Infrasonic Data
Captured elephant infrasonic datasets using our own Eloc devices and much advanced devices produced by the Cornell university need to be analysed to identify specific patterns related to elephant activities. Building techniques to automatically detect elephant infrasonic patterns in captured data and creating an elephant sound profile is the key objective of this part of the project.
If you are a student with the following skills, or willing to work hard and gain these skills along the way, this can be the project for you.
- Knowledge in digital signal processing, machine learning, and relevant software libraries.
- Comfortable with Unix-like environments.
- Programming skills in Python and C languages.
- Experienced with LaTeX document processing environment.
- Good reading and writing skills.
Please feel free to contact me if you are an interested student in joining any of these projects: asa[at]ucsc[dot]cmb[dot]ac[dot]lk
Asanka Sayakkara 