Hidden contaminants such as grass seeds or small rocks trapped in wool or lodged in livestock skin can cause infections, reduce wool quality, and lead to economic losses for farmers. Detecting these infestations early, quickly and non-invasively is a major challenge in agriculture.
This project explores how microwave sensing combined with machine learning can be used to detect such hidden contaminants. Instead of using optical imaging, this approach relies on analysing microwave signals measured using a Vector Network Analyzer (VNA). When microwaves interact with materials, they produce measurable spectral responses. The key challenge is how to process and interpret this raw measurement data to identify meaningful patterns linked to infestation.
In this project, the student will investigate how to:
- Process raw microwave measurement data obtained from a VNA using MATLAB
- Clean, visualise, and organise spectral data for analysis
- Extract relevant features from measured frequency spectra
- Apply machine-learning techniques to classify and detect infestations
- Evaluate detection accuracy and system performance
This project is strongly data-driven and combines signal processing with artificial intelligence. The student will gain practical experience in handling experimental measurement data and transforming it into useful diagnostic information. The project is suitable for students interested in electrical engineering, data science, applied physics, or computer engineering. Basic MATLAB or programming experience and an interest in data analysis will be highly beneficial.
School
Electrical Engineering and Telecommunications
Research Area
Microwave sensing and measurement | Signal processing | Machine learning and pattern recognition
Suitable for recognition of Work Integrated Learning (industrial training)?
No
- Research environment
- Expected outcomes
- Supervisory team
- Reference material/links
The project will be conducted within the Terahertz Innovation Group in School of Electrical Engineering and Telecommunications, led by A/Prof Shaghik Atakaramians. The student will work closely with Dr Amus Goay, Dr Qigejian Alfred Wang, Dr Deepak Mishra, A/Prof Shaghik Atakaramians, also interact with industry partners. The research setting integrates experimental measurements using a VNA with computational analysis in MATLAB. The student will gain exposure to real measurement data, structured data analysis workflows, and machine-learning model development within an active research environment.
By the end of the project, the student is expected to:
- Understand the fundamentals of microwave sensing using a VNA
- Develop skills in MATLAB-based signal processing
- Extract meaningful features from spectral measurement data
- Implement a basic machine-learning model for infestation detection
- Present results in a report or presentation
Strong outcomes may contribute to ongoing research on intelligent sensing solutions for agriculture and livestock monitoring.
The project will be conducted within the Terahertz Innovation Group in School of Electrical Engineering and Telecommunications, led by A/Prof Shaghik Atakaramians. The student will work closely with Dr Amus Goay, Dr Qigejian Alfred Wang, Dr Deepak Mishra, A/Prof Shaghik Atakaramians, also interact with industry partners.
- Thigale, S., Wang, Q., Mishra, D., Goldys, E.M. and Atakaramians, S. (2023). Terahertz imaging: a diagnostic technology for prevention of grass seed infestation. Optics Express, 31(22), 37030–37039.
- Wang, Q., Goay, A.C.Y., Mishra, D., Goldys, E.M. and Atakaramians, S. (2025). Diagnosing Grass Seed Infestation: Convolutional Neural Network-Based Terahertz Imaging. IEEE Access, 13, 16094–16102.