Project area/S
- Engineering
Project Details
The project aims to design an antenna array using a pre-trained neural network. You will leverage Pawsey HPC and machine learning to optimize antenna array performance, focusing on the radiation parameters such as beamwidth, beam pointing and side lobe levels. The project combines AI and RF engineering, providing hands-on experience in advanced antenna design and neural network training and optimization.
Student Attributes
Academic Background
Electronics Engineering
Computing Skills
Demonstrated ability to use Python or MATLAB and CAD tools
Training Requirement
Student will learn how to analyse the data from FEKO simulations and post-process the results
Project Timeline
Week 1 | Inductions and project introduction |
Week 2 | Initial presentation, understanding the antenna array design principle |
Week 3 | Using the pre-trained NN for array design |
Week 4 | Developing KPIs, fitness functions |
Week 5 | Investigating the use of MATLAB Antenna Array Design Toolbox and ML Toolbox (optional) |
Week 6 | Generating the optimal array layout |
Week 7 | Generating the optimal array layout |
Week 8 | Simulating the AI-designed antenna array, comparison with FEKO |
Week 9 | Final presentation, report writing |
Week 10 | Final report |
Associated Researchers
Professor David Davidson
Interim Discipline Lead, Electrical and Computer Engineering, Curtin University
Read More