Bio
My research interests cover a range of topics from bio-inspired computing to radar signal processing. As part of the Electromagnetic Environment programme (EME) I am currently exploring how different machine learning techniques can be applied to the analysis of radar signals.
If you are interested in any of the topics below, please feel free to reach out and discuss potential collaborations.
Radar signal processing
Detecting and processing radar pulses is a challenging task, especially in a noisy and congested environment. Extracting information about pulses and their potential origin are important tasks within this field. I am working with the EME group to explore how machine learning can aide us in these tasks. We are exploring a variety of machine learning architectures including deep neural networks, transformers and Bayesian inference techniques.
We are also exploring the use of novel hardware to carry out these computations in power constrained environments, such as spiking neural networks, as well as to achieve high throughputs of IQ data, optical computing.
Neuromorphic computing
In the field of bio-inspired computing, neuromorphic engineering, I have been particularly interested in the role of dendrites and their potential to carry out computations.
Memristors
I have also researched novel nanoelectronic devices such as memristors - a two terminal device whose resistance can be tuned.
With these devices I worked to understand a particular operating regime that occurs when the device is stressed with a constant current which is often referred to as the current-transient phenomenon or the subthreshold switching behaviour. My thesis documented how to induce this behaviour within MIM devices, developed theories about its origin and demonstrated uses of the behaviour in edge detection and in replicating homeostasis.