arXiv subjects

Dylan Black

Publications and source records attributed to Dylan Black.

4 recordsLinked to original sources

arXiv2026

Neuromorphic Infrared Fibre-Optic Event-Based Sensing with Fast and Efficient Photonic-Electronic Spiking Neurons

Current photonic technologies for remote sensing require the capture of large amounts of data, suffering as a result from high energy consumption, excessive data redundancy and storage, and costly data processing requirements, limiting their ability for direct and efficient edge-processing for rapid decision making and alarm triggering. In contrast, biological sensing systems, given their event-driven nature and in-sensor processing capabilities offer energy-efficient and practical alternatives. In this work, we draw direct inspiration from the neural spiking in biological sensory systems, to propose a novel neuromorphic event-based photonic technology permitting the remote sensing of environmental events-of-interest efficiently, at high speeds and across a wide dynamic frequency range. Our approach combines widely-deployed fibre-optic telecommunication links and photo-detecting resonant tunnelling diodes (pRTD) acting as light-triggered spiking neurons. We demonstrate that this new neuromorphic photonic sensing approach allows the remote detection of different types of environmental events, including temperature variations, strain-induced motion, audio and air turbulence, with high temporal resolution (encoding them with fast nanosecond-rate neural-like spikes). These results pave the way for novel light-enabled neuromorphic remote infrared fibre-optic sensing networks that are fast, efficient, event-driven, scalable, permit direct processing at-the-edge, and offer practical alternatives to current data-intensive approaches for photonic remote sensing.

physics.optics↗

arXiv2025

Ultrafast and compact photonic-electronic leaky integrate-and-fire circuits based upon resonant tunnelling diodes

This work provides a first report of ultrafast and compact photonic-electronic neuromorphic temporal leaky integrate-and-fire neuronal circuits built with Resonant Tunnelling Diodes (RTDs). We demonstrate experimentally that multiple fast (~100-200 ps) optical input pulses, arriving within a short (sub-ns long) temporal window, control the triggering of excitable responses in two different photonic-electronic RTD circuit architectures. These architectures are an electronic RTD coupled externally to a photodetector (referred to as a PD-RTD), and an integrated opto-electronic RTD device with inherent photodetection capability at infrared telecom wavelengths (referred to as an Optical RTD-PD). For both RTD systems, we reveal that the high-speed optically-triggered integrate-and-fire spiking operation can be precisely controlled by acting on the voltage bias applied to the RTD devices, or via the intensity of incoming optical pulses. Finally, we demonstrate the application of the leaky integrate-and-fire behaviour to a pattern recognition task at high-speed, with the systems triggering fast ns-long electrical spikes in response to optical inputs of weighted 4-bit digital headers.

physics.optics↗

arXiv2024

A manufacturable platform for photonic quantum computing

Whilst holding great promise for low noise, ease of operation and networking, useful photonic quantum computing has been precluded by the need for beyond-state-of-the-art components, manufactured by the millions. Here we introduce a manufacturable platform for quantum computing with photons. We benchmark a set of monolithically-integrated silicon photonics-based modules to generate, manipulate, network, and detect photonic qubits, demonstrating dual-rail photonic qubits with $99.98\% \pm 0.01\%$ state preparation and measurement fidelity, Hong-Ou-Mandel quantum interference between independent photon sources with $99.50\%\pm0.25\%$ visibility, two-qubit fusion with $99.22\%\pm0.12\%$ fidelity, and a chip-to-chip qubit interconnect with $99.72\%\pm0.04\%$ fidelity, not accounting for loss. In addition, we preview a selection of next generation technologies, demonstrating low-loss silicon nitride waveguides and components, fabrication-tolerant photon sources, high-efficiency photon-number-resolving detectors, low-loss chip-to-fiber coupling, and barium titanate electro-optic phase shifters.

quant-ph↗

arXiv2022

A Robust Framework for Deep Learning Approaches to Facial Emotion Recognition and Evaluation

Facial emotion recognition is a vast and complex problem space within the domain of computer vision and thus requires a universally accepted baseline method with which to evaluate proposed models. While test datasets have served this purpose in the academic sphere real world application and testing of such models lacks any real comparison. Therefore we propose a framework in which models developed for FER can be compared and contrasted against one another in a constant standardized fashion. A lightweight convolutional neural network is trained on the AffectNet dataset a large variable dataset for facial emotion recognition and a web application is developed and deployed with our proposed framework as a proof of concept. The CNN is embedded into our application and is capable of instant real time facial emotion recognition. When tested on the AffectNet test set this model achieves high accuracy for emotion classification of eight different emotions. Using our framework the validity of this model and others can be properly tested by evaluating a model efficacy not only based on its accuracy on a sample test dataset, but also on in the wild experiments. Additionally, our application is built with the ability to save and store any image captured or uploaded to it for emotion recognition, allowing for the curation of more quality and diverse facial emotion recognition datasets.

cs.CV↗