Scientists at the University of the Witwatersrand (Wits) have developed a “light-powered brain” that can recognise hidden information in complex patterns of light, potentially paving the way for faster communications systems.
The research brings together Wits University’s expertise in artificial intelligence (AI) and physics, demonstrating how light itself can be used to read information without the need for detailed images or extensive computer processing.
Published in Nature Communications , the research was conceived and supervised by Dr Isaac Nape and Professor Andrew Forbes at the Wits Structured Light Laboratory . It was led by MSc student Hadrian Bezuidenhout, with contributions from Cade Peters and Ram Nandan Kumar.
“Creating light patterns that form a large alphabets for communications is only useful if we can read them as fast as we can create them,” says Nape, the Optica Emerging Leader in Optics Chair and a Fellow of the Wits Machine Intelligence and Neural Discovery ( MIND ) Institute Fellow.
“Our work offers a simpler way to do that, picking out the information we need without first taking a detailed picture. This could help turn these unusual light structures into practical carriers of information for future communication systems.”
How does a light-powered brain work?
Light can carry information through specially shaped patterns, much like letters form words. The challenge is reading these patterns quickly, particularly when their appearance changes during transmission. The Wits researchers have developed a system that uses light to recognise features hidden within these patterns, even when they look different.
Nape explains that if you think of a ribbon that has been twisted, if you bend and fold that ribbon into many different shapes, the number of twists remains unchanged. Light can carry similar hidden features, which scientists call topology .
Instead of taking a detailed picture and analysing thousands of individual points, the new system passes light through specially designed layers that direct it towards detectors. Where the light lands reveals the information it carries.
“We don’t need to map every crease in our fabric of light to count the twists. Our network picks out the information that matters and helps us identify the topology quickly,” says Bezuidenhout, now an optics engineer at APSI.
After training on just 15 examples, the light-powered brain successfully recognised 81 different light structures within the family studied, including patterns it had not encountered before.
It also continued working when the researchers introduced noise into the detector, showing its potential for reading information under imperfect conditions.
Bringing AI and physics together
The discovery demonstrates the potential of combining expertise from two fields in which Wits is a leader on the continent: artificial intelligence and quantum science.
Professor Benjamin Rosman, Director of the MIND Institute, says the research highlights the importance of working across scientific disciplines.
“This research is a wonderful example of what becomes possible when we bring together ideas from artificial intelligence and fundamental physics. By embedding neural network principles directly into the behaviour of light, the team is opening up new possibilities for how we process information. It also demonstrates the enormous potential of interdisciplinary research, which is central to what we are building at the MIND Institute.”
Forbes, who directs the South African Quantum Technology Initiative (SA QuTI) and WitsQ, says the breakthrough helps address a major challenge in developing future communications technologies.
“At the Structured Light Laboratory, we are exploring how topology could support a future quantum and classical internet. But creating these light structures is only part of the challenge. We also need practical ways to read them, and that measurement toolbox is still taking shape. This work shows how reaching beyond photonics can help us build it.”
Nape, who is also affiliated with the MIND Institute, says the research could have applications beyond communications, particularly in quantum technologies.
“We hope this is just the beginning. Beyond topology, we want to explore how optical intelligence can contribute to quantum technologies in South Africa. Extracting information from individual particles of light presents even greater measurement challenges, and we are excited to investigate how these tools could help us tackle them.”
Nature Communications