‘A single optical fiber’: Scientists build a silicon-less computer that uses light waves and surpasses existing systems for classification – could this be the ultimate AI CPU?
In an era when the environmental footprint of large-scale AI systems is under increasing scrutiny, a team of international scientists has developed a potentially groundbreaking new solution.
The team, led by Prof. Mario Chemnitz and Dr. Bennet Fischer from the Leibniz Institute for Photonic Technology (Leibniz IPHT) in Jena, has built a computer that uses light waves instead of silicon, all through a single optical fiber .
The researchers have harnessed the unique interactions of light waves within optical fibers to create an advanced artificial learning system. This clever method eliminates the need for extensive electronic infrastructure, which is a marked departure from traditional systems that rely on computer chips containing thousands of electronic components.
Diagnosis of COVID-19 infections
“We use a single optical fiber to mimic the computing power of numerous neural networks,” explains Prof. Chemnitz. “By taking advantage of the unique physical properties of light, this system will enable the fast and efficient processing of large amounts of data in the future.”
If TechXplore explains that data, whether pixel values of images or frequency components of an audio track, are encoded on the color channels of ultra-short light pulses. These pulses carry the information through the fiber and undergo various combinations, amplifications or attenuations. The emergence of new color combinations at the output of the fiber enables the prediction of data types or contexts.
The team successfully applied this method in a pilot study to diagnose COVID-19 infections using voice samples from the University of Cambridge. The results achieved a detection rate that exceeds the best digital systems to date.
“We are the first to demonstrate that such a vivid interplay of light waves in optical fibers can classify complex information directly without any additional intelligent software,” said Prof. Chemnitz.
The article was published in Advanced science.
