In an era where AI computing is receiving increasing attention, a team of scientists in Germany has developed a potentially game-changing solution.
A research team led by Professor Mario Chemnitz and Dr. Bennett Fischer of the Leibniz Institute for Photonics Technology (Leibniz IPHT) in Jena, Germany, has reportedly developed a computer that uses light waves instead of chips—everything is done through a single optical fiber.
Researchers say they have created an advanced artificial intelligence learning system by utilizing the unique interactions of light waves in optical fibers. This approach eliminates the need for extensive electronic infrastructure, a stark contrast to traditional computing systems that rely on computer chips containing numerous electronic components.
Professor Chemnitz explained, "We can use a single optical fiber to simulate the computing power of numerous neural networks. By utilizing the unique physical properties of light waves, this system will be able to process large amounts of data quickly and efficiently in the future."
Tech Xplore interprets this as follows: Researchers encode various data (whether it's the pixel values of an image or the frequency components of an audio track) into the color channels of ultrashort light pulses. These pulses carry information through optical fibers, undergoing various combinations, amplification, or attenuation. The color combinations that appear at the output of the optical fiber make it possible to predict data types or scenes.
According to reports, the research team successfully applied this method in a pilot study using voice samples provided by Cambridge University to diagnose COVID-19 infection, achieving a detection rate that surpassed the best digital detection system to date.
Professor Chemnitz said, "For the first time, we have demonstrated that the dynamic interactions of light waves in optical fibers can directly classify complex information without the need for any intelligent software."
