Physicists have made a significant advancement in data processing using a technique known as ‘inverse-design.’ This innovative approach allows algorithms to set up a system based on specific needs, eliminating the need for tedious manual design and intricate simulations. The outcome is a smart ‘universal’ device that utilizes spin waves, called ‘magnons,’ to conduct various data processing tasks while maintaining outstanding energy efficiency. This development represents a major milestone in unconventional computing, promising great potential for future telecommunications, computing, and neuromorphic systems.
An international group of researchers, spearheaded by physicists from the University of Vienna, has made a significant advancement in data processing through an “inverse-design” strategy. This innovative approach enables algorithms to configure a system according to specific functions, eliminating the need for tedious manual design and complex simulations. The outcome is a smart “universal” device that utilizes spin waves (or “magnons”) to perform a range of data processing functions with remarkable energy efficiency. Published in Nature Electronics, this innovation signifies a crucial leap in unconventional computing, with considerable potential for next-generation telecommunications, computing, and neuromorphic systems.
Modern electronics are confronted with major issues, such as high energy usage and growing design complexity. In this regard, magnonics—the use of magnons or quantized spin waves found in magnetic materials—offers an exciting alternative. Magnons facilitate efficient data transportation and processing with minimal energy loss. As the demand for innovative computing solutions expands, from 5G to the anticipated 6G networks and neuromorphic computing (which mimics brain functionality), magnonics signifies a fundamental shift in how devices are engineered and function. The challenge of developing an advanced magnonic processor that supports highly adaptive and energy-efficient computing was successfully tackled by Andrii Chumak from the University of Vienna’s Nanomagnetism and Magnonics Group and his colleagues.
Success Through Experimentation
Noura Zenbaa, the lead author of the study, along with her team working with Dieter Süss from the Physics of Functional Materials at the University of Vienna, constructed a unique experimental apparatus featuring 49 independently controlled current loops on a yttrium-iron-garnet (YIG) film. These loops generated variable magnetic fields to manage and manipulate magnons. By using an “inverse-design” method, the team allowed algorithms to identify the best configurations for achieving the desired functionalities of the device, greatly simplifying the design process. After more than two years of development and trials, the team addressed numerous hurdles. “It was a challenging journey, but seeing everything come together with our first successful measurement was exceptionally rewarding,” commented Noura Zenbaa.
Advancing Eco-Friendly Technologies
The prototype created by the team exhibited two essential functions: serving as a notch filter (a device that blocks certain frequencies) and as a demultiplexer (a device that channels signals to different outputs). These features are vital for next-generation wireless communication technologies, including 5G and 6G. Unlike traditional systems that require customized parts, this adaptable hardware can cater to various applications, thus decreasing complexity, costs, and energy usage. Ongoing research indicates that the device can perform all logical operations on binary data, and with scalability, it could compete with traditional computers. The team aspires to integrate this technology into neuromorphic computing and other advanced systems. Although the current prototype is bulky and energy-intensive, reducing its size to below 100 nanometers could lead to remarkable efficiency, heralding the potential for low-energy, universal data processing and contributing to the development of greener computing technologies. “This project was a daring initiative with numerous unknowns,” reflects Andrii Chumak, the senior author of the study. “However, our initial measurements validated its feasibility—this concept works. Our findings underscore how artificial intelligence is transforming the field of physics, similar to how ChatGPT is revolutionizing text generation and education.”
