A surface plasmonic neural system to detect and process microwaves
Artificial neural networks (ANNs), based on AI, are now being used in many different settings to help humans solve problems more quickly and efficiently. Electronic engineers are exploring the possibility of running these algorithms on other platforms such as diffractive optic devices.
A team of researchers led by Prof. Tie Jun Cui, a professor at Southeast University in China, has developed a programmable neural net based on a surface electromagnetic waves called spoof surface plasmon polariton. This recently proposed surface plasmonic network (SPNN), introduced in a Nature Electronics paper, can detect microwaves and process them, which may be useful for wireless communications and other technological applications.
Qian Ma told Tech Xplore that optical neural networks and diffractive neural deep networks have recently been identified as promising solutions in digital hardware research to implement artificial neural networks. Previous research on optical neural network showed that it can be difficult to achieve simultaneous high-level programming and nonlinear computing. These ONN devices have usually been limited to certain tasks without programmability or applied only for simple recognition problems (i.e. linear problems).
Source:
https://techxplore.com/news/2023-05-programmable-surface-plasmonic-neural-network.html