Diffraction Neural Network for Multi‐Source Information of Arrival Sensing

All-optical diffractive neural networks (DNNs) based on passive structures have been shown to perform complicated functions by optical acceleration, thus reforming in situ applications requiring digital computing processing. Its stable, compact, and features make it suitable for many conventional complex electromagnetic environments. Here, a DNN multilayer metasurface array is presented estimate the information of arrival (IOA) over wide range frequency bands incident angles. Unlike traditional methods, multiple correlated or coherent signals interfere with each other, resulting difficult separation which fundamental obstacle multi-source IOA detection. proposed diffraction system creates separate virtual channels isolate process different incoming waves. It recognizes classifies beams, focuses them output plane divided angle working frequency, displays number sources real time at speed light. Furthermore, massive-parallelism data-post-processing-free strategy allows broader harsh environments, such as seismic detection underwater circumstances.