4 November 2020 in Journal of Lightwave Technology
Brillouin optical time-domain analyzer (BOTDA) is used to monitor the temperature and strain along a fiber. So far, neural network and machine learning methods have been successfully applied for temperature extraction. But for different frequency scanning steps, different networks should be designed and trained. Here, a BOTDA assisted by kernel extreme learning machine (K-ELM) with high generalization is proposed and experimentally demonstrated. By utilizing K-ELM, the raw Brillouin gain spectra measured from BOTDA system are classified into different temperature classes. The performance of K-ELM is investigated both in simulation and experiment under different cases of signal-to-noise ratios, pump pulse widths, and frequency scanning steps. Compared with curve fitting methods, the K-ELM algorithm has better measurement accuracy of 0.3 and it realizes great improvement of the processing speed over 120 times. The ultrafast processing speed, high accuracy and universality make K-ELM become a highly competitive candidate for the high-speed BOTDA sensing system.
14 November 2025 in Nature Photonics
Tensor processing is a cornerstone of many modern technological advancements, powering critical applications in data analytics and artificial intelligence. While optical computing offers exceptional advantages in bandwidth, parallelism and energy efficiency, existing methods optimized for scalar operations struggle to efficiently handle tensor-based tasks, limiting their applicability in complex applications, such as neural networks. Here we report parallel optical matrix–matrix multiplication (POMMM), which enables fully parallel tensor processing through a single coherent light propagation. This approach addresses key limitations of current optical methods, scaling the performance with data dimension, while improving theoretical computational power and efficiency. We demonstrate its high consistency with GPU-based matrix–matrix multiplication across both real-valued and complex-valued domains. Moreover, we showcase its adaptability, scalability and versatility in tensor processing applications such as convolutional and vision transformer neural networks. Furthermore, we analyse the theoretical compatibility and efficiency of POMMM in relation to existing optical computing paradigms, highlighting its potential to outperform current state-of-the-art methods. By enabling a variety of computational tasks and supporting multi-wavelength and large-scale expansion, POMMM provides a scalable, high-efficiency foundation for advancing next-generation optical computing.
8 October 2025 in Research Square
Wavelength-selective photodetectors are essential for applications such as hyperspectral imaging, biomedical diagnostics, and secure optical communication. Conventional photodetection systems typically rely on external filters or post-processing to resolve spectral information, leading to increased system complexity and data transfer overhead. Here we report a reconfigurable photodiode based on spatially patterned doped tungsten diselenide (WSe2), which exhibits two runtime switchable photodetection mode and a bidirectional wavelength-dependent conductance modulation across the visible spectrum. Under a broadband photodetection mode, the device exhibited a fast response and a high linear dynamic range of 72 dB. Meanwhile, under color-filtering mode, the device enables non-volatile and color-selective detection spanning from 445 to 780 nm, to experimently achieve the in-sensor spectral processing including color-based logic operations and object trajectory recognition within visible wavelength range. We further show its application in encrypted optical communication using chromatically encoded digit patterns, where the device selectively decodes spectral information via bias-controlled readout. Classification accuracies reach ~98.99% for red and 98.76% for green patterns using a standard convolutional neural network, highlighting the potential of this platform for hardware-level spectral-domain information processing with minimal computational load.
8 May 2024 in IEEE Transactions on Instrumentation and Measurement
This article presents a highly parallelized signal and image recognition method based on spatial optics by joint multiplexing of the spatial position and angular spectrum. By mapping different processor features into different spatial positions and overlaying different angular spectral components of different diffraction sources, this method can simultaneously process multiple sequences using multiple processors. Due to the decimeter-scale length of the optical path and the all-optical processing flow, this method has a nanosecond processing delay and a high-performance power consumption ratio. This method is strictly proven in theory, and a prototype is designed for experimental testing. With the modulation rate at 2 GHz, the experiment’s real-time recognition of over 50 TOPS [trillions (1012) of operations per second (OPS)] is achieved. This method provides a potential solution for applications with real-time requirements such as high-speed recognition of unknown signals, weak signal detection, and system warning supervision due to its extremely low processing delay and power consumption.