The explosive growth of data and information has motivated various emerging non-von Neumann computational approaches in the More-than-Moore era.Photonics neuromorphic computing has attracted lots of attention due to t...The explosive growth of data and information has motivated various emerging non-von Neumann computational approaches in the More-than-Moore era.Photonics neuromorphic computing has attracted lots of attention due to the fascinating advantages such as high speed,wide bandwidth,and massive parallelism.Here,we offer a review on the optical neural computing in our research groups at the device and system levels.The photonics neuron and photonics synapse plasticity are presented.In addition,we introduce several optical neural computing architectures and algorithms including photonic spiking neural network,photonic convolutional neural network,photonic matrix computation,photonic reservoir computing,and photonic reinforcement learning.Finally,we summarize the major challenges faced by photonic neuromorphic computing,and propose promising solutions and perspectives.展开更多
Analog feature extraction(AFE)is an appealing strategy for low-latency and efficient cognitive sensing systems since key features are much sparser than the Nyquist-sampled data.However,applying AFE to broadband radio-...Analog feature extraction(AFE)is an appealing strategy for low-latency and efficient cognitive sensing systems since key features are much sparser than the Nyquist-sampled data.However,applying AFE to broadband radio-frequency(RF)scenarios is challenging due to the bandwidth and programmability bottlenecks of analog electronic circuitry.Here,we introduce a photonics-based scheme that extracts spatiotemporal features from broadband RF signals in the analog domain.The feature extractor structure inspired by convolutional neural networks is implemented on integrated photonic circuits to process RF signals from multiple antennas,extracting valid features from both temporal and spatial dimensions.Because of the tunability of the photonic devices,the photonic spatiotemporal feature extractor is trainable,which enhances the validity of the extracted features.Moreover,a digital-analog-hybrid transfer learning method is proposed for the effective and low-cost training of the photonic feature extractor.To validate our scheme,we demonstrate a radar target recognition task with a 4-GHz instantaneous bandwidth.Experimental results indicate that the photonic analog feature extractor tackles broadband RF signals and reduces the sampling rate of analog-to-digital converters to 1/4 of the Nyquist sampling while maintaining a high target recognition accuracy of 97.5%.Our scheme offers a promising path for exploiting the AFE strategy in the realm of cognitive RF sensing,with the potential to contribute to the efficient signal processing involved in applications such as autonomous driving,robotics,and smart factories.展开更多
Analog-to-digital converters(ADCs)must be high speed,broadband,and accurate for the development of modern information systems,such as radar,imaging,and communications systems;photonic technologies are regarded as prom...Analog-to-digital converters(ADCs)must be high speed,broadband,and accurate for the development of modern information systems,such as radar,imaging,and communications systems;photonic technologies are regarded as promising technologies for realizing these advanced requirements.Here,we present a deep-learning-powered photonic ADC architecture that simultaneously exploits the advantages of electronics and photonics and overcomes the bottlenecks of the two technologies,thereby overcoming the ADC tradeoff among speed,bandwidth,and accuracy.Via supervised training,the adopted deep neural networks learn the patterns of photonic system defects and recover the distorted data,thereby maintaining the high quality of the electronic quantized data succinctly and adaptively.The numerical and experimental results demonstrate that the proposed architecture outperforms state-ofthe-art ADCs with developable high throughput;hence,deep learning performs well in photonic ADC systems.We anticipate that the proposed architecture will inspire future high-performance photonic ADC design and provide opportunities for substantial performance enhancement for the next-generation information systems.展开更多
Optical implementations of neural networks(ONNs)herald the next-generation high-speed and energy-efficient deep learning computing by harnessing the technical advantages of large bandwidth and high parallelism of opti...Optical implementations of neural networks(ONNs)herald the next-generation high-speed and energy-efficient deep learning computing by harnessing the technical advantages of large bandwidth and high parallelism of optics.However,due to the problems of the incomplete numerical domain,limited hardware scale,or inadequate numerical accuracy,the majority of existing ONNs were studied for basic classification tasks.Given that regression is a fundamental form of deep learning and accounts for a large part of current artificial intelligence applications,it is necessary to master deep learning regression for further development and deployment of ONNs.Here,we demonstrate a silicon-based optical coherent dot-product chip(OCDC)capable of completing deep learning regression tasks.The OCDC adopts optical fields to carry out operations in the complete real-value domain instead of in only the positive domain.Via reusing,a single chip conducts matrix multiplications and convolutions in neural networks of any complexity.Also,hardware deviations are compensated via in-situ backpropagation control provided the simplicity of chip architecture.Therefore,the OCDC meets the requirements for sophisticated regression tasks and we successfully demonstrate a representative neural network,the AUTOMAP(a cutting-edge neural network model for image reconstruction).The quality of reconstructed images by the OCDC and a 32-bit digital computer is comparable.To the best of our knowledge,there is no precedent of performing such state-of-the-art regression tasks on ONN chips.It is anticipated that the OCDC can promote the novel accomplishment of ONNs in modern AI applications including autonomous driving,natural language processing,and scientific study.展开更多
We demonstrate a photonic architecture to enable the separation of ultra-wideband signals.The architecture consists of a channel-interleaved photonic analog-to-digital converter(PADC)and a dilated fully convolutional ...We demonstrate a photonic architecture to enable the separation of ultra-wideband signals.The architecture consists of a channel-interleaved photonic analog-to-digital converter(PADC)and a dilated fully convolutional network(DFCN).The aim of the PADC is to perform ultra-wideband signal acquisition,which introduces the mixing of signals between different frequency bands.To alleviate the interference among wideband signals,the DFCN is applied to reconstruct the waveform of the target signal from the ultra-wideband mixed signals in the time domain.The channel-interleaved PADC provides a wide spectrum reception capability.Relying on the DFCN reconstruction algorithm,the ultra-wideband signals,which are originally mixed up,are effectively separated.Additionally,experimental results show that the DFCN reconstruction algorithm improves the average bit error rate by nearly three orders of magnitude compared with that without the algorithm.展开更多
We propose an optical tensor core(OTC) architecture for neural network training. The key computational components of the OTC are the arrayed optical dot-product units(DPUs). The homodyne-detection-based DPUs can condu...We propose an optical tensor core(OTC) architecture for neural network training. The key computational components of the OTC are the arrayed optical dot-product units(DPUs). The homodyne-detection-based DPUs can conduct the essential computational work of neural network training, i.e., matrix-matrix multiplication. Dual-layer waveguide topology is adopted to feed data into these DPUs with ultra-low insertion loss and cross talk. Therefore, the OTC architecture allows a large-scale dot-product array and can be integrated into a photonic chip. The feasibility of the OTC and its effectiveness on neural network training are verified with numerical simulations.展开更多
We experimentally demonstrate the ultra-high range resolution of a photonics-based microwave radar using a high repetition rate actively mode-locked laser(AMLL). The transmitted signal and sampling clock in the rada...We experimentally demonstrate the ultra-high range resolution of a photonics-based microwave radar using a high repetition rate actively mode-locked laser(AMLL). The transmitted signal and sampling clock in the radar originate from the same AMLL to achieve a large instantaneous bandwidth. A Ka band linearly frequency modulated signal with a bandwidth up to 8 GHz is successfully generated and processed with the electro-optical upconversion and direct photonic sampling. The minor lobe suppression(MLS) algorithm is adopted to enhance the dynamic range at a cost of the range resolution. Two-target discrimination with the MLS algorithm proves the range resolution reaches 2.8 cm. The AMLL-based microwave-photonics radar shows promising applications in high-resolution imaging radars having the features of high-frequency band and large bandwidth.展开更多
This Letter investigates the impact of the photodiode(PD) saturation in a sub-sampled photonic analogto-digital converter(PADC) with two individual pulse lasers. It is essentially proved that when the optical power to...This Letter investigates the impact of the photodiode(PD) saturation in a sub-sampled photonic analogto-digital converter(PADC) with two individual pulse lasers. It is essentially proved that when the optical power to the saturated PD increases, the optical–electrical conversion(OEC) responsivity and digitized output power of the PADC decrease. If femtosecond pulses are employed for the PADC sampling clock, the time-stretching process in a dispersive medium is necessary to decrease the impact of the PD saturation. In contrast, when the sampling clock with picosecond pulses is utilized, the PD saturation is more tolerable, and thus, the OEC responsivity can be improved by an increase of the optical power to the PD no matter if the time-stretching process is employed.展开更多
基金This work was supported in part by the National Outstanding Youth Science Fund Project of National Natural Science Foundation of China(62022062)the National Natural Science Foundation of China(61974177,61674119)the Fundamental Research Funds for the Central Universities.
文摘The explosive growth of data and information has motivated various emerging non-von Neumann computational approaches in the More-than-Moore era.Photonics neuromorphic computing has attracted lots of attention due to the fascinating advantages such as high speed,wide bandwidth,and massive parallelism.Here,we offer a review on the optical neural computing in our research groups at the device and system levels.The photonics neuron and photonics synapse plasticity are presented.In addition,we introduce several optical neural computing architectures and algorithms including photonic spiking neural network,photonic convolutional neural network,photonic matrix computation,photonic reservoir computing,and photonic reinforcement learning.Finally,we summarize the major challenges faced by photonic neuromorphic computing,and propose promising solutions and perspectives.
基金supported in part by the National Natural Science Foundation of China(Grant No.T2225023,62205203).
文摘Analog feature extraction(AFE)is an appealing strategy for low-latency and efficient cognitive sensing systems since key features are much sparser than the Nyquist-sampled data.However,applying AFE to broadband radio-frequency(RF)scenarios is challenging due to the bandwidth and programmability bottlenecks of analog electronic circuitry.Here,we introduce a photonics-based scheme that extracts spatiotemporal features from broadband RF signals in the analog domain.The feature extractor structure inspired by convolutional neural networks is implemented on integrated photonic circuits to process RF signals from multiple antennas,extracting valid features from both temporal and spatial dimensions.Because of the tunability of the photonic devices,the photonic spatiotemporal feature extractor is trainable,which enhances the validity of the extracted features.Moreover,a digital-analog-hybrid transfer learning method is proposed for the effective and low-cost training of the photonic feature extractor.To validate our scheme,we demonstrate a radar target recognition task with a 4-GHz instantaneous bandwidth.Experimental results indicate that the photonic analog feature extractor tackles broadband RF signals and reduces the sampling rate of analog-to-digital converters to 1/4 of the Nyquist sampling while maintaining a high target recognition accuracy of 97.5%.Our scheme offers a promising path for exploiting the AFE strategy in the realm of cognitive RF sensing,with the potential to contribute to the efficient signal processing involved in applications such as autonomous driving,robotics,and smart factories.
基金supported by the National Natural Science Foundation of China(grant nos 61822508,61571292,and 61535006)the Shanghai Municipal Science and Technology Major Project(2017SHZDZX03).
文摘Analog-to-digital converters(ADCs)must be high speed,broadband,and accurate for the development of modern information systems,such as radar,imaging,and communications systems;photonic technologies are regarded as promising technologies for realizing these advanced requirements.Here,we present a deep-learning-powered photonic ADC architecture that simultaneously exploits the advantages of electronics and photonics and overcomes the bottlenecks of the two technologies,thereby overcoming the ADC tradeoff among speed,bandwidth,and accuracy.Via supervised training,the adopted deep neural networks learn the patterns of photonic system defects and recover the distorted data,thereby maintaining the high quality of the electronic quantized data succinctly and adaptively.The numerical and experimental results demonstrate that the proposed architecture outperforms state-ofthe-art ADCs with developable high throughput;hence,deep learning performs well in photonic ADC systems.We anticipate that the proposed architecture will inspire future high-performance photonic ADC design and provide opportunities for substantial performance enhancement for the next-generation information systems.
基金This work is supported in part by the National Key Research and Development Program of China(Program no.2019YFB2203700)the National Natural Science Foundation of China(Grant no.61822508).
文摘Optical implementations of neural networks(ONNs)herald the next-generation high-speed and energy-efficient deep learning computing by harnessing the technical advantages of large bandwidth and high parallelism of optics.However,due to the problems of the incomplete numerical domain,limited hardware scale,or inadequate numerical accuracy,the majority of existing ONNs were studied for basic classification tasks.Given that regression is a fundamental form of deep learning and accounts for a large part of current artificial intelligence applications,it is necessary to master deep learning regression for further development and deployment of ONNs.Here,we demonstrate a silicon-based optical coherent dot-product chip(OCDC)capable of completing deep learning regression tasks.The OCDC adopts optical fields to carry out operations in the complete real-value domain instead of in only the positive domain.Via reusing,a single chip conducts matrix multiplications and convolutions in neural networks of any complexity.Also,hardware deviations are compensated via in-situ backpropagation control provided the simplicity of chip architecture.Therefore,the OCDC meets the requirements for sophisticated regression tasks and we successfully demonstrate a representative neural network,the AUTOMAP(a cutting-edge neural network model for image reconstruction).The quality of reconstructed images by the OCDC and a 32-bit digital computer is comparable.To the best of our knowledge,there is no precedent of performing such state-of-the-art regression tasks on ONN chips.It is anticipated that the OCDC can promote the novel accomplishment of ONNs in modern AI applications including autonomous driving,natural language processing,and scientific study.
基金the National Key R&D Program of China(No.2019YFB2203700)the National Natu ral Science Foundation of China(Nos.61822508 and 61571292).
文摘We demonstrate a photonic architecture to enable the separation of ultra-wideband signals.The architecture consists of a channel-interleaved photonic analog-to-digital converter(PADC)and a dilated fully convolutional network(DFCN).The aim of the PADC is to perform ultra-wideband signal acquisition,which introduces the mixing of signals between different frequency bands.To alleviate the interference among wideband signals,the DFCN is applied to reconstruct the waveform of the target signal from the ultra-wideband mixed signals in the time domain.The channel-interleaved PADC provides a wide spectrum reception capability.Relying on the DFCN reconstruction algorithm,the ultra-wideband signals,which are originally mixed up,are effectively separated.Additionally,experimental results show that the DFCN reconstruction algorithm improves the average bit error rate by nearly three orders of magnitude compared with that without the algorithm.
基金supported by the National Key R&D Program of China (No.2019YFB2203700)the National Natural Science Foundation of China (No.61822508)。
文摘We propose an optical tensor core(OTC) architecture for neural network training. The key computational components of the OTC are the arrayed optical dot-product units(DPUs). The homodyne-detection-based DPUs can conduct the essential computational work of neural network training, i.e., matrix-matrix multiplication. Dual-layer waveguide topology is adopted to feed data into these DPUs with ultra-low insertion loss and cross talk. Therefore, the OTC architecture allows a large-scale dot-product array and can be integrated into a photonic chip. The feasibility of the OTC and its effectiveness on neural network training are verified with numerical simulations.
基金partially supported by the National Natural Science Foundation of China(Nos.61571292and 61535006)by the State Key Lab Project of Shanghai Jiao Tong University(No.2014ZZ03016)by STCSM
文摘We experimentally demonstrate the ultra-high range resolution of a photonics-based microwave radar using a high repetition rate actively mode-locked laser(AMLL). The transmitted signal and sampling clock in the radar originate from the same AMLL to achieve a large instantaneous bandwidth. A Ka band linearly frequency modulated signal with a bandwidth up to 8 GHz is successfully generated and processed with the electro-optical upconversion and direct photonic sampling. The minor lobe suppression(MLS) algorithm is adopted to enhance the dynamic range at a cost of the range resolution. Two-target discrimination with the MLS algorithm proves the range resolution reaches 2.8 cm. The AMLL-based microwave-photonics radar shows promising applications in high-resolution imaging radars having the features of high-frequency band and large bandwidth.
基金supported by the National Natural Science Foundation of China(Nos.61822508,61571292,and 61535006)
文摘This Letter investigates the impact of the photodiode(PD) saturation in a sub-sampled photonic analogto-digital converter(PADC) with two individual pulse lasers. It is essentially proved that when the optical power to the saturated PD increases, the optical–electrical conversion(OEC) responsivity and digitized output power of the PADC decrease. If femtosecond pulses are employed for the PADC sampling clock, the time-stretching process in a dispersive medium is necessary to decrease the impact of the PD saturation. In contrast, when the sampling clock with picosecond pulses is utilized, the PD saturation is more tolerable, and thus, the OEC responsivity can be improved by an increase of the optical power to the PD no matter if the time-stretching process is employed.