【摘 要】
：
Matrix computation, as a fundamental building block of information processing in science and technology, contributes most of the computational overheads in modern signal processing and artificial intelligence algorithms. Photonic accelerators are designed
【机 构】
：
Wuhan National Laboratory for Optoelectronics,Huazhong University of Science and Technology,Wuhan 43
论文部分内容阅读
Matrix computation, as a fundamental building block of information processing in science and technology, contributes most of the computational overheads in modern signal processing and artificial intelligence algorithms. Photonic accelerators are designed to accelerate specific categories of computing in the optical domain, especially matrix multiplication, to address the growing demand for computing resources and capacity. Photonic matrix multiplication has much potential to expand the domain of telecommunication, and artificial intelligence benefiting from its superior performance. Recent research in photonic matrix multiplication has flourished and may provide opportunities to develop applications that are unachievable at present by conventional electronic processors. In this review, we first introduce the methods of photonic matrix multiplication, mainly including the plane light conversion method, Mach–Zehnder interferometer method and wavelength division multiplexing method. We also summarize the developmental milestones of photonic matrix multiplication and the related applications. Then, we review their detailed advances in applications to optical signal processing and artificial neural networks in recent years. Finally, we comment on the challenges and perspectives of photonic matrix multiplication and photonic acceleration.
其他文献
为精确规划激光跟踪仪的测量站位,提高飞机装配效率,本文提出基于虚拟力场的激光跟踪仪可测性量化判定方法.首先,根据测量中的距离约束、垂直测量角约束和障碍物可视性约束,构建激光跟踪仪可测性约束表达式.然后,引入虚拟力场,改进虚拟斥力公式,并定义阈值公式,构建障碍物可视性判定模型.最后,构建可测性矩阵,以实现激光跟踪仪的可测性量化判定.试验结果表明,与势能场法对比,基于虚拟力场的激光跟踪仪可测性量化判定方法具有较强的可行性和较高的准确率.
近年来,随着无线通信技术在军事领域的发展及各类无线通信装备数量的增加,战场电磁环境变得越来越复杂,其要求通信系统具有更强的抗干扰能力,而智能干扰识别是抗干扰的前提.基于传统机器学习的各类干扰识别算法存在干扰前期特征提取繁杂、低干噪比(JNR)下识别正确识别率低的问题.本文针对上述问题,引入了并行多路多尺度卷积,通过增加网络宽度、提升网络识别速度和精度,解决前期特征提取繁杂的问题;结合频带相关性长短时记忆网络(LSTM)创新性地提出了基于LSTM的频带相关性无线通信干扰智能识别网络,该方法利用LSTM探知干
Hyperbolic metamaterials with a unique hyperbolic dispersion relation allow propagating waves with infinitely large wavevectors and a high density of states. Researchers from Korea and Singapore provide a comprehensive review of hyperbolic metamaterials,
Biography:Elizabeth (Liz) Rogan was appointed Optica Chief Executive Officer in 2002. Previously, she served as the organization\'s Chief Operating Officer.rnUnder Rogan\'s guidance, the society\'s reserves have grown to US$160M, annual budget has g
The high degrees of freedom of light, various optical structures and optical materials can be explored and applied to develop optical encryption for securing information. An exciting optical image encryption approach has been proposed based on spatial non
Unknown speckle patterns can be used to image targets embedded in complex scattering media 100 times faster than previous techniques based on carefully calibrated illuminations.
The detection of orbital angular momentum usually relies on optical techniques, which modify the original beam to convert the information carried on its phase into a specific intensity distribution in output. Moreover, the exploitation of high-intensity b
The solar X-ray and Extreme Ultraviolet Imager (X-EUVI), which was developed by CIOMP, is China\'s first space-based solar X-ray and Extreme Ultraviolet (EUV) imager;it has been loaded onto the Fengyun-3E Satellite, which is supported by the China Meteo
受油包线范围的确定对硬式空中加油的成功对接、安全加油具有重要意义.本文提出了一种不同干扰等级下受油机飞行边界数值认定方法.首先,通过解算KC-135加油机的硬管包线给出受油机的加油可行范围.其次,在分析受油机纵向及横侧向运动模型的基础上,设计受油机多模态双闭环控制系统,使受油机速度、姿态角、位置等状态量能够稳定保持.最后,通过对受油机各模态的数字仿真分析,探析不同干扰等级下的受油机飞行边界范围,并规定受油机飞行安全品质等级.结果表明,受油机在Ⅳ级干扰的环境下不再适合空中加油任务,这对空中加油试飞验证具有很
Light:Science&Applications 11, 6 (2022)rnhttps://doi.org/10.1038/s41377-021-00694-4rnTwo-color infrared detection technology realizes target recognition in a complex environment by using the multi-spectral characteristics of the target. In the last decade