高剂量率单次快脉冲辐射参量测量

记述了“强光一号”高剂量率脉冲辐射装置在状态调试过程中对其出射波形和辐射剂量进行监测的基本情况，实际监测的结果与设计方监测的结果进行了比对，双方的剂量监测值基本一致，偏差在20％以内；双方的波形形状十分相似，但测量得到的波形半宽度却相差10～20ns，最后对双方束流监测产生的差异进行了定性分析。     

激光等离子实验中的低能X射线时间谱解谱方法

利用快速Fourier变换(FFT)算法和反卷积原理对Dante谱议测量波形进行了回推获得了激光等离子体源区低能X射线辐射时间波形。解谱方法采用了W.N.Mcelroy等人提出的SANDⅡ解谱方法并与文献[6]中介绍的限幅迭代和周期性光滑技术相配合计算出了能谱时间关系的16分区结果,给出了1989年6月LF-12强激光装置实验的双束靶、漏靶诊断口,入射口的X射线时间谱以及辐射温度时间关系T_R(i)。     

A coincidence transmission electron microscope with digital waveform measuring system for analytical microscopy

We have been developing a new analytical transmission electron microscope (TEM), called a coincidence TEM, which in principle enables elemental mapping images to be observed at a high signal‐to‐noise (S/N) ratio under very low dose radiation conditions. In this paper, we report the development of a coincidence TEM with a digital waveform measuring system for obtaining a coincidence elemental mapping image. In this system, analog signals detected by a Si(Li) detector and a multianode, position‐sensitive photomultiplier (PSPM) are continuously converted into 12‐bit digital waveform data at a rate of 100 MHz, and transferred to a PC. From the transferred digital waveform data, information on X‐ray photon energy, electron incident position, and detection times of both X rays and electrons are calculated by digital waveform measurement, which lead to the observation of a successful coincidence elemental mapping image. Copyright © 2005 John Wiley & Sons, Ltd.     

一种高质量的2 kb/s语言编码算法MWI

提出一种高质量的低速率语言编码算法ＭＷＩ,该算法对传统的波形插值算法ＷＩ进行了全面的改进。ＭＷＩ采用了一种更加合理的插值模型,包括统一的清音和浊音分析合成模型、新的典型波形的提取和表示方法,并且采用动态规划的算法增加基音周期估计的准确性。由于不依赖于准确的清浊音判别,ＭＷＩ具有更强的抗噪声能力。ＭＷＩ的参数设置也有利于更加有效的参数量化。随着编码速率的下降,ＭＷＩ的质量虽然呈下降的趋势,但在２ｋｂ／ｓ的速率下,它仍然能给出具有很高清晰度和自然度的合成语音。构造的完整的ＭＷＩ２ｋｂ／ｓ的声码器,具有很高的自然度和可懂度。     

超宽带无线通信概述

主要介绍超宽带(UWB)无线通信技术的基本概念、技术演化和主要优点．结合UWB技术的特点，介绍了包括超宽带接收的关键技术、天线、传输特性、信道模型、与其他通信系统的频谱共存兼容技术和基于UWB技术的通信网络协议等UWB系统设计关键领域中的最新成果和研究进展．     

数字合成波形发生器设计

以单片机为核心，对数字波形表各项进行比例换算和线性叠加的数字合成处理。采用定时器中断方式将处理结果经D／A口输出；并对输出信号进行放大、单到双极性变换和滤波处理，获得数字合成的波形信号．通过按键编辑任意波表，改变定时器参数和比例因子，达到任意波获取、波形频率和幅度数字调节目的．采用8住LED显示器动态刷新方式。显示波形类型、频率和幅度．     

时间选择性莱斯衰落信道条件下的OFDM系统

给出了时间选择性莱斯无线衰落信道条件下一种64—QAM 0FDM系统的理论论证，并给出了仿真结果。首先，对波形形成技术和TCM编码使AWGN(加性高斯噪声信道)条件下的0FDM系统性能得到改善进行了探讨；其次，讨论了COFDM(编码0FDM调制)系统抗时间选择性莱斯衰落性能。选择有效的信道编码和波形形成方法可以得到系统的理想抗衰落性能。     

Synthetic seismograms of ground motion near earthquake fault using simulated Green’s function method

Seismograms near source fault were synthesized using the hybrid empirical Green＇s function method where the discretely simulated seismic waveforms are used for Green＇s functions instead of the observed waveforms of small earthquakes. The Green＇s function seismic waveforms for small earthquake were calculated by solving wave equation using the pseudo-spectral method with the staggered grid real FFT strategy under a detailed 2-D velocity structure in Kobe region. Magnitude and seismic moment of simulated Green＇s function waveforms were firstly determined by using the relationship between fault length and corner frequency of source spectrum. The simulated Green＇s function waveforms were employed to synthesize seismograms of strong ground motion near the earthquake fault. The synthetic seismograms of the target earthquake were performed based on the model with multiple source rupture processes. The results suggest that synthesized seismograms coincide well with observed seismic waveforms of the 1995 Hyogo-ken Nanbu earthquake. The simulated Green＇s function method is very useful for prediction of the strong ground motion in region without observed seismic waveforms. The present technique spreads application field of the empirical Green＇s function method.     

深度学习在引力波数据处理中的应用初探

截至2018-01-16,LIGO已成功探测引力波事件6次.可以预期,引力波探测事件会越来越多,引力波天文学会很快进入到大数据阶段.深度学习在大数据处理方面近年来得到迅速发展.它在数据处理速度,准确度等方面都表现出极大的优势.深度学习在引力波数据处理中的应用讨论还不多.本文引入此问题,并对其进行初步研究.引力波数据最大的特点是强噪声、弱信号.现行的数据处理方法是利用匹配滤波的方式把引力波信号从强噪声中挖掘出来.同时,匹配滤波方法还可以确定引力波源的性质,定量确定其参数.匹配滤波方法的弱点是计算量巨大.这导致数据处理速度很慢.对于将来的大数据引力波天文学,这更将是一个巨大的隐患.匹配滤波方法的另一个潜在问题是,完备准确的理论波形模板是其工作的前提条件.这个潜在问题的后果是很难找到理论预期之外的引力波信号.深度学习的数据处理方法有可能在这些问题上提供出路.同时,深度学习也会遇到其自身的若干困难和问题.本文将从网络结构、训练数据制备、训练优化、对信号识别的泛化能力、对数据的特征图表示以及对特征数据遮挡的响应等方面来展开讨论.     

Optical pulse shaper with integrated slab waveguide for arbitrary waveform generation using optical gradient force

Integrated optical pulse shaper opens up possibilities for realizing the ultra high-speed and ultra wide-band linear signal processing with compact size and low power consumption. We propose a silicon monolithic integrated optical pulse shaper using optical gradient force, which is based on the eight-path finite impulse response. A cantilever structure is fabricated in one arm of the Mach–Zehnder interferometer（MZI） to act as an amplitude modulator. The phase shift feature of waveguide is analyzed with the optical pump power, and five typical waveforms are demonstrated with the manipulation of optical force. Unlike other pulse shaper schemes based on thermo–optic effect or electro–optic effect, our scheme is based on a new degree of freedom manipulation, i.e., optical force, so no microelectrodes are required on the silicon chip,which can reduce the complexity of fabrication. Besides, the chip structure is suitable for commercial silicon on an insulator（SOI） wafer, which has a top silicon layer of about 220 nm in thickness.