宽带毫米波雷达目标时延神经网络识别新方法

基于一维距离像和神经网络研究宽带毫米波雷达目标识别问题,研究了用于雷达距离像序列识别的时延神经网络模型及其学习算法,并提出了基于距离像序列的宽带雷达目标时延神经网络识别方法,还利用三种飞机缩比模型的暗室测量数据,研究了时延神经网络分类器中时延单元数目对分类精度的影响以及分类器的分类性能。实验结果表明：本方法具有良好的应用前景。     

Method for Overflow Attack Defense of SDN Network Flow Table Based on Stochastic Differential Equation

In order to avoid the overflow problem of network flow table caused by hackers attacking the network in the process of using the network, a method for overflow attack defense of SDN network flow table based on stochastic differential equation is proposed. In this method, the stochastic differential equation is first proposed, and the drift coefficient and diffusion coefficient of the equation are expanded and adjusted by Taylor. By using the limit theorem, the spillover attack of SDN network is weakly converged to an approximate two-dimensional Markov diffusion process, and the improved stochastic differential equation is obtained. Then, according to the stochastic nature of SDN network attack, the stochastic differential equation is transformed into an amplitude equation, which is based on the amplitude. The equation establishes a SDN attack detection scheme based on flow table statistics, which detects the spillover attacks of SDN network flow tables. Finally, according to the test results, it is proposed to use other switches instead of network flow table overflow switches to control the data upload rate, thus reducing the possibility of network crash and meeting the attack defense requirements of flow table overflow. The simulation results show that the proposed method has better detection performance and shorter running time, and can provide help for network security related work.

     

三维点云中关键点的配准与优化算法

传统三维(3D)点云配准过程中存在配准误差高、计算量大及耗时长等问题,针对该问题,提出了一种3D点云中关键点的配准与优化算法.在关键点选取阶段,用边缘点检测算法剔除边缘关键点,以提高关键点特征描述的全面性和重复性,降低3D点云配准误差.在3D点云配准阶段,用K-维树(KD-tree)加速的最近邻算法和迭代最近点算法剔除...     

基于改进残差网络的道口车辆分类方法

为了提高模型在道口环境下的车辆图像的特征提取和识别能力,提出了一种基于改进残差网络的车辆分类方法。首先以残差网络为基础模型,改进了残差块中激活函数的位置,并将残差块中的一般卷积用分组卷积代替,引入注意力机制,用焦点损失函数替换交叉熵损失函数。实验部分先用公开数据集StanfordCars进行预训练,再用自建的道口车辆数据集进行迁移学习。结果表明,改进模型在两个数据集中的准确率均优于几种经典的深度学习模型。     

激光辐照铝材表面温度场特征演化的数值模拟

建立激光辐照铝材料的有限元分析模型,对材料表面的温度场进行数值模拟。研究了激光光束在对材料表面扫描过程中激光扫描速度、TEM00及TEM10两种理想模式的叠加比例η的取值、材料厚度等因素对扫描结果的影响。分析了在材料上所取的几个目标点的温度场变化情况。仿真结果表明扫描的速度快慢决定了材料表面可以吸收激光能量的多少,影响材料的最高温度;η的取值决定了激光光束的能量分布情况,η值越高激光光束能量越集中,在扫描过程中目标点的温度变化越剧烈;随着深度的增加,材料内部的温度的最高值逐渐降低,温度的升高趋势逐渐趋于平缓。     

基于业务类型的微波与激光混合链路中继卫星接入控制

针对用户星数据中继接入请求超额定购问题,分析了用户星业务种类,研究了中继卫星接入控制所受多类约束的特点,建立了约束满足模型;以接入业务总优先级权值最高为原则,根据业务的类型、时间有效性和业务优先级设置链路接入灵活度和业务冲突集,以业务的接入顺序和选择接入的中继星链路为优化对象,提出了一种基于业务类型和自适应小窗口蚁群算法双层优化的中继卫星混合链路接入控制算法。仿真实验表明,所建模型可实现多类型业务与多种类天线的接入控制,提出的双层优化算法较单层寻优效果更佳,且具有较快的收敛速度和较强的全局优化能力,因而所提算法可有效解决多约束条件下多业务类型、多种类天线和多中继星的接入控制问题。     

Single event soft error in advanced integrated circuit

As technology feature sizes decrease, single event upset (SEU), and single event transient (SET) dominate the radiation response of microcircuits. Multiple bit upset (MBU) (or multi cell upset) effects, digital single event transient (DSET) and analogue single event transient (ASET) caused serious problems for advanced integrated circuits (ICs) applied in a radiation environment and have become a pressing issue. To face this challenge, a lot of work has been put into the single event soft error mechanism and mitigation schemes. This paper presents a review of SEU and SET, including: a brief historical overview, which summarizes the historical development of the SEU and SET study since their first observation in the 1970's; effects prominent in advanced technology, which reviews the effects such as MBU, MSET as well as SET broadening and quenching with the influence of temperature, device structure etc.; the present understanding of single event soft error mechanisms, which review the basic mechanism of single event generation including various component of charge collection; and a discussion of various SEU and SET mitigation schemes divided as circuit hardening and layout hardening that could help the designer meet his goals.     

Effect of Ge-GeO2 co-doping on non-ohmic behaviour of TiO2-V2O5-Y2O3 varistor ceramics

An investigation was made into the effect of doping with the elemental crystal Ge or/and GeO₂ on the TiO₂-V₂O₅-Y₂O₃ varistor ceramics. The result shows that as the doping contents of V₂O₅ and Y₂O₃ are 0.5 mol%, respectively, co-doping with 0.3 mol% Ge and 0.9 mol% GeO₂ makes the highest α value (α = 12.8), the lowest breakdown voltage V_1mA (V_1mA = 15.8 V/mm) and the highest grain boundary barrier Φ_B (Φ_B = 1.48 eV), which is remarkably superior to the TiO₂-V₂O₅-Y₂O₃ varistor ceramics undoped with Ge and GeO₂ and mono-doped with Ge or GeO₂. The TiO₂-V₂O₅-Y₂O₃-Ge-GeO₂ ceramic has the prospect of becoming a novel varistor ceramic with excellent electrical properties.     

无线激光通信系统中DPIM技术研究

为了改善通信的带宽效率、功率需求,以及信道中信号传输的抗干扰性能,对应用于无线激光通信系统中的DPIM(数字脉冲间隔调制)方式进行了研究。通过仿真分析,比较了OOK(开关键控)、PPM(脉冲位置调制)、DPPM(差分脉冲位置调制)以及DPIM方式的功率、带宽、单位传信率和差错性能。根据DPIM的理论分析,设计了一种以FPGA(现场可编程门阵列)为核心的DPIM系统。结果表明,该调制系统性能稳定,可以达到较高的调制速率。     

基于共轭噪声组的宽带噪声雷达机动目标参数估计

宽带噪声雷达参数估计时通常会采用宽带互模糊函数的方法,但是在处理机动目标时这种方法需要在距离、速度及加速度3维搜索而导致运算量巨大,为此该文提出一种基于共轭噪声组的机动目标参数估计算法。该算法首先根据回波伸缩效应预设多路通道,每路通道截取固定长度的噪声组内信号进行混频,然后利用分数阶傅里叶变换(FrFT)估计混频信号的多普勒相位,根据相位信息构造补偿函数,并对补偿后的噪声组信号利用频域尺度相关(FSC)算法估计回波的时延,最后联立多普勒相位及时延信息获取目标的距离、速度和加速度。该算法避免了目标参数3维搜索的过程,无需时域重构回波信号,较宽带互模糊函数方法极大地降低了运算量,整个算法都可通过快速傅里叶变换(FFT)实现,便于系统实时处理。仿真结果验证了该算法的有效性及优势。