基于节点相关性的网络不动点理论研究

现代复杂的通信网络内部存在着广泛的幂律现象，网络节点之间存在相关特性. 根据这种相关特性，提出了网络不动点理论. 将Banach不动点理论引入网络模型，证明了网络不动点理论的正确有效性. 证明过程是把通信网络看作由路径预测算法产生的似马尔可夫链的路由节点迭代序列形成的网络空间. 由节点相关性可知，此空间中的节点序列相对越长就越能折射出搜索的目标所在，预测准确率也会逐步增加，可以更好地进行目标定位、数据挖掘等. 通过某种路由准则的算子从源节点最终映射到的目的节点与Banach空间的不动点相对应，即为网络空间的不动点. 当网络发展到能为用户提供真正的无处不在的连接时，网络不动点理论的物理特性将非常明显. 因为网络规模越大，节点间的群体作用越显著，就越能显现网络不动点理论的物理特性. 关键词： 计算机网络 长程相关 不动点 幂律     

基于平滑高斯半马尔可夫模型的移动无线传感器网络生成树算法

针对以往移动无线传感器网络研究只是单纯地对移动群体进行分簇而没有充分利用组群移动的内部稳定性的问题,结合组移动模型中节点运动的规律和内聚性原理,采用平滑高斯半马尔可夫移动模型刻画组内单个节点移动特征,构建了一种适合移动网络的稳定生成树算法(GM-base stable spanning tree algorithm,简称GSST)。实验证明,该算法从单个节点运动变化入手,在预测未来节点运动情况,选择稳定的链路构建网络结构方面,提高了移动网络的稳定性;同时,利用树的分层特征,简化移动网络的组网过程,并实现网络重组局部化。该算法有效延长节点存活率,均衡数据传输量。     

基于SOM和RBF网络的软测量方法研究

针对氢粉碎过程中钕铁硼粉碎状态不可知,为有效预测合金的反应状态,提出了一种基于自组织特征映射(SOM)神经网络和径向基函数(RBF)神经网络结合构建的网络模型。在该模型中,SOM神经网络作为聚类网络,采用无教师学习算法对输入样本进行自组织分类,并将分类中心及其对应的权值向量传递给RBF神经网络,作为径向基函数的中心;RBF神经网络作为基础网络,采用高斯函数作为径向基函数实现从输入到隐含层的非线性映射,输出层则采用有教师学习算法训练网络的权值,从而实现输入层到输出层的线性映射。并以钕铁硼氢粉碎过程合金中氢含量为检测对象,运用上述方法在MATLAB平台上建立了合金中氢含量预测模型,并完成了仿真验证。     

一种基于β-warping变换算子的被动声源距离估计方法

基于浅海简正波水平波数差与波导不变量之间的关系, 本文提出了一种适用于水平不变浅海声波导中接收信号自相关函数的频域卷绕变换算子. 该算子可以将接收信号自相关函数中的简正波互相关成分变换为时域上可分离的脉冲序列, 且脉冲序列的相对延迟时间包含声源距离信息. 利用已知距离的引导声源, 由单水听器记录的脉冲信号即可实现被动声源距离估计. 对仿真和实验获得的脉冲信号数据处理结果验证了该变换算子用于被动声源距离估计的有效性.     

环形加权网络的时空混沌延迟同步

研究了环形加权网络的时空混沌延迟同步问题.以随时间和空间演化均呈现混沌行为的时空混沌系统作为网络的节点,通过环形加权连接使所有节点建立关联.基于线性稳定性定理,通过确定网络的最大Lyapunov指数,得到了实现网络延迟同步的条件.在最大Lyapunov指数小于零的区域内,任取节点之间耦合强度的权重值,均可以使整个网络实现延迟同步.采用具有时空混沌行为的自催化反应扩散系统作为网络节点,仿真模拟验证了该方法的有效性. 关键词： 延迟同步 加权网络 时空混沌 Lyapunov指数     

一种自动分割跟踪视频运动对象的方法

针对视频图像中单个运动对象的分割和跟踪问题,提出了一种基于时空离散度的视频对象分割跟踪算法。该算法首先采用时空离散度进行运动对象区域的粗定位;然后通过形态学算子来进行细定位,并获取运动物体的二值模型;再然后使用部分Hausdorff距离实现后续帧中运动对象模型的匹配;最后为了容纳运动对象在视场中形状的变化,在图像距离空间中引入了欧几里德范数作为约束条件来完成运动物体模型的更新。实验表明,在运动对象跟踪过程中相邻位置差未超过2个像素点,从而验证了算法的有效性。     

激光时空混沌模型的加权网络投影同步

提出了一种实现加权网络时空混沌投影同步的方法.通过构造合适的Lyapunov函数,确定了加权网络中连接节点之间耦合函数的结构以及网络节点状态方程中分离配置的线性项的系数矩阵的取值范围.以Bragg声光双稳系统作为局域函数,单向耦合映像格子作为空间扩展系统构成激光时空混沌模型.通过仿真模拟检验了采用激光时空混沌模型作为网络节点的加权网络的投影同步效果.结果显示,对于任意的节点之间耦合强度的权重值,加权网络的投影同步均可以实现. 关键词： 投影同步 加权网络 时空混沌 Bragg声光双稳系统     

节点结构互异的复杂网络的时空混沌反同步

研究了节点结构互异的离散型时空混沌系统构成复杂网络的反同步问题. 通过构造合适的Lyapunov函数, 确定了复杂网络中连接节点之间的耦合函数的结构以及控制增益的取值范围. 以物理中具有时空混沌行为的激光相位共轭波空间扩展系统、Gibbs电光时空混沌 模型、Bragg声光时空混沌模型以及一维对流方程的离散形式作为 网络节点构成的复杂网络为例进行了仿真模拟, 发现整个网络存在稳定的混沌反同步现象.     

电流控制二次型Boost变换器的动力学研究

通过对二次型Boost变换器开关状态的完整描述, 推导了两个电感电流边界, 建立了电流控制二次型Boost变换器的分段光滑迭代映射模型. 对比分析了以输入电感电流或储能电感电流作为电流反馈量的非线性分岔行为. 通过稳定性和工作模式分析, 得到了电流控制二次型Boost变换器从稳定的周期1工作状态到次谐波振荡状 态转移以及从电感电流不连续导电模式到连续导电模式转移的条件, 并采用参数空间映射图, 对二次型Boost变换器的工作状态域进行了估计. 设计了实验电路, 由实验结果证明了不同的参数变化有着不同的分岔路由, 存在工作模式转移现象, 电流控制二次型Boost 变换器呈现复杂的动力学行为, 实验结果验证了理论分析的正确性. 关键词： 二次型Boost变换器 开关状态 模式转移 边界碰撞分岔     

基于有向渗流理论的关联微博转发网络信息传播研究

微博网络的快速性、爆发性和时效性, 以及用户复杂的行为模式, 使得研究其信息传播模型及影响因素成为网络舆情的热点方向. 利用压缩映射定理, 分析不动点迭代过程的收敛条件, 得到有向网络信息传播过程的渗流阈值和巨出向分支的数值解法; 通过可变同配系数生成模型, 分析关联特征对信息传播的影响; 最后利用微博转发网络数据进行仿真对比实验. 结果表明: 虽然四类关联特征同时体现出同配、异配特征, 但信息传播结果更多受入度-入度相关性、入度-出度相关性影响; 通过删除少量节点的方法, 提取边同配比例, 验证大部分节点的四类关联特征呈现一致性.     

Spatio-temporal segmentation of moving objects using edge features in infrared videos

In this paper, a novel method is proposed for spatio-temporal segmentation of moving objects using edge features in infrared videos. We define motion saliency of edge (MSoE) to generate the MSoE-map. The seeds of moving objects are extracted from the MSoE-map by using Otsu's method and subsequently compensated by historical data. An improved layer-based region growing method is applied to the seeds to achieve spatial segmentation of moving objects. The region growing method has an adjustable growing threshold. So, one of the focuses of our work is how to determine the best growing threshold. A Markov Random Field (MRF) based criterion with maximum a posterior (MAP) estimation principle is proposed for performance evaluation of moving object segmentation without ground truth (GT) in infrared videos. This criterion can be considered as an object function of threshold determination during global searching. The global optimum is accomplished by using simulated annealing (SA) algorithm to obtain the best growing threshold. The final segmentation mask of moving objects is grown from the seeds with the best growing threshold. Experimental results are provided to illustrate that the proposed method has better performance for moving object segmentation with fewer effects of object-background misclassification in infrared videos.     

基于大功率LED的闪光成像速度测量系统

设计出一种非接触式高速运动物体速度测量系统,在CCD单帧曝光时间内利用闪光灯对飞行中的物体进行两次闪光照明,CCD对两个不同位置的飞行物体成像,设定两个光脉冲之间的时间间隔以及测量出两个物体像之间的距离即可算出其飞行速度.该系统能够进行实时测量,突破了空间单点测量技术的局限,可在同一幅图像上记录下不同时刻多个空间点目标的图像信息,以及同一时刻不同空间点目标的图像信息,从而可以获得目标运动的瞬时速度.     

基于大功率LED的闪光成像速度测量系统(英文)

设计出一种非接触式高速运动物体速度测量系统,在CCD单帧曝光时间内利用闪光灯对飞行中的物体进行两次闪光照明,CCD对两个不同位置的飞行物体成像,设定两个光脉冲之间的时间间隔以及测量出两个物体像之间的距离即可算出其飞行速度.该系统能够进行实时测量,突破了空间单点测量技术的局限,可在同一幅图像上记录下不同时刻多个空间点目标的图像信息,以及同一时刻不同空间点目标的图像信息,从而可以获得目标运动的瞬时速度.     

Passive ranging and a three-dimensional imaging system through wavefront coding

A single-image passive ranging and three-dimensional(3 D) imaging system with chiral phase encoding was proposed in 2011 [Opt. Lett. 36, 115(2011)]. A new theoretical analysis of the system in space domain is presented in this paper. We deduce the analytic relationships between the object distance and the point spread function, and between the object distance and the encoded image, respectively. Both the point spread function and the processed spectrum of the encoded image have two spots, which will rotate with the variation of the object distance. Then the depth map is extracted from the encoded image and it can be used to set up 3 D images. The theoretical analysis is verified by a wavefront coding system with a chiral phase which is generated by a phase-only liquid-crystal spatial light modulator. The phase generated by the liquid-crystal spatial light modulator is more flexible than the fixed phase mask and can be adjusted in real time. It is especially suitable for observing the object with a large depth of field.     

透过散射介质对直线运动目标的全光成像及追踪技术

光散射是限制光传输以及降低和破坏光学成像性能的主要因素,透过复杂散射介质对运动目标的全光成像是光学领域极具挑战性的技术之一.本文提出一种利用散斑差值自相关透过散射介质对运动目标进行实时追踪的方法.采用赝热光照明,基于光学记忆效应理论,通过对运动目标采集的两帧散斑做差值,然后做自相关运算,计算目标移动的距离,实现对目标的实时追踪,并且利用相位恢复算法进行简单处理就可以重建隐藏目标.对该方法进行了实验验证,成功地对隐藏的运动目标实现了成像与追踪.这种透过散射介质对运动目标的全光成像及实时追踪技术,在生物医学等领域具有重要应用潜力.     

On-line phase measuring profilometry based on phase matching

On-line phase measuring profilometry based on phase matching is proposed. While just one fixed sinusoidal fringe is projected on a measured object moving with the pipeline, deformed patterns modulated by the object moving at the same distance moment are captured synchronously by the CCD camera. The phase information of the object in those captured deformed patterns can be predicted using FTP method to assist the pixel matching so as to realize the point-to-point correspondence of the object in the captured deformed patterns. Meanwhile, the equivalent phase-shifting deformed patterns can be extracted. So the three dimensional shape of the object can be reconstructed successfully with an equal phase-shifting algorithm. Numerical simulation and experiments show feasibility and validity of the proposed method.     

Ultrasonic distance and velocity measurement using a pair of LPM signals for cross-correlation method: improvement of Doppler-shift compensation and examination of Doppler velocity estimation

Real-time distance measurement of a moving object with high accuracy and high resolution using an ultrasonic wave is difficult due to the influence of the Doppler effect or the limit of the calculation cost of signal processing. An over-sampling signal processing method using a pair of LPM signals has been proposed for ultrasonic distance and velocity measurement of moving objects with high accuracy and high resolution. The proposed method consists of cross correlation by single-bit signal processing, high-resolution Doppler velocity estimation with wide measurement range and low-calculation-cost Doppler-shift compensation. The over-sampling cross-correlation function is obtained from cross correlation by single-bit signal processing with low calculation cost. The Doppler velocity and distance of the object are determined from the peak interval and peak form in the cross-correlation function by the proposed method of Doppler velocity estimation and Doppler-shift compensation. In this paper, the proposed method of Doppler-shift compensation is improved. Accuracy of the determined distance was improved from approximately within ±140 μm in the previous method to approximately within ±10 μm in computer simulations. Then, the proposed method of Doppler velocity estimation is evaluated. In computer simulations, accuracy of the determined Doppler velocity and distance were demonstrated within ±8.471 mm/s and ±13.87 μm. In experiments, Doppler velocities of the motorized stage could be determined within ±27.9 mm/s.     

Point-based integration for 3D object reconstruction from Ladar range images

A method of point-based integration from multiple registered laser radar (Ladar) range images for 3D object reconstruction is presented in this literature. Our method operates surface point data directly and provides a direct way for overlapping data removal. The overlapping areas were detected using a single distance threshold. In order to set the distance threshold, a local point density approach is introduced to solve it. Compared with the mesh-based, volumetric and other points-based integration approaches, our method is simple and fast, and need less storage memory. The approach is performed on various objects with different geometric shapes. The experimental results demonstrate the efficiency and feasibility of our method.     

The non-stationary freefield response for a moving load with a random amplitude

This paper presents a stochastic solution procedure for the calculation of the non-stationary freefield response due to a moving load with a random amplitude. In this case, a non-stationary autocorrelation function and a time-dependent spectral density are required to characterize the response at a fixed point in the freefield. The non-stationary solution is derived from the solution in the case of a moving load with a deterministic amplitude. It is shown how the deterministic solution can be calculated in an efficient way by means of integral transformation methods if the problem geometry exhibits a translational invariance in the direction of the moving load. A key ingredient is the transfer function between the source and the receiver that represents the fundamental response in the freefield due to an impulse load at a fixed location. The solution in the case of a moving load with a random amplitude is formulated in terms of the double forward Fourier transform of the non-stationary autocorrelation function. The solution procedure is illustrated with an example where the non-stationary autocorrelation function and the time-dependent standard deviation of the freefield response are computed for a moving harmonic load with a random phase shift. The results are compared with the response in the deterministic case.