基于遗传算法的动态Bayesian网结构学习的研究

动态Bayesian网是复杂随机过程的图形表示形式,从数据中学习建造动态Bayesian网是目前的研究热点问题.本文针对该问题提出了一种遗传算法.文中设计了结合数学期望的适应度函数,该函数利用进化过程中的最好动态Bayesian网把不完备数据转换成完备数据,使动态Bayesian网的学习分解为两个Bayesian网(初始网和转换网)的学习,简化了学习的复杂度.此外,文中给出了网络结构的编码方案,设计了相应的遗传算子.模拟实验结果表明,该算法能有效地从不完备数据序列中学习动态Bayesian网,并且实验结果说明了隐藏变量的作用和遗传控制参数对结果模型的影响.     

无线电波传播中的随机射线及其应用

对建模无线电波传播的随机射线进行了详细的阐述,从随机网格信道引出随机射线的概念,介绍了随机射线的产生背景和分类,并给出一种产生随机射线的数学模型随机桥过程。使用最大熵原理得到发生后次反射随机射线的概率分布。最后介绍使用随机射线建模超宽带室内信道的多径传播轨,在此基础上可以得到信道特性建模的基本随机变量,以及无线电波传播的概率模型等应用。     

Performance Analysis of Millimeter Wave SWIPT System Using Ginibre Point Process

This paper deals with the coverage probability analysis of mmWave simultaneous wireless information and power transfer (SWIPT) system using stochastic geometry. mmWaves are higher frequency waves results in dense deployment of base stations (BS) leading to a higher data rate. Stochastic geometry provides a mathematical model to the randomly distributed nodes in the network. The Poisson point process (PPP) is the most analytically tractable and widely used model for the location of BSs, but it fails to capture the underlying separation between the BSs of cellular networks. Ginibre point process (GPP) model is used for the repulsion between the two nodes and provides a realistic model compared to PPP. In this paper, mmWave interfering base stations are considered to be GPP for deriving the coverage probability analysis. The proposed model considers the interfering base stations for analyzing the coverage probability. The closed form expression for the coverage probability is derived and is analyzed. The theoretical results are validated through simulation results.

     

Relationships between probabilistic Boolean networks and dynamic Bayesian networks as models of gene regulatory networks

A significant amount of attention has recently been focused on modeling of gene regulatory networks. Two frequently used large-scale modeling frameworks are Bayesian networks (BNs) and Boolean networks, the latter one being a special case of its recent stochastic extension, probabilistic Boolean networks (PBNs). PBN is a promising model class that generalizes the standard rule-based interactions of Boolean networks into the stochastic setting. Dynamic Bayesian networks (DBNs) is a general and versatile model class that is able to represent complex temporal stochastic processes and has also been proposed as a model for gene regulatory systems. In this paper, we concentrate on these two model classes and demonstrate that PBNs and a certain subclass of DBNs can represent the same joint probability distribution over their common variables. The major benefit of introducing the relationships between the models is that it opens up the possibility of applying the standard tools of DBNs to PBNs and vice versa. Hence, the standard learning tools of DBNs can be applied in the context of PBNs, and the inference methods give a natural way of handling the missing values in PBNs which are often present in gene expression measurements. Conversely, the tools for controlling the stationary behavior of the networks, tools for projecting networks onto sub-networks, and efficient learning schemes can be used for DBNs. In other words, the introduced relationships between the models extend the collection of analysis tools for both model classes.     

Performability modelling of wireless communication systems

The high expectations of performance and availability for wireless mobile systems has presented great challenges in the modelling and design of fault tolerant wireless systems. The proper modelling methodology to study the degradation of such systems is so‐called performability modelling. In this paper, we give overview of approaches for the construction and the solution of performability models for wireless cellular networks. First, we start with the Erlang loss model, in which hierarchical and composite Markov chains are constructed to obtain loss formulas for a system with channel failures. Consequently, we develop two level hierarchical models for the wireless cellular system with handoff and channel failures. Then, for a TDMA system consisting of base repeaters and a control channel, we build a hierarchical Markov chain model for automatic protection switching (APS). Finally, we discuss stochastic reward net (SRN) models for performability analysis of wireless systems. Copyright © 2003 John Wiley & Sons, Ltd.     

Network planning with random demand

We study a planning problem associated with networks for private line services. In these networks, demands are known to exhibit considerable variability, and as such, they should be treated as random variables. The proposed planning model is a two-stage stochastic linear program (SLP) with recourse. Due to the enormous size of the deterministic equivalent, we choose a sampling based algorithm calledstochastic decomposition (SD). For very large-scale SLPs, such as the ones solved in this application, SD provides an effective methodology. The model presented in this paper is validated by using a detailed simulation of the network. We report results with a network that has 86 demand pairs, 89 links and 706 potential routes.This work was supported in part by Grant No. NSF-DDM-9114352 from the National Science Foundation.     

基于泛函网络求解随机优化问题的EM算法

研究了一类带有随机变量的不确定优化问题。为了解决这类随机优化问题,首先用随机仿真为泛函网络产生训练样本,用泛函网络逼近随机函数的均值,将随机函数的均值转为不合随机变量的确定性函数,然后用类电磁算法求解确定性函数的最优解。最后,给出一些数值例子验证了所提算法的有效性。     

基于MIV和BRBP神经网络的电路板红外诊断方法

针对BP神经网络对于海量数据训练及多维数据训练收敛困难的问题,在使用增加动力项、自适应学习速率等方法的基础上,引入均值影响度算法(MIV)构造了贝叶斯正则化反向传播(BRBP)神经网络,以此提高电子线路板红外故障诊断算法的效率。利用红外测温方式,获取了不同室温及运行状态下电路板中21个元器件温度数据。将此21个参数作为故障诊断模型的初始输入变量,经过MIV算法简约为12个参数输入至BRBP神经网络,进行故障评估和诊断。结果表明:相对于传统的BRBP神经网络,本文设计的基于MIV和BRBP神经网络模型诊断方法极大简化了数据训练的数据量并解决了数据收敛的困难,因此效率更高,用时更省。     

Performance study of hybrid spectrum access scheme in millimeter wave cellular network

In millimeter wave cellular networks,the spectrum efficiency of multi-operator dual frequency hybrid spectrum access scheme under open access mode was studied.Firstly,Poisson point process was used to model the distribution of base stations of each operator,the base station and carrier frequency were selected by users based on maximum received power criterion and high frequency carrier signal to interference plus noise ratio (SINR) threshold access criterion respectively.Secondly,the rate coverage was used as performance criteria to study the spectrum efficiency,and the theoretical expression of rate coverage was derived based on stochastic geometry theory according to channel model,path loss model and antenna model.Finally,the influence of user density,base station density and antenna gain on rate coverage was analyzed by simulation.The experimental results show that the multi-operator dual frequency hybrid spectrum access scheme has better spectrum utilization efficiency.     

Stochastic Modeling of TCP in Networks with Abrupt Delay Variations

An analytical model of TCP (Transport Control Protocol) over an end-to-end path with random abrupt round-trip time (RTT) changes is presented. Modeling the RTT as a stochastic process, we analytically quantify and compare between the degree of degradation of the steady-state average throughput and window size due to spurious retransmissions for the different versions of TCP (Reno/NewReno versus Tahoe). The modeling methodology in this paper is used for evaluating different design alternatives for TCP for highly dynamic networks.     

Sleep and Wakeup Strategies in Solar-Powered Wireless Sensor/Mesh Networks: Performance Analysis and Optimization

A queuing analytical model is presented to investigate the performances of different sleep and wakeup strategies in a solar-powered wireless sensor/mesh network where a solar cell is used to charge the battery in a sensor/mesh node. While the solar radiation process (and, hence, the energy generation process in a solar cell) is modeled by a stochastic process (i.e., a Markov chain), a linear battery model with relaxation effect is used to model the battery capacity recovery process. Developed based on a multidimensional discrete-time Markov chain, the presented model is used to analyze the performances of different sleep and wakeup strategies in a sensor/mesh node. The packet dropping and packet blocking probabilities at a node are the major performance metrics. The numerical results obtained from the analytical model are validated by extensive simulations. In addition, using the queuing model, based on a game-theoretic formulation, we demonstrate how to obtain the optimal parameters for a particular sleep and wakeup strategy. In this case, we formulate a bargaining game by exploiting the trade-off between packet blocking and packet dropping probabilities due to the sleep and wakeup dynamics in a sensor/mesh node. The Nash solution is obtained for the equilibrium point of sleep and wakeup probabilities. The presented queuing model, along with the game-theoretic formulation, would be useful for the design and optimization of energy-efficient protocols for solar-powered wireless sensor/mesh networks under quality-of-service (QoS) constraints     

Detecting change in a time-series (Corresp.)

A method is presented which provides a criterion for detecting a change in the structure of a model generating a stochastic sequence. Models that can be represented by a sequence of predictive probability distributions are considered. The method is based on the transformation of the observed sequence{x_{n}}into a sequence of partial sums of the general innovations, computed for the sequence{-log f(x_{n}|x_{n-1},x_{n-2}, cdots ,x_{0})}. If no change occurs the transformed sequence behaves like a Wiener process, but its mean will exhibit a monotonic growth after the process changes. Based on the properties of this transformation, fixed sample size and sequential tests for the change are constructed. The technique is applied to test for a change in the mean vector in a sequence of (generally dependent) Gaussian random variables, a change of coefficients of an autoregressive process, and a change of distribution in a sequence of discrete independent identically distributed random variables.     

车载网运动模型的复杂网络解析方法

车载网快速移动的特点导致其结构具有强烈的复杂性,传统的网络研究方法无法对这种复杂性进行分析.针对此问题将复杂网络解析方法运用于车载网的研究之中,依据车载网组网过程建立车载网运动模型,通过复杂网络研究手段与运动模型相结合的方式对车载网的度分布、鲁棒性以及网络连接变化率进行分析与仿真实验.结果表明,所构建的运动模型能够反应车载网构网过程,复杂网络解析方法在车载网性质分析方面十分有效.     

Anti-Synchronization of Markovian Jumping Stochastic Chaotic Neural Networks with Mixed Time Delays

The anti-synchronization control is investigated for a class of uncertain stochastic chaotic neural networks with both Markovian jump parameters and mixed delays. The mixed delays consists of discrete and distributed time-varying delays. First, by combining the Lyapunov method and a generalized Halanay-type inequality for stochastic differential equations, a delay-dependent criterion is established to guarantee the state variables of the discussed stochastic chaotic neural networks to be globally exponential anti-synchronized. Next, by utilizing a novel lemma and the Jensen integral inequality, a delay-dependent criterion is proposed to achieve the globally stochastic robust anti-synchronization. With some parameters being fixed in advance, the proposed conditions are all expressed in terms of linear matrix inequalities, which can be solved numerically by employing the standard Matlab LMI toolbox package. Finally, two examples are proposed to demonstrate the effectiveness and usefulness of the proposed results.     

Performance evaluation on a double‐layered satellite network

How to evaluate the performance of satellite networks is a prerequisite to the construction of satellite networks, and is also one of challenges in the researches on satellite networks. In this paper, generalized stochastic Petri net (GSPN) models are presented to carry out the performance analysis of a double‐layered satellite network. Firstly, the GSPN model of a double‐layered satellite network is simplified by proper analysis. Then, two sets of experiments are conducted to analyse the performance of the satellite networks, and show that the double‐layered satellite network outperforms single‐layered ones on the heavy traffic load. Finally, the feasibility and effectiveness of the proposed approach is verified by simulation experiments. Copyright © 2005 John Wiley & Sons, Ltd.     

Traffic models for wireless communication networks

Introduces a deterministic fluid model and two stochastic traffic models for wireless networks. The setting is a highway with multiple entrances and exits. Vehicles are classified as calling or noncalling, depending upon whether or not they have calls in progress. The main interest is in the calling vehicles; but noncalling vehicles are important because they can become calling vehicles if they initiate (place or receive) a call. The deterministic model ignores the behavior of individual vehicles and treats them as a continuous fluid, whereas the stochastic traffic models consider the random behavior of each vehicle. However, all three models use the same two coupled partial differential equations (PDEs) or ordinary differential equations (ODEs) to describe the evolution of the system. The call density and call handoff rate (or their expected values in the stochastic models) are readily computable by solving these equations. Since no capacity constraints are imposed in the models, these computed quantities can be regarded as offered traffic loads. The models complement each other, because the fluid model can be extended to include additional features such as capacity constraints and the interdependence between velocity and vehicular density, while the stochastic traffic model can provide probability distributions. Numerical examples are presented to illustrate how the models can be used to investigate various aspects of time and space dynamics in wireless networks     

Topology Design of OXC-Switched WDM Networks

This paper attempts to find a new analytical framework to identify cost-effective topological architectures of optical cross-connect (OXC)-switched wavelength-division multiplexing (WDM) networks. To keep the analysis tractable, we first focus on regular networks and a deterministic uniform traffic model. Regular topologies with symmetries are good approximations of metropolitan area and local area networks but only a fair resemblance of wide area networks. We find that for a regular topology the minimum-hop distance, which is normally used for gauging the size of the network, is an important parameter in dimensioning switching resources. By setting up a first-order cost model and evaluating the tradeoff between network resources, we obtain closed form solutions for the optimal node degree and network cost. For network design under stochastic traffic, we study the worst case capacity dimensioning and blocking probability among the set of all possible stochastic traffic distributions with the same mean and variance. The analytical approach presented in this paper helps us gain insights of parametric dependency of an optimal network architecture on key network design parameters.     

Performance and dependability analysis of fault-tolerant networks

Fault-tolerant networks continue their operation even in the presence of some failures. However, failures of components or subsystems can considerably affect the performance behaviour. This paper illustrates the use of Stochastic Reward Models (SRMs) for evaluating combined performance and dependability of fault-tolerant networks. A SRM is composed of a stochastic process, describing the evolution of the system, and a superimposed reward structure, reflecting different performance levels. To study the overall network behaviour, different network performance/dependability are introduced and discussed. Due to the arising complexity of exact SRMs a complexity reduction approach is proposed. For highly reliable networks an approximation of the model is presented.     

传输线的随机建模及瞬态响应数值实验分析

该文考虑集成电路制造过程中传输线制造工艺参数随机扰动对传输线传输性能的影响,建立了传输线的随机模型,针对该模型提出了基于蒙特卡洛法的传输线瞬态响应统计分析方法。统计分析中采用精细积分算法求解传输线样本瞬态响应,并对模型输出的正态性进行偏度-峰度检验,给出了最差情况估计。试验结果表明该文提出的传输线随机模型及统计分析方法可以对传输线的传输性能进行有效的评估,对于传输线制造过程的控制及优化有着重要意义。