Design of an optimal BER monitoring scheme for preventive maintenance of an ATM network

In today's telecommunication networks it is well known that unexpected failures in the physical links may cause significant loss or interruption of services. It is desirable if the deterioration of links can be recognized early and maintenance is carried out before failure occurs. Asynchronous transfer mode (ATM) networks have recently received much attention and are considered to be the prime universal networks of the future. Unfortunately, there has been very little research in the areas of preventive maintenance and performance monitoring of the ATM network. In this paper we propose a preventive maintenance monitoring scheme for an ATM network where a performance index called the bit error rate (BER) index of each service of the ATM network is continuously monitored. The BER index of each path in the ATM network is assumed to be slowly deteriorating and is modelled as a Brownian motion process with positive drift. If any of the BER indices exceeds its action limit or threshold, a repair process is initiated. The optimal action limit is derived by minimizing the modified long-run average maintenance cost. We found that the optimal action limit depends strongly on the multiservice characteristics of the ATM network and on the network topology, which relates to the virtual path (VP) bandwidth management and routing protocol used. To obtain the optimal action limit, an optimization model for VP bandwidth management is set up and the multidimensional Erlang loss model is used to determine the path selecting probabilities. Some numerical examples are given to illustrate the proposed model and the optimization results. © 1998 John Wiley & Sons, Ltd.     

Blind deconvolution of symmetric noncausal impulse responses usingtwo-sided linear prediction

Considers the problem of estimating the time-symmetric, noncausal impulse response of a linear time-invariant system from measurements of the response of the system to an unknown input signal, which is assumed to be a realization of a white random process. The symmetric impulse response is modeled by a two-sided AR or ARMA system model. The two-sided AR coefficients are estimated using a two-step procedure. First, an estimate of an unconstrained parameter vector is computed by solving a close-to-Toeplitz-plus-Hankel system of equations using previously developed fast algorithms. Then, the polynomial square root of the result is obtained by solving a constrained least-squares problem which has a simple solution. Unlike previous methods, this approach requires no iterative procedure. However, it may lead to an unstable model in some extreme cases. Simulation results illustrate the performance of the proposed methods     

Optimal Synthesis of Distributed Parameter Systems

A technique is developed for the synthesis and design of a distributed parameter system guiding waves from one point in space to another. The parameter distributions are assumed to be nonrestricted except for the upper and lower bounds resulting from the imposition of physical realizability. The problem is similar to the "sensitivity" problem encountered in the optimal control of the systems. An improved version of the first-order gradient technique is used to obtain the optimal distributions of the parameters. The first-order gradient technique is sensitive to the form of the arbitrary distributions assumed at the start of the iterations. This technique has serious convergence problems associated with it. The problem is particularly severe and is encountered in "singular" optimal control problems. The algorithm devised here improves the first-order gradient technique so that it becomes less sensitive to the initial assumed distributions and virtually eliminates the convergence problems generated because of the bounds on the parameter distributions. A transmission line with distributed series r, l and shunt c is a particular case of the distributed parameter system. The optimal design of a notched filter employing a thin film circuit is a successful example of the application of the improved gradient technique. These distributions have been obtained by the use of a hybrid computer.     

Incremental output resistance of a Wilson-current mirror

The use of the 'Early' voltage parameter facilitates the derivation of simple limit expressions for the low-frequency incremental output resistance of a Wilson-current source in which the constituent bipolar devices are assumed to have characteristics that are not necessarily identical.     

On the Performance of a Two-User MIMO Downlink System in Heavy Traffic

A multiple-input multiple-output (MIMO) downlink system in which data is transmitted to two users over a common wireless channel is considered. The channel is assumed to be fixed for all transmissions over the period of interest and the ratio of anticipated average arrival rates for the two users, also known as the relative traffic rate, is the system design parameter. A packet-based traffic model is considered where data for each user is queued at the transmit end. A queueing analog for this system leads to a coupled queueing system for which a simple policy is known to be throughput-optimal under Markovian assumptions. Since an exact expression for the performance is not available, as a measure of performance (in heavy traffic), a diffusion approximation is established. This diffusion process is a two-dimensional (2-D) Semimartingale Reflecting Brownian Motion (SRBM) living in the positive quadrant of 2-D space     

A model of reservation systems

The author considers the performance evaluation of an advanced capacity reservation system using the CCITT defined reserved call establishment mode. A simple model for the traffic process is introduced: calls are assumed to be offered uniformly to the reservation book. The state of the system can then be described statistically in terms of the population of various types of islands of free time slots. The limit where the length of the reservation book tends to infinity is, in particular, considered. For single-server systems with a deterministic holding time of one or two time slots, a closed-form solution is found. For a general holding time distribution, an asymptotic analytical solution is derived and corrections to this asymptotic solution are calculated numerically. For multiserver systems with geometric holding time distribution, the author presents numerical results obtained by simulations. An interpolation formula valid for all traffic intensities is developed. A procedure is given to extrapolate the results to larger systems than were actually simulated     

The implementation of the power law model in silicon oxidation process simulation

The purpose of this work is to extend the power-law model¹ to a semi-empirical oxidation model for process simulation. The proposed model benefits from a very simple model formulation, when compared to an existing model there are fewer parameter variables and simpler expressions of parameter dependence for a wide range of oxidation conditions while maintaining a very good overall fitting accuracy. The model covers dry oxidation for different pressures, temperatures, orientations, oxidation in a HC1/O₂ ambient, doping enhanced oxidation phenomena, as well as wet oxidation for different pressure, orientation and temperature conditions. To develop the model, a strategy for parameter extraction,² combined with optimal adaptation is proposed. This approach includes two steps. One is the determination of the parameter correlation matrix and the e-indifference range for each oxidation condition(e.g. temperature). The second step is the parameter adaptation within the ε-indifference range to obtain a simple parameter variation over a wide range of oxidation conditions. As a consequence of this work, the proposed model is a possible replacement for the oxidation model of SUPREM-3³ in process simulation. The optimized parameter extraction strategy can be applied to a large class of modelling problems.     

An optimal maintenance model using a number of different actions

In this paper, we study an optimal maintenance model. The state of a system is determined by the distribution of its operating time. Whenever the system fails, a number of actions can be chosen, including the repair actions, the replacement actions and the action of discarding the system. The objective of this paper is to determine an optimal policy which maximizes the expected total discounted reward. By using the semi-Markov decision process approach, the method of successive approximations is suggested for determining the optimal reward function and the corresponding optimal policy.     

A Nonstationary Model for the Electromyogram

A theoretical model of the electromyographic (EMG) signal has been developed. In the model, the neural pulse train inputs were considered to be point processes which passed through linear, time-invariant systems that represented the respective motor unit action potential. The outputs were then summed to produce the EMG. It was assumed, that in the production of muscle force, the controlled parameter was the number of active motor units, n(t). The model then showed that the EMG can be represented as an amplitude modulation process of the form EMG = [Kn(t)1/2 w(t) with the stochastic process, w(t), having the spectral and probability characteristics of the EMG during a constant contraction. Various assumptions made in the model development have been verified by experiments.     

A Model for Studying Velocity Variations in Unmyelinated Axons

A simple electrical model of the nerve axon is presented. The only non-linear element in the model is the sodium conductance, which is assumed to vary in a step-like way. This allows the propagation velocity for action potentials to be expressed analytically. The aftereffects from an action potential are generally slow. By considering them quasi-stationary, the velocity of a subsequent action potential can be calculated and compared with that of the first one. Considering the effects of afterpotential and afterconductance, it is shown that the theory adequately describes the variations in propagation velocity in the squid giant axon for action potentials separated by 3 ms to 14 ms.     

一种模糊规则的确定方法

在模糊神经网络结构的设计中，对模糊规则数目的获取，传统方法是凭经验来确定，这种方法复杂且难度大。本文提出通过模糊数据曲线来确定模糊规则，这样网络结构的建立只需要根据系统的输入输出数据来获得。并将其用于共面波导CPW(Coplanar Waveguide)的参数提取，试验数据表明该方法具有学习速度快，建模精度高等特点。     

Optimal identification of regression-type processes underadaptively controlled observations

The paper presents a new approach for the design and analysis of adaptive systems for the optimal identification of physical processes and signals described by linear regression-type equations. Contrary to the traditional methods, a compound model of the observed process is proposed. This model describes an unobservable process that is subject to identification and the observing device (sensor) separately. The introduced adaptive model of the sensor with bounded linear range of its characteristic is more general and adequate than the commonly used ones. It is shown that optimal adaptive control of the sensor parameters and its fit to the statistics of the identified process significantly improve the accuracy of the parameter estimates and increase their convergence rate. Results of the theoretical part of the paper are illustrated by a simple analytic example and confirmed via simulation     

相关竞争失效场合雷达功率放大系统可靠性评估

针对相关竞争失效场合难以获取高可靠部件的性能分布信息,无法对系统可靠性进行准确估计的问题.提出了相关竞争失效场合下考虑认知不确定性的多态系统可靠性评估方法.该方法首先通过假定部件突发失效阈值为递减型随机过程来表征累积退化与突发失效的相关性,同时为降低对部件认知不确定性的影响,假定冲击引起的部件性能损伤分布参数和突发失效参数均为区间变量,建立基于区间变量的部件性能分布模型;而后对传统的通用生成函数方法进行改进,给出了区间通用生成函数的定义及其运算法则;最后对某型雷达功率放大系统的可靠性进行分析.该方法不仅克服了部件的失效模式复杂、状态信息少的不足,且方法简单、思路清晰,具有很强的通用性和工程应用价值.     

Signal Design for Low-Error Probability in Fading Dispersive Channels

The signal design problem for FSK communication via fading dispersive channels is considered. The channel is modeled as a linear filter whose time-varying impulse response is a sample function from a zero-mean Gaussian random field of arbitrary WSSUS type. The additive noise component in the received waveforms is supposed to be a zero-mean white Gaussian random process, and maximum likelihood demodulation is assumed. The signal design procedure here adopted consists of minimizing a known upper bound on the error probability, whereas the previous similar design method by Daly intended maximizing an upper bound on the detection probability for radar-astronomy targets. Though with slightly different optimal numerical values, here, as in Daly's problem, the signal design depends on a single parameter which is a simple functional of the channel timefrequency covariance function and of the signal envelope ambiguity function. A detailed example shows how the results of this concise paper can be used to optimize signal parameters and to predict the performance loss due to nonoptimal signal envelopes.     

A threshold criterion for phase-lock demodulation

An analytical threshold criterion has been developed for the general phase-lock receiver utilizing Booton's quasi-linearization technique. This criterion is established for arbitrary information and noise spectral densities. The information is assumed phase- or frequency-encoded on the received signal. Explicit results are centered around the case of additive white Gaussian noise. The principal nonlinearity is assumed to be the phase detector which is represented as a product device. Threshold curves are derived for three types of signals: 1) Bandlimited phase-encoded white Gaussian signals, optimal receiver; 2) Bandlimited phase-encoded white Gaussian signals, second-order receiver; 3) Frequency-encoded white signals, optimal receiver. The phase-encoded white Gaussian signal threshold is then compared with Shannon's results. It was found that the optimal receiver threshold occurs 10 log10(e) or 4.34 db above Shannon's limit. The second-order receiver was found to threshold 2 to 3 db above the optimal receiver in the region of normally encountered output signal-to-noise power ratios. Frequency-encoded white signals represent the character of residual noise in a communication link oscillator system. Residual frequency noise is induced by the ever present thermal noise in oscillator circuits. This particular thermal-induced noise cannot be removed entirely. For this fundamental noise process maximum receiver sensitivities are derived. An interesting result of quasi-linearization is that, for the signals considered, previous solutions of the Wiener-Hopf equation may be applied with only slight modifications.     

Intelligent optimal control of single-link flexible robot arm

This paper addresses the design and properties of an intelligent optimal control for a nonlinear flexible robot arm that is driven by a permanent-magnet synchronous servo motor. First, the dynamic model of a flexible robot arm system with a tip mass is introduced. When the tip mass of the flexible robot arm is a rigid body, not only bending vibration but also torsional vibration are occurred. In this paper, the vibration states of the nonlinear system are assumed to he unmeasurable, i.e., only the actuator position can be acquired to feed into a suitable control system for stabilizing the vibration states indirectly. Then, an intelligent optimal control system is proposed to control the motor-mechanism coupling system for periodic motion. In the intelligent optimal control system a fuzzy neural network controller is used to learn a nonlinear function in the optimal control law, and a robust controller is designed to compensate the approximation error. Moreover, a simple adaptive algorithm is proposed to adjust the uncertain bound in the robust controller avoiding the chattering phenomena. The control laws of the intelligent optimal control system are derived in the sense of optimal control technique and Lyapunov stability analysis, so that system-tracking stability can be guaranteed in the closed-loop system. In addition, numerical simulation and experimental results are given to verify the effectiveness of the proposed control scheme.     

基于半马尔科夫过程的变压器故障率预测及研究

设备的故障率曲线是制定维修策略的重要指标之一。为了合理估计变压器的故障率,本文提出了基于半马尔科夫过程的预测模型。模型中考虑了变压器的热老化过程和随机故障过程。热老化过程根据绝缘纸平均聚合度的取值范围划分为4个状态,状态间的转移时间假设服从威布尔分布,威布尔分布的尺度参数由经验公式获得;随机故障过程考虑了由雷击和短路冲击引起的2个随机故障状态,老化状态和随机故障状态之间的转移时间假设服从指数分布,转移率由故障统计数据获得。对模型中威布尔分布的形状参数进行适当取值后,可获得符合实际统计结果的变压器故障率曲线。     

A stochastic model of TCP/IP with stationary random losses

In this paper, we present a model for TCP/IP congestion control mechanism. The rate at which data is transmitted increases linearly in time until a packet loss is detected. At this point, the transmission rate is divided by a constant factor. Losses are generated by some exogenous random process which is assumed to be stationary ergodic. This allows us to account for any correlation and any distribution of inter-loss times. We obtain an explicit expression for the throughput of a TCP connection and bounds on the throughput when there is a limit on the window size. In addition, we study the effect of the Timeout mechanism on the throughput. A set of experiments is conducted over the real Internet and a comparison is provided with other models that make simple assumptions on the inter-loss time process. The comparison shows that our model approximates well the throughput of TCP for many distributions of inter-loss times.     

基于参数优化的类机器人运动规划研究

在类人机器人的运动过程中,各种不确定因数可能导致其发生摔倒.当摔倒不可避免的情况下,本文研究了类人机器人倒地动作的最优控制问题,通过对倒地动作深入分析,引入一种参数优化技术对倒地动作进行最优控制.针对参数优化方法对求解问题初值的敏感性和在求解二次规划问题(SQP)中的不足,使用初值状态筛选器和一种基于改进的二次规划滤子算法对该方法的寻优过程进行了改进,并和极小值原理控制方法进行了比较.最后仿真和实验显示了本文所提控制方法的有效性.     

A New Nonparametric Growth Model

This paper proposes a new nonparametric reliability growth model for the analysis of the failure rate of a system that is undergoing development test. The only restrictions on the actual, unknown failure distribution for each stage of testing is that it be continuous, have only one unknown parameter ?, and have an associated unimodal likelihood function. No assumptions regarding the parametric form of the failure rate of the development process are made, only that there is no decay in the reliability of the system during the design changes. The parameters are assumed to be ordered from one test stage to the next such that ?1 ? ?2 ? ... ? ?m. The new model performs very well based on relative error and mean square error. The model is generally superior to the popular AMSAA model, regardless of the actual underlying failure process. In addition, the results indicate a notable bias in the AMSAA model, early in the development process, regardless of the actual underlying failure process.