A General Neural Network Model for Estimating Telecommunications Network Reliability

This paper puts forth a new encoding method for using neural network models to estimate the reliability of telecommunications networks with identical link reliabilities. Neural estimation is computationally speedy, and can be used during network design optimization by an iterative algorithm such as tabu search, or simulated annealing. Two significant drawbacks of previous approaches to using neural networks to model system reliability are the long vector length of the inputs required to represent the network link architecture, and the specificity of the neural network model to a certain system size. Our encoding method overcomes both of these drawbacks with a compact, general set of inputs that adequately describe the likely network reliability. We computationally demonstrate both the precision of the neural network estimate of reliability, and the ability of the neural network model to generalize to a variety of network sizes, including application to three actual large scale communications networks.      

New generation network architecture and software design

This article deals with new telecommunications networks, the reasons for changes, and the options telecommunications are facing today. The new generation of network architecture is discussed, and two network evolution case studies are presented as well as the new role of powerful applications in the future development of networks. Special emphasis is on software research and development supporting this “new telecom world”     

Mis-Specification Analysis of Linear Degradation Models

Degradation models are widely used to assess the lifetime information of highly reliable products if there exists quality characteristics whose degradation over time can be related to reliability. The performance of a degradation model depends strongly on the appropriateness of the model describing a product's degradation path. In this paper, motivated by laser data, we propose a general linear degradation path in which the unit-to-unit variation of all test units can be considered simultaneously with the time-dependent structure in degradation paths. Based on the proposed degradation model, we first derive an implicit expression of a product's lifetime distribution, and its corresponding mean-time-to-failure (MTTF). By using the profile likelihood approach, maximum likelihood estimation of parameters, a product's MTTF, and their confidence intervals can be obtained easily. In addition, laser degradation data are used to illustrate the proposed procedure. Furthermore, we also address the effects of model mis-specification on the prediction of the product's MTTF. It shows that the effect of the model mis-specification on the predictions of a product's MTTF is not critical under the case of large samples. However, when the sample size and the termination time are not large enough, a simulation study shows that these effects are not negligible.     

A comparative study of pattern recognition techniques for qualityevaluation of telecommunications software

The extreme risks of software faults in the telecommunications environment justify the costs of data collection and modeling of software quality. Software quality models based on data drawn from past projects can identify key risk or problem areas in current similar development efforts. Once these problem areas are identified, the project management team can take actions to reduce the risks. Studies of several telecommunications systems have found that only 4-6% of the system modules were complex [LeGall et al. 1990]. Since complex modules are likely to contain a large proportion of a system's faults, the approach of focusing resources on high-risk modules seems especially relevant to telecommunications software development efforts. A number of researchers have recognized this, and have applied modeling techniques to isolate fault-prone or high-risk program modules. A classification model based upon discriminant analytic techniques has shown promise in performing this task. The authors introduce a neural network classification model for identifying high-risk program modules, and compare the quality of this model with that of a discriminant classification model fitted with the same data. They find that the neural network techniques provide a better management tool in software engineering environments. These techniques are simpler, produce more accurate models, and are easier to use     

Evaluating Reliability of Telecommunications Networks Using Traffic Path Information

We propose a reliability model for representing telecommunications networks that does not focus on topological information, but rather traffic path information. Mapping from traffic paths to physical elements and capacities enables the model to express simply how terrible performance degradations occur. Existing models, such as probability graph models, and probability-capacity graph models, do not adequately address actual telecommunication network designs. The probability graph model never considers performance degradations, while the probability-capacity model unreasonably assumes that we can estimate performance degradations from only the network topology. This paper also proposes an algorithm for evaluating the reliability of our new model. A numerical example shows that the algorithm is reasonably efficient for even large telecommunications networks.     

Signaling alternatives in a wireless ATM network

The world of wireless telecommunications is rapidly changing. The capabilities of wireless networks are improving at a steady pace. This paper presents two possible protocols for implementing mobility for wireless users in an asynchronous transfer mode (ATM) network. The vision of the authors is of one “wireless ATM telecommunications network” that is capable of supporting a variety of today's applications with room to grow for advanced applications of the future. We first visit database architectures that can support mobility in a wireless ATM network. We then discuss one of two signaling architecture alternatives, the “overlay signaling”, for overlay support of mobile users in the ATM-based wireless telecommunications network. “Overlay signaling” aims at minimizing the modification needed to the existing ATM protocols. We then describe a native “migratory signaling” approach that further integrates wireless and wireline users into one global wireless ATM network at the expense of requiring some modifications to the existing ATM protocols. A performance analysis of the proposed signaling architecture alternatives is also presented. We conclude by pointing out some challenges in merging ATM with wireless telecommunications     

提高通信网络可靠性的方法探析

从产品的可靠性定义出发,通过对通信网络可靠性的分析得出了影响通信网络可靠性的因素。从提高节点／设备的可靠性、优化拓扑结构、使用网络保护与恢复技术、使用差错控制技术等几个方面详细介绍了提高通信网络可靠性的方法和需要注意的问题,并简要说明了其它的方法。最后,指出这些方法已经运用于工程实践并被证明切实有效。     

Software Reliability Growth Models with Testing-Effort

Many software reliability growth models have been proposed in the past decade. Those models tacitly assume that testing-effort expenditures are constant throughout software testing. This paper develops realistic software reliability growth models incorporating the effect of testing-effort. The software error detection phenomenon in software testing is modeled by a nonhomogeneous Poisson process. The software reliability assessment measures and the estimation methods of parameters are investigated. Testing-effort expenditures are described by exponential and Rayleigh curves. Least-squares estimators and maximum likelihood estimators are used for the reliability growth parameters. The software reliability data analyses use actual data. The software reliability growth models with testing-effort can consider the relationship between the software reliability growth and the effect of testing-effort. Thus, the proposed models will enable us to evaluate software reliability more realistically.     

Cost impact of telecommunications switch software problems ondivested Bell operating companies

Bellcore has developed a model to support the periodic calculation and reporting of customer costs resulting from poor quality in telecommunications switching system software. This model is designed to provide, as a primary output, total switching system software cost of poor quality, along with a breakdown of this total by major cost component. The primary input to the model is software RQMS data, described in Bellcore's reliability and qualify measurements for telecommunications systems, TR-TSY-000929, and reported by switching system suppliers to the divested Bell operating companies (telcos) on systems operated by these companies. Telco and switching system supplier alike are expected to benefit from the ongoing application of the model, through which accumulated maintenance cost information will be made available to support telco switching system management functions and supplier quality improvement efforts     

Managing a broadband environment: “you can't buck themarket”

Demand for bandwidth seems unstoppable, and investment in network infrastructure is massive. There is a desire that network operators be able to manage these networks in an efficient manner, with flexibility to support new customer service opportunities faster and cheaper. Network operators are looking to standards to help them meet this challenge. The telecommunications management network (TMN) model, so effective in describing how management information must support needs at the element, network, and service levels, provides a good starting point. However, if the supporting standards are not developed and used wisely, a TMN approach can have the opposite effect of adding cost and removing flexibility. Technology-specific “stovepipes” may make it impossible to manage services across a range of technologies. The authors express their views on the need for a service-driven approach to standards implementation. They point out specific challenges operators will face as they approach their management systems' needs for broadband services, and encourage service providers and network operators to take a more proactive role in defining their needs     

Advanced Multi-Objective Evolutionary Algorithms Applied to Two Problems in Telecommunications

Most research in optimisation is aimed at single objective problems, where the aim is to find a solution which maximises or minimises a single quality measure. However, as in nature, many problems in telecommunications are fundamentally multi-objective, particularly where the issues involved are related to quality of service, or cost/reliability trade-offs. There has been considerable research in multi-objective optimisation, but, until recently, the most prominently known multi-objective optimisation algorithms have tended to be rather slow, and there has been no universally accepted way to properly compare the performance of different methods. Here, we describe two evolutionary computation-based multi-objective optimisation methods which have recently been shown both to be considerably faster than the classical set of such methods, and to outperform existing methods on a wide range of test problems. We focus on two application areas in telecommunications — the adaptive distributed database management problem, and the offline-routeing problem. The speed and quality of these new methods suggest that their adoption in live applications of these and other telecommunications-related problems is feasible.     

Improved survivability with multi-layer dynamic routing

The article deals with the integrated design and analysis of a telecommunications network to achieve specified levels of survivability under various failure modes and conditions. The main goal of the article is: given a criterion for network survivability, develop a unified theory for network planning at different “layers” of the network, and for coordinated restoration methods after a failure. While most of the discussion relates to modern circuit-switching architectures, for completeness, asynchronous transfer mode technology is also considered     

A synchronous digital hierarchy network management system

The critical issues in multivendor environments and operation systems of telecommunication networks essential to speeding up the handling of service orders and service restoration after failures are discussed. To overcome these problems as the software backlog grows, the authors propose the application of a telecommunications management network (TMN) architecture together with an object-oriented network resource (ONR) model. The second phase of the synchronous digital hierarchy (SDH) network management system, which is based on the TMN architecture and ONR model, is examined     

基于决策树的软件可靠性模型选择

软件可靠性建模是软件可靠性评估的主要方法之一。现在还没有一个可适用于所有软件项目的通用模型,所以可靠性模型的选择已成为一个重要的研究方向。决策树是数据挖掘的一种算法。文中首先介绍数据挖掘与软件可靠性模型选择的结合应用概念,然后重点分析决策树的生成算法。最后以一组数据仿真决策树生成过程,并验证此方法的可行性和准确性。     

Using distributed object technologies in telecommunications networkmanagement

The Telecommunications Management Network (TMN) has been developed as the framework to support administrations in managing telecommunications networks. It suggests the use of OSI Systems Management (OSI-SM) as the technology for management information exchanges. Distributed object technologies, such as the Common Object Request Broker architecture (CORBA), address the use of software application program interfaces (APIs) in addition to interoperable protocols. Their use in TMN has been the subject of intensive research in previous years, with most approaches focusing on interoperability aspects with OSI-SM. We examine the issues behind using distributed object technologies in TMN via a native fashion, with network elements supporting distributed objects directly, e.g., a “CORBA to the switch” approach. The proposed solution tries to maintain the full OSI-SM expressive power in a way that other solutions have not attempted before. Performance and scalability issues are considered, while the approach has been validated through implementation     

网管系统可靠性指标分配方法研究

为了研究网管系统可靠性指标分配,介绍了几种系统可靠性基本模型、特点和使用的系统,提出了可靠性指标分配的方法和原则,并进行了案例研究。针对特定通信网的综合网络管理系统功能,构建了该网络管理系统的可靠性模型,并依据研制合同中规定的可靠性指标要求进行可靠性指标分配,给出了满足研制总要求的可靠性指标分配结果,解决了网管系统可靠性指标分配关键技术。     

Nonhomogeneous Poisson process software-debugging models withlinear dependence

Past research in software reliability concentrated on reliability growth of one-version software. This work proposes models for describing the dependency of N-version software. The models are illustrated via a logarithmic Poisson execution-time model by postulating assumptions of dependency among nonhomogeneous Poisson processes (NHPPs) of debugging behavior. Two-version, three-version, and general N-version models are proposed. The redundancy techniques discussed serve as a basis for fault-tolerant software design. The system reliability, related performance measures, and parameter estimation of model parameters when N=2 are presented. Based on the assumption of linear dependency among the NHPPs, two types of models are developed. The analytical models are useful primarily in estimating and monitoring software reliability of fault-tolerant software. Without considering dependency of failures, the estimation of reliability would not be conservative     

Visionen zum Entwurf universeller Kommunikationsnetze von morgen

The design of conventional telecommunications is based on high reliability, availability, security and, most important scarcity, of resources. About 90% of switch software and hardware is needed to comply with these requirements. However, are these assumptions still valid in a world where one fibre can deliver all the traffic in the Austrian switching network, where processing power and storage capacity double every one to one and a half years and where Wi-Fi networks promise ubiquitous wireless broadband access? Internet is an option, but it is not designed for the particularities of voice traffic. This article revisits the underlying design assumptions for today’s telecommunications network and tries to answer the question: How would you design a telecom network if bandwidth, processing power, storage capacity and frequencies were abundant?