Showing papers on "Network planning and design published in 1992"

A Simulated Annealing Approach to the Network Design Problem with Variational Inequality Constraints

Abstract: The equilibrium network design problem can be formulated as a mathematical program with variational inequality constraints. We know this problem is nonconvex; hence, it is difficult to solve for a globally optimal solution. In this paper we propose a simulated annealing algorithm for the equilibrium network design problem. We demonstrate the ability of this algorithm to determine a globally optimal solution for two different networks. One of these describes an actual city in the midwestern United States.

Computational results with a cutting plane algorithm for designing communication networks with low-connectivity constraints

Abstract: We describe a cutting plane approach to the problem of designing survivable fiber optic communication networks. This problem can be formulated as a minimum cost network design problem with certain low-connectivity constraints. Computational results on real-world telephone network design problems demonstrate the effectiveness of our cutting plane method. The facet-inducing inequalities for the convex hull of the solutions to this problem on which our algorithm is based are studied in detail in a companion paper.

Water Quality Monitoring Network Design: a Problem of Multi-Objective Decision Making

Abstract: Hydrologic data network design is a fairly complicated problem where questions as to the number of gages required, time frequencies to be selected, and benefits/costs of monitoring still remain unresolved. These issues are intensified in case of water quality variables as they are more error-prone, costly, and time consuming to sample. The basic difficulty underlying the design and evaluation of monitoring systems is the lack of an objective criterion to assess: (a) the efficiency, and (b) cost-effectiveness of a network. A statistical procedure based on the entropy principle of information theory is proposed to address the evaluation of both factors. Efficiency is measured quantitatively in terms of the information produced by a network. Similarly, benefits of monitoring are described by informative measures for an objective evaluation of cost-effectiveness. The study presented demonstrates the applicability of the entropy method in assessing the efficiency and the benefits of an existing water quality monitoring network. The method is applied for temporal and spatial features of monitoring, handled as both separate and combined problems. The results are shown in the case of the highly polluted Porsuk River in Turkey. The strengths and shortcomings of the proposed methodology are discussed, with recommendations for future research on the application of the entropy principle in network design.

Congestion tuning of telecommunications networks

Abstract: A system for optimizing the traffic carrying capacity of a telecommunications network having a plurality of elements and a plurality of routes connecting those elements to one another. The controlled congestion in the network is controlled by identifying and limiting defective devices and routes with exaggerated levels of congestion. Disturbance alarm levels are also adjusted to allow increased traffic through the network and the traffic within the network is reconfigured in real time in response to the occurrence of network events to optimize traffic capacity.

Power and ground network topology optimization for cell based VLSIs

Abstract: A new power and ground network design problem for cell-based VLSIs is discussed. In contrast to the conventional method, the network topology is optimized, or wiring resource consumption subject to electromigration and voltage drop constraints is minimized. The proposed method has been implemented. Using several examples, the validity of the problem formulation and the solution method was confirmed. Experimental results showed that brute power bus enhancement was meaningless and smart power and ground topologies significantly reduced the consumption of wiring resources. >

Facets for Polyhedra Arising in the Design of Communication Networks with Low-Connectivity Constraints

Abstract: This paper addresses the important practical problem of designing survivable fiber optic communication networks. This problem can be formulated as a minimum-cost network design problem with certain low-connectivity constraints. Previous work presented structural properties of optimal solutions and heuristic methods for obtaining “near-optimal” network designs. Some facet-inducing inequalities for the convex hull of the solutions to this problem are given. A companion paper describes computational results on real-world telephone network design problems with a cutting plane method based on this work. These computational results are summarized in the last section of this paper.

Transmission Networking: Sonet and the Synchronous Digital Hierarchy

Abstract: Telecommunications network evolution functional architecture of transport networks network node interface (NNI) based on the new synchronous digital hierarchy (SDH) synchronization performance monitoring and performance objectives management and control equipment specification and implementation network planning and deployment.

Solving nonlinear bilevel programming models of the equilibrium network design problem: a comparative review

Abstract: Nonlinear bilevel programming problems, of which the equilibrium network design problem is one, are extremely difficult to solve. Even if an optimum solution is obtained, there is no sure way of knowing whether the solution is the global optimum or not, due to the nonconvexity of the bilevel programming problem. This paper reviews and discusses recent developments in solution methodologies for nonlinear programming models of the equilibrium network design problem. In particular, it provides a primer for descent-type algorithms reported in the technical literature and proposes certain enhancements thereof.

Efficient implementation of heuristics for the continuous network design problem

Abstract: In this paper, we present an efficient implementation of heuristic procedures for solving the continuous network design problem where network users behave according to Wardrop's first principle of traffic equilibrium. Numerical results involving a “standard” benchmark problem are given. Also, it is shown that the cost mapping arising in the Iterative-Optimization-Assignment algorithm is integrable if and only if the volume-delay function is of either the BPR or some logarithmic form.

Construction of a real-world bilevel linear programming model of the highway network design problem

Abstract: The formulation of the highway network design problem (NDP) as a bilevel linear program (BLP) allows more realistic solutions taking into account the reaction of users to improvements made by the system. In this paper, a conceptual framework for the optimization of investments in inter-regional highway networks in developing countries is proposed. The model is applied to the Tunisian network using actual data. Considerable effort was expended to make the implementation as realistic as possible, taking into consideration travel times, operating costs, accident costs, improvement costs, conservation laws, and effects of intra-regional flows. A new formulation allowing the incorporation of any improvement cost functions, including non-convex and non-concave functions, is introduced.

Network Design Formulations in Schedule Synchronization

Abstract: This paper deals with problems in schedule synchronization for public transit networks. First we reformulate the problem of minimizing waiting times of passengers changing communication routes at certain transfer stations in a network. In this problem all transit lines only meet at certain points or transfer stations. A second problem, in addition, considers the case where different lines partly use the same tracks implicating that security distances have to be observed.

When Cycles Collapse: A General Approximation Technique for Constrained Two-Connectivity Problems

Abstract: We present a general approximation technique for a class of network design problems where we seek a network of minimum cost that satisfies certain communication requirements and is resilient to worst-case single-link failures Our algorithm runs in $O(n^2 \log n)$ time on a graph ith $n$ nodes and outputs a solution of cost at most thrice the optimum We extend our technique to obtain approximation algorithms for augmenting a given network so as to satisfy certain communication requirements and achieve resilience to single-link failures Our technique allows one to find nearly minimum-cost two-connected networks for a variety of connectivity requirements For example, our result generalizes earlier results on finding a minimum-cost two-connected subgraph of a given edge-weighted graph and an earlier result on finding a minimum-cost subgraph two-connecting a specified subset of the nodes Using our technique, we can also approximately solve for the first time a two-connected version of the generalized Steiner network problem and a two-connected version of the non-fixed point-to-point connection problem

Method and an arrangement for performance monitoring in a telecommunications network

Abstract: The invention relates to a method and an arrangement for terminal-oriented performance monitoring in a telecommunications network having a public side and a mobile subscriber side with terminals. Information with respect to network performance is collected by the subscriber terminals or in connection with the subscriber terminals and transferred to the public side. The information includes the grade of service, information on error conditions and information on the quality of the transmission channel in the network and non-connection information. With the aid of the invention the network operator is provided with a picture of the performance from the point of view of the customer and he can thus operate the network more effectively at the same time as the performance experienced by the customer is kept under control.

Ground water monitoring network design using multiple criteria decision making ant) geostatistics

Abstract: A multi-criteria approach to ground water quality monitoring network design is developed. The methodology combines multi-criteria decision making (MCDM) and modifications of geostatistical variance reduction analysis. Composite programming, a distance based optimization algorithm that employs a hierarchial structure, is used for the MCDM component of the design methodology. MCDM allows the consideration of numerous, often conflicting, design criteria. The methodology is useful for identifying the preferred combination of direct borehole and indirect geo-electric data. It also permits the use of prior information during initial stages of network development. Multi-variate kriging is employed to evaluate network performance using the combination of direct borehole data and indirect geoelectric data. Weighted measures of estimation variance are used as primary measures of performance, with the reduction in estimation variance being computed by the fictitious point method. Case study results demonstrate that the network design methodology can be used in both early and late phases of network development. It also leads to selection of the preferred combination and spatial orientation of direct and indirect data sources while considering cost-effectiveness and performance of alternative designs.

The OSI 95 Connection-Mode Transport Service: The Enhanced QoS

Abstract: During the last ten years, tremendous changes have taken place in the communication environments. First, there has been a continuous increase in network performance that has led, for instance, to increasingly high access data rates available in the lower layers. Furthermore, the changes in the offered network services have raised the issue of providing, at the transport level, services already provided at the subnetwork level, such as multicast or synchronous services for example. With the arrival of new applications, such as multimedia or client/server applications, a widening of the application requirements has also been observed. It is this evolving environment that has been at the origin of the ESPRIT II Project OSI 95. An important task in the framework of this project is the definition of an enhanced Transport Service taking account of the aforementioned evolutions. The enhanced Transport Service specified for OSI 95 consists of several types of service. The paper presents the connection-mode Transport Service. We focus mainly on the following original features of our connection-mode service: a new semantics for QoS parameters and the associated negotiation and re-negotiation.. Keyword Codes: C.2.1; C.2.2; C.2.m

Maximizing entropy defined reliability of water distribution networks

Abstract: A new measure is proposed for assessing reliability in water distribution networks. The measure used fundamental entropy theory and is based upon the consideration that redundancy and reliability of networks are highly inter-related. The measure is formulated in such a way that it is easily incorporated into optimization design models for water distribution networks. Optimization design models incorporating the measure are able to design both the layout and component sizes for a network. Evaluation, by the traditional network reliability measure, Nodal Pair Reliability, of network designs developed by network design models incorporating the measure showed the measure to be capable of identifying reliable networks. Furthermore, the changes in the value of the entropic redundancy measure with different network designs closely matched the changes in the Nodal Pair Reliability for those networks. As yet there is no comprehensive information on what a particular level of the entropy measure means in t...

Visualization in network topology optimization

Abstract: The intrinsic complexity and the geometric nature of network topology optimization lends itself to visual human-computer interaction. We have applied techniques of scientific visualization to all the aspects of network design, and have designed graphical user interface which takes full advantage of human expertise in the context of automated design processes.

Artificial intelligence-based system representation and search procedures for transit route network design

Abstract: An artificial intelligence (AI)-based representation of transportation networks is described. Such representation facilitates the development of efficient AI search algorithms that make up the bulk of the computational effort involved in the design and analysis of transportation networks. The novel representation is demonstrated, as are its advantages as implemented in the AI search algorithms developed for the design and analysis of a particular type of transportation network, namely, transit bus routes networks.

Genetic algorithms and fitness variance with an application to the automated design of artificial neural networks

Abstract: Existing genetic algorithm (GA) theory addresses how schema fitness serves as a measure of the expected increase or decrease of schema representation within the population. The work presented here considers how schema fitness variance affects schema representation through GA decision-making. It has long been known that the more significant bits of binary coded parameters converge before bits of lesser significance. This phenomenon, called domino convergence, is explored using the identity problem, f(x) = x. Sometimes convergence stops prematurely (a phenomenon called convergence stall), depending upon the relative magnitude of the mutation rate and the length of the encoding string. Analyses and models are presented exploring various aspects of these phenomena. GA convergence occurs in competition partitions. Each partition has an associated signal (measure of the force tending towards correct decision-making within that partition) and noise (measure of the force hindering correct decision-making). Which has the upper hand within a particular partition determines if the GA chooses correctly between competing schemata, which in turn determines convergence in the partition. Signal, noise, and the signal-to-noise ratio (SNR) are each defined in terms of fitness variance, with the SNR reconciling the conflicting effects of signal and noise. Formulas for the flat-population schema fitness variance, signal, noise, and SNR are derived using the Walsh basis. Both domino convergence and convergence stall are examined from the signal versus noise perspective. Designing an artificial neural network (ANN) for a specified problem can be difficult. Since the design of biological neural networks is a result of evolution, evolutionary search techniques may be well suited to network design. Back-propagation is known to generalize well on the contiguity problem (counting the number of clumps of 1s in a binary input field) when hidden layer receptive fields are narrow, but with high performance variance (noise) due to local minima. Evolutionary network design is used as a case study in applying GAs to a difficult, noisy problem. A program called GAND, genetic algorithms for network design, is described and tested on the contiguity problem. A number of techniques are presented that allow GAND, starting with randomly generated network interconnections, to evolve architectures rivaling the best produced by hand.

Spatial topology and dynamics in CDMA cellular radio

Abstract: The architecture and dynamics of a code-vision multiaccess (CDMA) cellular radio network are discussed. The network's layout, power control equation, cell coverage, signal quality, cell radio design, and heterogeneous cell clustering are described. It is shown that the CDMA system is adaptive to the channel and to the traffic load and offers flexibility in the network design and operation. >

A Review of Pipe Network Optimization Techniques

Abstract: In the water, gas and oil industries, capital investment in pipe networks is very high. Despite this, there has until recently been little effort to adopt formal optimization techniques in pipe network design.

Currency exchange rate forecasting by error backpropagation

Abstract: The paper describes a neural network system for forecasting time series and its application to a non-trivial task in forecasting currency exchange rates. The architecture consists of a two-layer backpropagation network with a fixed number of inputs modelling a window moving along the time series in fixed steps to capture the regularities in the underlying data. Several network configurations are described and the results are analysed. The effect of varying the window and step size is also discussed as are the effects of overtraining. The error backpropagation network was trained with currency exchange data for the period 1988-9 on hourly updates. The first 200 trading days were used as the training set and the following three months as the test set. The network is evaluated both for long term forecasting without feedback (i.e. only the forecast prices are used for the remaining trading days) and for short term forecasting with hourly feedback. By careful network design and analysis of the training set, the backpropagation learning procedure is an active way of forecasting time series. The network learns the training set near perfect and shows accurate prediction, making at least 20% profit on the last 60 trading days of 1989. >

Optical distribution channel: an 'almost-all' optical LAN based on the field-coding technique

Abstract: The authors introduce the design of the optical distribution channel, which serves as a local-area network and facilitates high-speed interconnection among heterogeneous machines in a way that is economically justifiable. In particular, they describe a design of an almost-all-optical local-area network that is capable of providing a gigabit per second transmission rate directly to the user. The network design is based on the 802.6 topology, a field coding technique in which the header and the data fields are encoded at different rates, and the principle of almost-all-optical switching. >

Planning for optimal expansion of leased line communication networks

Abstract: We report a new heuristic algorithm useful for developing least-cost expansion plans for leased telecommunications networks in which there is a hierarchy of possible transmission facilities. The problem combines network topology design and capacity expansion problems. Implemented on a PC, the algorithm has successfully developed topology growth plans for twenty periods on networks with up to 31 nodes.

NETPLAN-a telecommunications network planning support system

Abstract: A software system, NETPLAN, for supporting telecommunications network planning has been developed. It is implemented using Prolog with a user-friendly graphical interface based on X Windows and Motif. The three basic activities of network planning (routing design, one-way and both-way junction circuit design, and signaling network design for circuit-switched networks) are fully supported. The dimensioning of the junction circuits is optimized by considering noncoincidental busy hour traffic characteristics of the various routes using the 24-hour traffic profile. The object-oriented approach adopted by NETPLAN eliminates the need for defining traffic and routing matrices as required by conventional approaches. The usefulness of the system is illustrated by its application to the planning of a network with ten switching nodes and twenty interconnected links. >