Interconnection Of Local Area Networks Research Articles

QoS Enhancement of Wireless Campus Network Using Throughput Parameter

In recent years, broadband wireless systems developed themselves as one of the fastest growing and emerging sector in the field of telecommunications. It is a network of multiple interconnected local area networks in a limited geographical area. Quality of Service (QoS) is considered as one of the key issue in Wireless Campus Network due to the bottleneck and the competition among different kinds of traffic flow on the network. In this paper a Short-Term and Long-Term models are developed to enhancements QoS performance for Wireless Campus Network using throughput parameter. The models were designed and simulated using Optimized Network Engineering Tool, OPNET IT GURU Academic Edition with linkages between wireless workstations suitable for implementation to determine the performance of the network throughput. The two versions contain the same number of wireless workstations, but with some additional access points and multiple processors in the long-term model, a point-to - point protocol (PPP) was used to link the access points to the routers and the cloud while using 100BaseT cable to connect. In each case, the wireless network throughput was obtained. The findings obtained from these models showed that, relative to the short-term model, the Long-Term model achieves higher wireless network performance. That may be due to the fact that while in both models the total number of wireless workstations is the same, the long-term model topology includes additional access points and double processors resulting in the observed minimum throughput.

Practical data communications

From the Publisher: This book serves the practitioner. Its wide discussion of data network enhancements and design is built upon a thorough introduction of generic and specific data networks, focusing at length on transmission limitations, protocols and how they work, and questions of standards and related topics. The book covers local area networks (LANs) based on IEEE and ANSI standards, and wide area networks (WANs) built around TCP/IP, SNA, and OSI related protocols. It also covers VSAT (very small aperture terminal) satellite networks for data connectivity, and frame relay and SMDS for the interconnection of LANs over long distances. The reader will share the benefits of the author's extensive experience, detailing actual implementation methods for enterprise networks, considering particular corporate and government requirements, describing wiring concepts and specific models, and providing comparisons between different systems in terms of performance and complexity. Throughout, data communication is incorporated into the bigger picture of telecommunications, including the public switched digital network (PSDN), questions of high-capacity transmission media such as fiber optics, wire pair, and microwave and satellite communications. There is emphasis on the trend toward the integration of data, voice, and image into a single telecommunication system - popularly termed multimedia communications.

Wavelength assignment algorithms for wavelength routed interconnection of LANs

This paper addresses the wavelength assignment issues in interconnecting optical local area networks (LANs) in which a wavelength cannot be reused for local connections. Multiple LANs are connected using a wavelength routing bridge that may or may not support wavelength conversion. Static and dynamic approaches to partitioning of wavelengths for local and global traffic are compared using analysis and simulations. Under the static scenario, the wavelength set is partitioned into fixed sets of local wavelengths and global wavelengths. Dynamic wavelength assignment allows all traffic to use all the wavelengths and thus provides greater statistical multiplexing at the expense of increased implementation complexity. It is found that static wavelength assignment, the easiest algorithm to implement, is significantly outperformed by dynamic algorithms. Several dynamic assignment algorithms are developed, and architectural issues in interconnecting optical networks are discussed. The dependence of the call blocking performance on system parameters, such as the focal and global traffic statistics, and the number of available wavelengths is examined in detail. An elegant outcome is development of a simple, yet accurate, model to predict approximate blocking performance with an arbitrary number of LANs.

GSPN models of bridged LAN configurations

Communication systems comprising several local area networks (LANs) interconnected by bridges are modeled with Generalized Stochastic Petri Nets (GSPNs). GSPN models are developed at different levels of detail, progressively abstracting from the detailed behavior of real systems, and exploiting symmetries present in simple interconnected LAN topologies by “folding” the detailed GSPN models. The numerical results obtained from the solution of GSPN models are validated against results of very detailed simulation experiments, and found in good agreement, even for the most abstract GSPN models. Advantages and drawbacks of the GSPN modeling approach are discussed at the end of the paper.

Approximate bounds and expressions for the link utilization of shortest-path multicast network traffic

Multicast network traffic is information with one source node, but many destination nodes. Rather than setting up individual connections between the source node and each destination node, or broadcasting the information to the entire network, multicasting efficiently exploits link capacity by allowing the source node to transmit a small number of copies of the information to mutually-exclusive groups of destination nodes. Multicasting is an important topic in the fields of networking (video and audio conferencing, video on demand, local-area network interconnection) and computer architecture (cache coherency, multiprocessor message passing). In this paper, we derive approximate expressions for the minimum cost (in terms of link utilization) of shortest-path multicast traffic in arbitrary tree networks. Our results provide a theoretical best-case scenario for link utilization of multicast distribution in tree topologies overlaid onto arbitrary graphs. In real networks such as the Internet MBONE, multicast distribution paths are often tree-like, but contain some cycles for purposes of fault tolerance. We find that even for richly-connected graphs such as the shufflenet and the hypercube, our expression provides a good prediction of the cost (in terms of link utilization) of multicast communication. Thus, this theoretical result has two applications: (1) a lower bound on the link capacity required for multicasting in random tree topologies, and (2) an approximation of the cost of multicasting in regular LAN and MAN topologies.

Design of interconnection of local area networks

This paper addresses the interconnection of local area networks (LANs) using bridges. The interconnection problem is difficult to solve and feasible solutions are usually obtained through heuristics. The description of a simplified approach to solve the problem of interconnecting LANs is followed by the formulation of the problem and the development of a Lagrangean relaxation. A Tabu search metaheuristic for designing minimum cost spanning tree topologies is also presented. Computational results are given for the interconnection of up to ten local area networks.

Design of Interconnection of Local Area Networks

This paper addresses the interconnection of local area networks (LANs) using bridges. The interconnection problem is difficult to solve and feasible solutions are usually obtained through heuristics. The description of a simplified approach to solve the problem of interconnecting LANs is followed by the formulation of the problem and the development of a Lagrangean relaxation. A Tabu search metaheuristic for designing minimum cost spanning tree topologies is also presented. Computational results are given for the interconnection of up to ten local area networks.

Frame Relay to ATM interworking

Frame Relay is a well-established technology that has attracted a large number of customers world-wide. It is typically used to interconnect large, multi-site company local area networks (LANs). These data connections are usually at low to medium speeds. Asynchronous transfer mode (ATM) is a newer, broadband technology that is capable of very high speed LAN interconnection. In addition, it is arguably an ideal transport medium for legacy protocols such as Frame Relay.This paper describes the data transport features of each technology and discusses how ATM may be used to interconnect existing Frame Relay networks, and how an ATM user may exchange data with a Frame Relay user.

A congestion-avoidance scheme for LAN interconnection over high-speed networks

Organizations engaged in distributed computing have made a substantial investment in local area networks (LANs) (e.g., Ethernet, token rings and FDDIs). The number of connected servers and workstations, as well as their corresponding storage and service devices, are rapidly increasing. This leads to a substantial increase in LAN traffic, making current LAN interconnection technology ineffective. High-speed backbone networks may offer the suitable interconnecting media for LANs supporting traffic-intensive applications. However, the interconnection of dissimilar LANs and devices over a high-speed backbone network can create congestion at a destination LAN (output) gateway when more than one device or LAN attempt is made to communicate simultaneously with the same device or LAN. We address the issue of congestion avoidance in LANs interconnected by a high-speed backbone. A congestion-avoidance scheme that requires input gateways to secure permits to the output gateway before transmitting is proposed. An implementation of the proposed congestion-avoidance scheme in an asynchronous transfer mode switch is shown. Performance results indicate that packet dropping rates are greatly reduced.

Evolution of hospital communication systems towards multimedia applications

A study of the communication requirements and a detailed analysis of the network infrastructure necessary to efficiently support multimedia applications in a typical mid-range general hospital is presented. The study focuses on the consequences that the introduction of multimedia objects has on existing interconnected networks and has been based on the analysis of the performance of installed interconnected local area networks within a distributed medical database environment. Particular attention has been paid to the problems related to the introduction of multimedia technology in inter-hospital communication systems. The results of a number of tests undertaken in a real-life hospital environment on the transfer of objects simulating multimedia information in interconnected IEEE 802-3 networks are also presented and discussed. Various computational environment scenarios have been considered for this purpose. The operational parameters of the network have been measured in detail. The results obtained and ...

Network management for the NCIH

The North Carolina Information Highway (NCIH) will support many applications, including distance learning, LAN interconnection, video conferencing, law enforcement, and health care. Initially, the NCIH is supporting distance learning and local area network (LAN) interconnection applications for the North Carolina State Government. The operations plan for the NCIH was driven by end-user service requirements, network operator requirements, and the network architecture. The NCIH is a multi-vendor, multi-carrier ATM network offering a 155-Mb/s OC-3c SONET/ATM customer access interface to support many applications. BellSouth, along with Bellcore, Fujitsu, NTI, and other carriers, has developed and implemented an operations plan for managing the NCIH.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

High-speed bridges for fiber optic local area networks

Progress in optical fiber communications allows ever-increasing channel capacity for local area networks (LANs). However, there are still many problems to solve with respect to bridges interconnecting these LANs, which constitute a system bottleneck. To eliminate the bridge bottleneck in the interconnection of fiber optic LANs, a high-speed bridge using optical technology is proposed. The proposed optical bridge uses bistable optical devices, such as interference filters, as essential components of its design. As these devices can be easily fabricated, the implementation of this optical bridge is feasible.

Transparent (cut-through) bridging of CSMA/CD networks: performance analysis and implementation

The increasing popularity of local area networks (LANs) and the limitations of a single LAN (on its geographical coverage, information-carrying capacity, and number of stations) have triggered tremendous interest in the study and implementation of interconnected LANs. The authors address the problem of interconnecting the widely deployed CSMA/CD LANs (or LAN segments) via a transparent bridge. A transparent bridge is so called because it about the network structure and organization during its operation, i.e., it learns about the location of the network stations relative to the bridge in order to do the forwarding of frames selectively; also, the stations need not be aware of the presence of the bridge. The authors make two important contributions. First, they present an approximate analytical model of a transparent CSMA/CD bridge which can be either of the normal store-and-forward type or of the novel cut-through variety (introduced by Kwok and Mukherjee). Using the analytical model, which is also verified via simulation, the performance of the cut-through bridge is compared with (and found to be superior than) that of the normal bridge. Secondly they present the design of a cut-through bridge which can be easily implemented using off-the-shelf components. >

Performance analysis of interconnected LANs with server/client configuration

In this paper, we study the end-to-end performance of interconnected local area networks (LAN) with server/client configuration. The system uses bridges to connect two token-ring LANs through a high-speed communication link. A server station located on one LAN receives requests from client-stations on the same LAN as well as on the remote LAN, processes the requests, and returns responses to the client-stations. The end-to-end connections of the interconnected network are modelled as single-chain and multiple-chain closed queueing systems, which are solved by an iterative algorithm based on the MVA (mean value analysis) method. The performance examples are shown in terms of various system parameters such as the window size, server processing speed and internetwork transmission capacity, and are verified by computer simulations.

Fiber-optic interconnection of local area networks: physical limitations of topologies

This paper examines and quantifies the physical limitations imposed by basic topologies and access couplers in fiber-optic interconnection of local area networks (LAN's). To that end a detailed comparison of the basic topologies and access couplers is provided. The comparison is in terms of amount of fiber required, average propagation delay, fault tol- erance, maintainability, number of access couplers required, and maximum bit rate. These results lead to simple guidelines for the choice of topology and the type of access couplers that may be used in practical LAN interconnection. Our results indicate that for 100 node LAN interconnection, the star topology requires 3.8 times as much fiber as a dual ring, which may be of significant concern if the interconnection network is in an urban business area. The average propagation delay of a dual ring for a 100 node interconnection, on the other hand, is 3.3 times that of the star. This may be a significant disadvantage in LAN interconnection networks where the distance between nodes is large. The maximum bit rate, by assuming simple direct detection, is calculated for a fiber network based on the topology and type of access coupler. Our results show that the more fault tolerant an access coupler is, the lower is the maximum bit rate it can support.

Access methods for fiber-optic interconnection of LAN's

This paper examines the use of conventional access methods from the point of view of fiber-optic interconnection of local area networks (LAN's). It is shown that their characteristics depend on the number of nodes, packet length, and distance between nodes. A statistical analysis is carried out to compute access delay, delay variance, the number in the buffer, and analyze the effect of the interconnection network on the LAN's. Results show that, with the exception of polling, the network char- acteristics are not accurately predicted by the Poisson models. Compared to ring and bus topologies the star has better power budget characteristics, which are useful for WDM, but it is more difficult to implement access methods on. The addition of more wavelengths multiplies the throughput, but has little effect on the total delay.

A scheme for high-performance LAN interconnection across public MAN's

The proliferation of local area networks (LANs) and the emergence of high-bandwidth public networks will create an environment in which it will become increasingly simple and attractive to communicate among different organizations across a common public networking infrastructure. The open bridging scheme proposed, which employs inherent features of forthcoming cell-based metropolitan area networks (MANs) in order to achieve efficient interenterprise communications is introduced. It uses enhanced remote LAN bridging techniques to ensure router-level functionality with bridge-level simplicity and performance. The practicality of open bridging is shown by describing how it can be incorporated in networks using current standard protocol stacks, such as open systems interconnnection (OSI) and transmission control protocol/internet protocol (TCP/IP). >

A Lagrangian Relaxation Technique for Optimizing Interconnection of Local Area Networks

This paper addresses the problem of interconnecting a group of Local Area Networks (LANs) with bridges. The network designer would like to minimize the cost of connecting the LANs while maintaining acceptable traffic intensity levels on each of the LANs and the bridges. This problem belongs to the class of fixed charge network flow problems. A Lagrangian relaxation algorithm is proposed that incorporates two sets of constraints into the objective function. The relaxed problem has a special structure and can be solved as a minimum spanning tree problem and a set of shortest path problems. The subgradient algorithm is then used to maximize the Lagrangian dual. A Lagrangian heuristic algorithm is also proposed to find good primal feasible solutions. For problems with five LANs, the heuristic solutions are normally within 1% of the optimum. For larger problems the heuristic solutions are close to the lower bounds generated by the Lagrangian relaxation algorithm. This suggests that the quality of the heuristic solutions does not deteriorate as the dimension of the problem grows.

A model for interconnected LANs with general arrivals

We present an analytic model for the performance evaluation of interconnected local area networks. The system is decomposed into three elementary subnetworks—routers, Ethernets with 10 Mbps capacity, and an FDDI backbone with 100 Mbps capacity—each of which is modeled by a GI/G/1 queue. Routers are modeled as polling machines with the assumption that packets arriving at the router are not copied when they are moved from input queues to output queues. An approximate analysis based on the first two moments of interarrivai time and packet length distribution is applied to each subnetwork to obtain the performance. A variety of studies are conducted to evaluate those parameters affecting the performance of system. The optimum network structure with least waiting time could be found from our model by varying the number of routers connected to the FDDI and the number of Ethernets connected to routers.

Improved routing algorithm for interconnected local area networks

An improved version of the ISO intermediate system to intermediate system (IS–IS) routing algorithm is presented. Through simulation, it is shown that the modification provides a higher network power than the IS–IS algorithm. This improvement is achieved without a significant increase in computational complexity.