Wdm Optical Multicasting

Relevant for computer systems and communications engineers as well as networking vendors. Explores the optical multicast wavelength division multiplexing (WDM) technology and the media access control (MAC) protocols that supports multicast traffic in local WDM networks. Presents a comprehensive and comparative study of different challenges faced in the design of multicast technique and schemes. Investigates design issues and considerations for flexible and scalable network architecture such as multicast wavelength assignment problem, transceivers tuning limitations, multicast scheduling algorithms, reconfiguration techniques, and survivability strategies. Provides a thorough analysis of the single-hop WDM optical network based on the broadcast-and-select with passive star coupler (PSC) system.

Ethernet is the dominant technology for local-area networks (LANs) due to its simplicity, flexibility, and cost efficiency. In order to carry Ethernet traffic over backbone networks of metropolitan-area networks (MANs) and wide-area networks (WANs), the current mode of operation requires the conversion of Ethernet signals to technologies used by the backbone networks using complex and expensive equipment. In order to avoid this sort of unnecessary conversion, efforts are now under way to extend the usage of Ethernet from LANs to carrier-grade Ethernet in backbone networks of different service providers, and to provide a unified platform for transporting traffic. With the advent of muticast applications such as Internet Protocol TV (IPTV) and Video-on-Demand, there is a need for techniques to efficiently route multicast traffic over Ethernet backbone networks. Here, we address the problem of Routing and Wavelength Assignment (RWA) of a set of multicast requests in a Multi-Line-Rate (MLR) Ethernet backbone network with the objective of minimizing the cost of setting up the network, in terms of the Service Provider's Capital Expenditure (CAPEX). We present an Auxiliary Graph-based heuristic algorithm that routes each multicast request on a light-tree structure, and assigns minimum-cost wavelengths along the route. The algorithm utilizes the flexibility extended by MLR wavelength-division-multiplexed (WDM) optical networks used to construct the Ethernet backbone, by re-using channels, and using appropriate weight-assignment schemes to make a cost-efficient choice of wavelengths. We compare the properties of the algorithm to the optimal solution given by a mathematical model formulated as an Integer Linear Program (ILP), and show that they compare very well for small networks. The algorithm is tested on bigger networks, and it is shown that the algorithm is most cost-effective when the incoming traffic requests are processed in descending order of their bandwidth requirements, while using the linear weight-assignment scheme with volume discount.

Many applications and telecommunications services in futurehigh-speed networks will require some form of multipointcommunication. The problems associated with providing network support formultipoint communication have been widely studied within anumber of different networking contexts. As current network technologies evolve to an all-optical, largely passive infrastructure, these problems takeon new significance and raise a number of challenging issuesthat require novel solutions. We consider the problem of supporting multipoint communication at the media access control (MAC) layer of broadcast-and-select Wavelength Division Multiplexed (WDM) networks. In this environment, bandwidth consumption and channel utilization arise as two conflicting objectives in the design of scheduling algorithms for multicast traffic. We present a new technique for the transmission of multicast packets which is based on the concept of a virtual receiver. This is a set of physical receivers which behave identically in terms of tuning. We focus on the problem of optimally selecting the virtual receivers, and prove that it is NP-complete. We then present four heuristics of varying degrees of complexity for obtaining virtual receivers that provide a good balance between the two conflicting objectives. The dynamic nature of multicast traffic could affect the balance obtained with the virtual receivers when the network conditions change. We study the sensitivity to changes of the virtual receiver sets and the cost associated with handling the changes. Also, the cost of three different approaches to handling the changes is analyzed. Finally, we study the performance of various strategies for scheduling a combined load of unicast and multicast traffic in a broadcast WDM network. Three different scheduling strategies are presented, namely: separate scheduling of unicast and multicast traffic, treating multicast traffic as a number of unicast messages, and treating unicast traffic as multicasts of size on.

In modern networks, the demand for bandwidth and high quality of service (QoS) requires the efficient utilisation of network resources such as transmitters, receivers and channel band~idth. One method for conserving these resources is to employ efficient implementations of multicasting wherever possible. Using multicasting, a source sending a message to multiple destinations may schedule a single transmission which can then be broadcasted to multiple destinations or forwarded from one destination to another, thus conserving the source transmitter usage and channel bandwidth. This thesis investigates the behaviour of single-hop WDM optical networks when they carry multicast traffic. Each station in the network has a fixed-wavelength transceiver and is set to operate on its own unique wavelength as a control channel. Each station also has a tuneable wavelength transceiver in order to transmit or receive signals to or from all the other stations. A transmission on each channel is broadcasted by a star coupler to all nodes. Multicasting in single-hop WDM networks has been studied with different protocols. This thesis studies the multicasting performance adopting receiver collision avoidance (RCA) protocol as a multicasting protocol. This study takes into consideration the effect of the tuneable transceiver tuning time which is the time required to switch from one wavelength to another, and the propagation time required by a packet to propagate from one node to another. The strategy in RCA protocol is that nodes request transmission time by sending a control packet at the head of their queues. Upon receipt of this information all nodes run a deterministic distributed algorithm to schedule the transmission of the multicast packet. With the control information, nodes determine the earliest time at which all the members of the multicast group can receive the packet and the earliest time at which it can be transmitted. If a node belongs to the multicast group addressed in the control packet, its receiver must become idle ~ntil all nodes in the group have tuned to the appropriate wavelength to receive the packet. This problem leads to poor transmission and consequently low channel utilisation. However, throughput degradation due to receiver conflicts decreases as the multicast size increases. This is because for a given number of channels, the likelihood of a receiver being idle decreases as the number of intended recipients per transmission increases. The number of wavelengths available in a WDM network continues to be a major constraint. Thus in order to support a large number of end users, such networks must use and reuse wavelengths efficiently. This thesis also examines the number of wavelengths needed to support multicasting in single-hop optical networks. This thesis shows that the system under study can accommodate large tuning delays and keeps with suitable throup) hput when the number of wavelength is equal to the number of nodes. When the number of wavelengths is comparable to the number of users the tuning time influence on the packet delay increases. This thesis also shows that when the size of multicast packet and receiver tuning delay increase the throughput decreases.