This paper looks at several aspects of optical layer protection techniques from an implementation perspective. We discuss the factors that affect the complexity of optical protection schemes, such as supporting mesh instead of ring protection, handling low-priority traffic, and dealing with multiple types of failures. The paper also looks at how the client layer interacts with the optical layer with respect to protection, in terms of how client connections are mapped into the optical layer, and how protection schemes in both layers can work together in efficient ways. Finally, we describe several interesting optical protection implementations, focusing on the ones that are different from conventional SONET-like implementations.

We provide network designs for optical add-drop wavelength-division-multiplexed (OADM) rings that minimize overall network cost, rather than just the number of wavelengths needed. The network cost includes the cost of the transceivers required at the nodes as well as the number of wavelengths. The transceiver cost includes the cost of terminating equipment as well as higher-layer electronic processing equipment, which in practice can dominate over the cost of the number of wavelengths in the network. The networks support dynamic (i.e., time-varying) traffic streams that are at lower rates (e.g., OC-3, 155 Mb/s) than the lightpath capacities (e.g., OC-48, 2.5 Gb/s). A simple OADM ring is the point-to-point ring, where traffic is transported on WDM links optically, but switched through nodes electronically. Although the network is efficient in using link bandwidth, it has high electronic and opto-electronic processing costs. Two OADM ring networks are given that have similar performance but are less expensive. Two other OADM ring networks are considered that are nonblocking, where one has a wide-sense nonblocking property and the other has a rearrangeably nonblocking property. All the networks are compared using the cost criteria of number of wavelengths and number of transceivers.

This paper proposes algorithms for allocating wavelengths to connections (lightpaths) in optical wavelength division multiplexed networks, predominantly for ring topologies. A worst-case model is considered, where no blocking of lightpaths is allowed, and there are no assumptions made on the traffic arrival and holding times. The traffic is characterized only by its load L, which is the maximum number of lightpaths that can be present on any link, assuming no blocking. A dynamic traffic model is considered where requests to set up lightpaths arrive over time and, must be accommodated without rerouting existing lightpaths, and lightpaths may be terminated over time as well. For networks without wavelength conversion, we show that at least 0.5Llog/sub 2/N wavelengths are required by any dynamic algorithm for rings of N nodes and present an algorithm that uses at most Llog/sub 2/N+L wavelengths for rings and 2(L-1)log/sub 2/N for trees. We also study the worst-case behavior of the well-known first-fit algorithm, and show that it requires at most 2.52Llog/sub 2/N+5L wavelengths (small variants of these constants are proven as well). When limited wavelength conversion is allowed, we first show how to use expanders to insure no blocking in arbitrary topologies. Then, we present conversion patterns for rings with conversion degree d=2, which require Llog/sub 2/L+4L or 2Llog/sub 2/log/sub 2/L+4L wavelengths, thereby eliminating the dependence (that exists without wavelength conversion) between the number of wavelengths and N. We also consider different traffic models where lightpath setup requests arrive over time, but once set up, lightpaths are never taken down. For this model, the number of wavelengths needed is shown to be only max{0,L-d}+L for a conversion degree of d.

We provide network designs for optical wavelength division multiplexed (OWDM) rings that minimize overall network cost, rather than just the number of wavelengths needed. The network cost includes the cost of the transceivers required at the nodes as well as the number of wavelengths. The transceiver cost includes the cost of terminating equipment as well as higher-layer electronic processing equipment, and in practice, can dominate over the cost of the number of wavelengths in the network. The networks support dynamic (time varying) traffic streams that are at lower rates (e.g., OC-3, 155 Mb/s) than the lightpath capacities (e.g., OC-48, 2.5 Gb/s). A simple OWDM ring is the point-to-point ring, where traffic is transported on WDM links optically, but switched through nodes electronically. Although the network is efficient in using link bandwidth, it has high electronic and opto-electronic processing costs. Two OWDM ring networks are given that have similar performance but are less expensive. Two other OWDM ring networks are considered that are nonblocking, where one has a wide sense nonblocking property and the other has a rearrangeably nonblocking property. All the networks are compared using the cost criteria of number of wavelengths and number of transceivers.

This paper presents methods for recovering from channel failures, link failures, and node failures in wavelength-division multiplexed (WDM) point-to-point links and ring networks with limited wavelength conversion/switching capabilities at the nodes. Different recovery schemes are presented to handle each type of failure. Each scheme is evaluated based on the network hardware configuration required to support it and the performance and management overheads associated with fault recovery. Although similar recovery techniques have been used in conventional networks such as SONET, the constraints due to limited wavelength conversion require new and more complex solutions.

This paper proposes optical wavelength division multiplexed (WDM) networks with limited wavelength conversion that can efficiently support lightpaths (connections) between nodes. Each lightpath follows a route in a network and must be assigned a channel on each link along the route. The load /spl lambda//sub max/ of a set of lightpaths is the maximum over all links of the number of lightpaths that use the link. At least /spl lambda//sub max/ wavelengths will be needed to assign channels to the lightpaths. If the network has full wavelength conversion capabilities, then /spl lambda//sub max/ wavelengths are sufficient to perform the channel assignment. Ring networks with fixed wavelength conversion capability within the nodes are proposed that can support all lightpath sets with load /spl lambda//sub max/ at most W-1, where W is the number of wavelengths in each link. Ring networks with a small additional amount of wavelength conversion capability within the nodes are also proposed that allow the support of any set of lightpaths with load /spl lambda//sub max/ at most W. A star network is also proposed with fixed wavelength conversion capability at its hub node that can support all lightpath sets with load /spl lambda//sub max/ at most W. These results are extended to tree networks and networks with arbitrary topologies. This provides evidence that significant improvements in traffic-carrying capacity can be obtained in WDM networks by providing very limited wavelength conversion capability within the network.

This paper presents methods for recovering from channel failures, link failures, and node failures in wavelength-division multiplexed (WDM) point-to-point links and ring networks, with limited wavelength conversion/switching capabilities at the nodes. Different recovery schemes are presented to handle each type of failure. Each scheme is evaluated based on the network hardware configuration required to support it, and the performance and management overheads associated with fault-recovery.

This paper proposes optical wavelength division multiplexed (WDM) networks with limited wavelength conversion that can efficiently support lightpaths (connections) between nodes. Each lightpath follows a route in the network and must be assigned a channel along each link in its route. The load /spl lambda//sub max/ of a set of lightpath requests is the maximum over all links of the number of lightpaths that use the link. At least /spl lambda//sub max/ wavelengths will be needed to assign channels to the lightpaths. If the network has full wavelength conversion capabilities then /spl lambda//sub max/ wavelengths are sufficient to perform the channel assignment. We propose ring networks with a fixed wavelength conversion capability within the nodes that can support all lightpath request sets with load /spl lambda//sub max/ at most W-1, where W is the number of wavelengths in each link. We also propose ring networks with selective pairwise wavelength conversion capability within the nodes that can support all lightpath request sets with load /spl lambda//sub max/ at most W. We also propose a star network with fixed wavelength conversion capability at its hub node that can support all lightpath request sets with load /spl lambda//sub max/ at most W. We extend this result to tree networks and networks with arbitrary topologies. These results show that significant improvements in traffic-carrying capacity can be obtained in WDM network by providing a very limited wavelength conversion capability within the network.

This paper describes an architecture for controlling a wavelength-routed optical network. The optical network provides reconfigurable connections that can be used to carry different types of data, at possibly different bit rates. A connection consists of a path in the network and a wavelength on each link in that path. This work focuses on the mechanisms for controlling the optical connections. Distributed control protocols are provided for setting up and taking down connections reliably. These protocols allow connection originators to obtain resources with high probability by minimizing reservation conflicts, allow connections to stay up even if controllers along the path in the network fail, ensure that controllers in the network have a consistent view of the state of each connection, and ensure that all resources taken up by a connection are released once the connection is taken down. Fast protocols are also proposed for handling link failures and wavelength failures on a link. These protocols are general in that they can be used in networks with: (1) different kinds of node architectures (electrical/optical switching, with or without wavelength conversion, etc.), (2) different route-computation algorithms that reflect the underlying constraints imposed by the nodes and the physical layer, and (3) any kind of underlying mechanism using which the network controllers can communicate with each other.

This paper describes an architecture for adding an underlying wavelength-routed optical layer to a standard ATM network. The optical layer provides reconfigurable lightpaths that serve as links between electronic ATM switches. A lightpath is carried on a particular wavelength and path in the underlying optical wavelength-routed network. This work focuses on the mechanisms for controlling the optical layer. Distributed control protocols are provided for setting up and taking down lightpaths reliably and for updating the topology of the network, including the lightpaths. These protocols allow lightpath originators to obtain resources with high probability by minimizing reservation conflicts, allow lightpaths to stay up even if controllers along the path in the network fail, ensure that controllers in the network have a consistent view of the state of each lightpath, and ensure that all resources taken up by a lightpath are released once the lightpath is taken down.

This paper studies the problem of dedicating routes to connections in optical networks. In optical networks, the vast bandwidth available in an optical fiber is utilized by partitioning it into several channels, each at a different optical wavelength. A connection between two nodes is assigned a specific wavelength, with the constraint that no two connections sharing a link in the network can be assigned the same wavelength. This paper considers optical networks with and without switches, and different types of routing in these networks. It presents optimal or near-optimal constructions of optical networks in these cases and algorithms for routing connections, specifically permutation routing for the networks constructed here.

In computer networks using wavelength-division multiplexing (WDM), it is often necessary to extend the length of a given WDM path beyond that of a single segment whose length Is limited by the link budget. While wavelength-flattened optical amplifiers are the most obvious solution, this paper calls attention to a simple (albeit less efficient) alternative that works for any number of wavelength channels, but only a modest number of segments and modest bit rates. This scheme substitutes for each optical amplifier stage a multiwavelength 2R remodulator consisting of a WDM demultiplexor, followed by wavelength-by-wavelength zero-crossing detectors, then remodulation wavelength by wavelength, and finally wavelength-division multiplexing back onto a single fiber. Theory and experimental results confirm the usefulness of this primitive technique, which can also be used for add-and-drop links, for wavelength routing or wavelength conversion in scalable wavelength routing networks, for internetwork gateways, and for other applications.

This paper describes the Rainbow-II optical metropolitan area network (MAN), which supports 32 nodes each at 1 Gbit/s over a distance of 10-20 km. Rainbow-II uses optical wavelength-division multiplexing (WDM), in a broadcast star architecture. Each node uses a separate fixed wavelength for transmitting data and a tunable receiver for receiving one of several data streams. The network is implemented in the form of optical network nodes, each attached to a host computer via the high-performance parallel interface (HIPPI). Each network node contains protocol processing hardware to offload the protocol processing work from the host computer onto the node. The goal is to provide full gigabit-per-second bandwidth to end-user supercomputer applications. Preliminary protocol performance measurements in a testbed network are given.

The problem of designing a logical topology over a wavelength-routed all-optical network (AON) physical topology is studied. The physical topology consists of the nodes and fiber links in the network. On an AON physical topology, we can set up lightpaths between pairs of nodes, where a lightpath represents a direct optical connection without any intermediate electronics. The set of lightpaths along with the nodes constitutes the logical topology. For a given network physical topology and traffic pattern, our objective is to design the logical topology and the routing algorithm so as to minimize the network congestion while constraining the average delay seen by a source-destination pair and the amount of processing required at the nodes (degree of the logical topology). Ignoring the delay constraints can result in fairly convoluted logical topologies with very long delays. On the other hand, in all our examples, imposing it results in a minimal increase in congestion. While the number of wavelengths required to imbed the resulting logical topology on the physical all optical topology is also a constraint in general, we find that in many cases of interest this number can be quite small. We formulate the combined logical topology design and routing problem described above as a mixed integer linear programming problem which we then solve for a number of cases of a six-node network. This programming problem is split into two subproblems: logical topology design, and routing. We then compare the performance of several heuristic topology design algorithms against that of randomly generated topologies, as well as lower bounds.

Considers routing connections in a reconfigurable optical network using WDM. Each connection between a pair of nodes in the network is assigned a path through the network and a wavelength on that path, such that connections whose paths share a common link in the network are assigned different wavelengths. The authors derive an upper bound on the carried traffic of connections (or equivalently, a lower bound on the blocking probability) for any routing and wavelength assignment (RWA) algorithm in such a network. The bound scales with the number of wavelengths and is achieved asymptotically (when a large number of wavelengths is available) by a fixed RWA algorithm. The bound can be used as a metric against which the performance of different RWA algorithms can be compared for networks of moderate size. The authors illustrate this by comparing the performance of a simple shortest-path RWA (SP-RWA) algorithm via simulation relative to the bound. They also derive a similar bound for optical networks using dynamic wavelength converters, which are equivalent to circuit-switched telephone networks, and compare the two cases. Finally, they quantify the amount of wavelength reuse achievable in large networks using the SP-RWA via simulation as a function of the number of wavelengths, number of edges, and number of nodes for randomly constructed networks as well as de Bruijn networks. They also quantify the difference in wavelength reuse between two different optical node architectures.>

The author consider the problem of routing connections in an optical network using wavelength division multiplexing, where each connection between a pair of nodes in the network is assigned a path through the network and a wavelength on that path, such that connections whose paths share a common link in the network are assigned different wavelengths. They derive an upper bound on the carried traffic of connections (or equivalently, a lower bound on the blocking probability) for any routing and wavelength assignment (RWA) algorithm in such a network. The bound scales with the number of wavelengths and is achieved asymptotically (when a large number of wavelengths is available) by a fixed RWA algorithm. Although computationally intensive, this bound can be used as a metric against which the performance of different RWA algorithms can be compared for networks of moderate size. They illustrate this by comparing the performance of a simple RWA algorithm via simulation with our bound. They also derive a similar bound for optical networks using dynamic wavelength converters, which are equivalent to circuit-switched telephone networks, and compare the two cases for different examples.>

We propose a new architecture for a high-speed packet-switched metropolitan-area lightwave network using a combination of wavelength division multiplexing over fiber and subcarrier multiplexing within each wavelength. Each station transmits on a fixed-wavelength (that may also be used by some other stations) at any one of many available subcarrier frequencies within that wavelength, and receives on some fixed wavelength and fixed subcarrier frequency. Rapid tuning between subcarrier frequencies allows a station to transmit packets to all stations that receive on some subcarrier channel on its transmit wavelength. Packets intended for stations that do not receive on the wavelength of the transmitting station must be routed through intermediate stations. By avoiding rapid tuning between wavelengths, we realise a practical multihop architecture for packet switching with one optical transmitter shared among many stations and one optical receiver per station. This architecture imposes a constraint on the multihop connectivity pattern in the network since all stations using a common transmit wavelength send packets directly to the same set of other stations. The same constraint is applicable to shared-channel multihop networks with a single transmitter and receiver per station. We introduce the notion of "clusterable" directed graphs to represent the connectivity patterns that are possible, and using the properties of these graphs, show that nodes can be added to and removed from the network with minimal disruption to the other nodes. We show that some well-known multihop topologies such as shufflenets and de Bruijn graphs are clusterable. Power budget calculations indicate the feasibility of a 32-wavelength network with five 200 Mb/s FSK subcarrier channels on each wavelength. Our throughput calculations indicate that such a network could support 160 stations with a throughput per station of 44 Mb/s.>

Proposes de Bruijn graphs as logical topologies for multihop lightwave networks. After deriving bounds on the throughput and delay performance of any logical topology, the authors compute the throughput and delay performance of de Bruijn graphs for two different routing schemes and compare it with their bounds and the performance of shufflenets. For a given maximum nodal in- and out-degree and average number of hops between stations, a logical topology based on a de Bruijn graph can support a larger number of stations than a shufflenet and this number is close to the maximum that can be supported by any topology. The authors also propose de Bruijn graphs as good physical topologies for wavelength routing lightwave networks consisting of all-optical routing nodes interconnected by point-to-point fiber links. The worst-case loss experienced by a transmission is proportional to the maximum number of hops (diameter). For a given maximum nodal in- and out-degree and diameter, a physical topology based on a de Bruijn graph can support a large number of stations using a relatively small number of wavelengths.>

A media-access protocol for high-speed packet-switched multichannel networks that are based on a broadcast topology, such as optical passive star networks using wavelength-division multiple access, is described. The protocol supports connection-oriented traffic with or without bandwidth reservation, as well as datagram traffic to integrate transport-layer functions with the media-access layer. It uses the bandwidth efficiently while keeping the processing requirements low by requiring stations to compute their transmission and reception schedules only at the start and end of each connection. Analysis results show that low blocking probabilities for connections and high network throughput can be achieved.>

This paper proposes a new media-access protocol for high-speed packet-switched multichannel networks based on a broadcast topology, for example, optical passive star networks using wavelength-division multiple access. The protocol supports connection-oriented traffic with or without bandwidth reservation as well as datagram traffic, in an attempt to integrate transport-layer functions with the media-access layer. It utilizes the bandwidth efficiently while keeping the processing requirements low by requiring stations to compute their transmission and reception schedules only at the start and end of each connection. A simple analysis shows that we can achieve low blocking probabilities for connections, as well as high network throughput.

The authors de Bruijn graphs as logical topologies for multihop lightwave networks. The throughput and delay performance of networks based on de Bruijn graphs are compared with those of shufflenets. It is shown that for a given maximum nodal in- and out-degree and average number of hops between stations, a network based on a de Bruijn graph can support a much larger number of stations than a shufflenet. This increase in the number of stations is obtained at the expense of some nonuniformity in edge loading when the traffic distribution over the source-destination pairs is uniform. In the 1024-node example, this results in a slightly lower throughput for the network based on the de Bruijn graph compared to the shufflenet, but the former has lower delay at low offered loads.>

A dynamic time-wavelength division multiaccess protocol (DT-WDMA) is proposed for metropolitan-sized multichannel optical networks employing fixed wavelength transmitters and tunable optical receivers. Control information is sent over a dedicated signaling channel and data are sent over channels owned by the transmitters. Time is divided into slots on each channel and slots on the control channel are further split into mini-slots. Fixed time-division multiaccess (TDM) is used within each slot on the control channel. Transmitters indicate their intention to transmit a packet by transmitting the destination address during their appropriate mini-slot in the control channel and then transmit their packet in the next slot on their data channel. Receivers listen to the control channel and tune to the appropriate channel to receive packets addressed to them. A common but distributed arbitration algorithm is used to resolve conflicts when packets from many transmitters contend for the same receiver. Each receiver executes the same deterministic algorithm to choose one of the contending packets. Each transmitter uses the same algorithm to determine the success or failure of its packet.>

A first-generation design, called Rainbow, for optical wavelength division multiaccess (WDMA) computer networks is described. The Rainbow research prototype takes the form of a direct detection, circuit-switched metropolitan-area-network (MAN) backbone consisting of 32 IBM PD/2's as gateway stations, communicating with each other at 200-Mb/s data rates and submillisecond switching times. The prototype architectural options for realizing WDMA networks are discussed, along with reasons for choosing this design point. Experimental measurements on the prototype are presented. Ways of extending this prototype to more stations, higher bit rates, and faster setup times are given.>

A distributed algorithm is presented for obtaining an efficient and conflict-free broadcasting schedule in a multi-hop packet radio network. The inherent broadcast nature of the radio channel enables a node's transmission to be received by all other nodes within range. Multiple transmissions can be scheduled simultaneously because of the multi-hop nature of the network. It is first shown that the construction of a broadcasting schedule of minimum length is NP-complete, and then a centralized algorithm based on a sequential graph-coloring heuristic is presented to construct minimal-length schedules. A distributed implementation of this algorithm is then proposed, which is based on circulating a token through the nodes in the network.>