Search

Ring networks have drawn wide attention due to token ring and FDDI standards. This paper presents a scalable multihop WDM passive ring architecture for local area or metropolitan area networks. The network consists of optical interfaces, which are connected into a ring by fibers. Each optical interface is completely passive and unpowered. We apply the "multihop" principle by introducing a regular multiloop ring called "wheel" as the optical connectivity and embedding it onto a single-loop fiber ring. Each network access node requires only a small number of fixed-tuned transceivers and an ATM switch. The network can incrementally scale up from a smaller ring to a larger ring by deploying more access nodes and optical interfaces. When scaling up the network, not only the number of wavelengths required is small but also no more wavelengths are required. This architecture has an excellent ability for alternative routing and optimal wavelength reuse. We present three distributed adaptive wavelength routing algorithms and an optimal wavelength assignment algorithm which minimizes the number of wavelengths used. A method yielding the lowest possible call blocking probability for an "ideal" network with a fixed number of transceivers per node is presented and used as a yardstick to evaluate the proposed networks. The numerical results show that the architecture is very efficient.

In this paper, we present three logarithmically scalable routing algorithms for very large optical networks. The algorithms are based on a hierarchical approach. Each algorithm requires a different amount of information to be stored at each node, and results in a different efficiency. The network capacity provided by each algorithm is compared with that provided by the global shortest path algorithm (which is not scalable). It is found that the network capacities of the three algorithms are almost the same as that of the global shortest path algorithm. For fixed routing algorithms, we use an approximate analysis, appeared in the literature, from which the blocking probability at a fixed network load can iterately be obtained (we use call blocking probability as our quality objective and compare network capacity by observing the difference in the offered load which produces the same blocking probability). The approximate analysis is validated through simulation and is found to be very accurate. Scalable, quasi-dynamic routing algorithms are also studied. Numerical results show that, when the node-capacity is equal to the capacity of a single wavelength and the bandwidth required per call is large (greater than 24% of the link capacity), the capacity provided by the quasi-dynamic routing algorithm is very close to that of an infinite capacity centralized switch (lowest possible call blocking caused exclusively by congestion on the finite capacity user input/output links, never by the switch fabric itself). In all the examples considered, the improvement of the quasi-dynamic routing algorithm over the fixed routing algorithm is significant.

This paper proposes scalable conflict-free media-access protocols for WDM optical passive star networks using shared parallel queueing and wormhole scheduling. Shared parallel queueing is to associate several parallel queues with a data channel, rather than one deep queue (as in DT-WDMA) or N static queues (as in [5] and [6], N is the number of stations). This significantly reduces both the hardware resource requirement and the control signalling demand. Wormhole scheduling is to schedule multiple successive packets which are destined to the same receiver at one time, decreasing the overhead of tuning caused by relatively slow filter and relieving the electronic processing bottleneck. An analytical model is presented and verified through simulation. Numerical results from analysis and simulation indicate that with only 5-6 parallel queues, the protocols are capable of achieving high throughput. Thus the protocols can be scaled up for networks with a large number of stations and could lead to a potential implementation.

This paper proposes scalable conflict-free media-access protocols with a novel architecture called parallel queueing for WDM optical passive star networks. The processing and hardware resource requirements are low and the control signalling demand is significantly reduced by using a small number of parallel queues in each station. A new scheme called wormhole scheduling is introduced to compensate the relatively slow laser-filter tuning speed. The protocol using wormhole scheduling decreases the overhead of tuning (thus increases the throughput) and relieves the electronic processing bottleneck. The key ideas are: (1) to associate several parallel queues with a data channel, rather that one deep queue (as in DT-WDMA) of N static queues, N is the number of stations; and (2) to schedule multiple successive packets which are destined to the same receiver at one time. An analytical model is presented and verified through simulation. Numerical results from analysis and simulation indicate that with only 5-6 parallel queues, the protocols are capable of achieving high throughput. Thus the protocols can be scaled up for networks with a large number of stations and could lead to potential implementation. >

Ring networks have drawn wide attention due to token ring and FDDI standards. This paper presents a scalable multihop WDM passive ring architecture for local-area or metropolitan area networks. The network consists of optical interfaces, which are connected into a ring by fibers. Each optical interface is completely passive and unpowered. We apply the "multihop" principle by introducing a regular multi-loop ring called "wheel" as the optical connectivity and embedding it onto the single-loop fiber ring. Each network access node requires only a small number of fixed-tuned transceivers and an ATM switch. The network can incrementally scale up from a smaller ring to a large ring by deploying more access nodes and optical interfaces. When scaling up the network, not only the number of wavelengths required is small but also no more wavelengths are required. This architecture has an excellent ability for alternative routing and optimal wavelength reuse. We present three distributed adaptive wavelength routing algorithms and an optimal wavelength assignment algorithm which minimizes the number of wavelengths used. A method yielding the lowest possible call blocking probability for an "ideal" network with a fixed number of transceivers per node is presented and used as a yardstick to evaluate the proposed networks. The numerical results show that the architecture is very efficient.

Multihop lightwave networks enjoy increasing prominence as viable candidates for high-speed local, metropolitan, and wide-area networks. Of interest is the end-user capacity which may be afforded by a multihop lightwave network. For uniform traffic patterns, it is possible to compute both the expected number of hops taken by a representative packet and the total network capacity. When the traffic patterns are non-uniform, simple expressions for network capacity cannot be produced. The authors apply information theoretic concepts to find an upper-bound on the capacity which may be produced for an arbitrary non-uniform user to-user traffic pattern, by any multihop network, along with a lower bound on the expected number of hops. Such bounds are useful as yardsticks against which the performance of specific approaches may be compared or ensure that, the network can meet the intended traffic load. Furthermore, they are equally applicable to any type of networks characterized by packet switching nodes/access stations and directed connectivity graphs.

ShuffleNet as an optical connectivity has been proposed extensively for use in the implementation of high-throughput multihop lightwave networks. A ShuffleNet uses a small, fixed number of transceivers per network access node and provides nearly optimal throughput. However, building multihop networks using ShuffleNets requires either a large number of fibers or a large number of wavelengths. We for the first time show that topologically a ShuffleNet is the multiplication of hypercube and ring, similar to cube-connected-cycle (CCC). This finding relates ShuffleNet to the well-understood hypercube and enables us to embed ShuffleNet on mesh network using a small number of fiber buses and an even smaller number of wavelengths. This architecture provides the practicality for implementation. For example, with two transceivers per node, a 384-node ShuffleNet uses 96 fiber buses and 6 wavelengths. We conclude that the WDM mesh-based ShuffleNet is suited for high-speed local-area or metropolitan-area networks.

Proposes a new media-access protocol for WDM star coupler networks of which the stations have arbitrary distances to the star coupler. The network architecture is called CC-FTFT-FRTR, in which each station sends packets at a unique wavelength using a fixed-tuned transmitter and receives packets using a tunable receiver. The authors propose a novel synchronization scheme which allows stations to have arbitrary (or dynamic) propagation delays to the star coupler. In addition, a new control signalling scheme is also presented to lower the signalling bandwidth demand. A new protocol, called systolic sliding windows (SSW), can yield the same throughput as the protocols in Chen and Yum (1991) and Chipalkatti et al. (1993); compared to the conventional approach that extra delay lines are used to make all stations have the same propagation delay to the star coupler, this protocol is capable of coping with the dynamic change of propagation delays from the stations to the star coupler, incurs less network delay, and requires no accurate measurement of propagation delays from stations to the star coupler. It is the first single-hop protocol based on CC-FTFT-FRTR structure that can achieve higher throughput than DT-WDMA for networks of which the stations can have arbitrary or dynamic propagation delays to the star coupler.

In reference one, a network architecture called hyper-cluster employing wavelength-routing, multi-hop packet switching and optical reconfiguration is presented. It is modularly scalable to very large configurations on both hardware and operational bases. A hyper-cluster uses a logical hierarchy for the purpose of addressing but guarantees that all access nodes have a constant number of transceivers. It is a cluster of regular graphs; the clustering structure follows the traffic distribution in a grand granularity. The issue of operational scalability is addressed by presenting a scalable routing protocol and a scalable reconfiguration protocol. When using shuffle-net as the building block, a novel routing scheme called quantified deflection routing is presented. The scheme improves call blocking performance significantly. In this paper, we propose a distributed reconfiguration protocol. The network throughput and virtual call blocking performance is obtained via simulation on large networks (with size beyond 200 nodes). Numerical results show that the dynamic self- routing protocol, combined with quantified deflection routing for shuffle-net, can achieve excellent resource utilization efficiency for very large networks. When the call arrival rate is below 0.3, the capacity provided by the hyper-cluster dynamic routing algorithm is close to that of an infinite capacity centralized switch (lowest possible call blocking caused exclusively by congestion on the finite capacity user input/output links, never by the switch fabric itself).© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.