WDM光网络中支持QoS的波长分配算法

基于在波长可变光网络中的波长分层图模型,本文提出了一种在动态业务情况下的支持不同QoS要求的路由波长分配算法,根据客户层业务不同的QoS要求,通过分层图模型中参数的不同取值,对其光路建立请求区别对待,提供不同级别的光路建立.仿真结果表明该算法具有很好的性能,提高了全网的资源利用率,满足了客户层业务在建立光通路时不同的QoS要求对网络阻塞率的要求.     

全光网络迂回路由的算法及性能分析

分别对波长通道（ＷＰ）和虚波长通道（ＶＷＰ）两种机制的全光网络迂回路由算法的阻塞性能进行了数值分析，并给出了一个任意拓扑全光网络的计算机数值模拟，对比了两种机制的网络阻塞性能。     

干扰长度对WDM光网络阻塞性能的影响

引入WDM光网络中静态业务下一种适合于计算各种网络阻塞概率的分析模型，计算出波长连续网络和波长可转换网络的阻塞慨率，采用波长转换增益定性分析了干扰长度对这两种网络阻塞性能的影响，并提出关于干扰长度的相对波长转换增益系数和相对通道长度系数的概念。Matlab仿真的结果表明，波长转换增益最大值大约为相对通道长度系数的一半；当干扰长度较大时，波长连续网络中的波长利用率较高，波长转换增益不明显。应用到网络拓扑设计时，可考虑相对通道长度系数，适当减小通道长度而增大干扰长度，从而减少波长转换器的数目，降低网络的硬件花费，提高网络的使用效率。     

WDM网络中波长转换稀疏性与有限波长转换能力的研究

阐述稀疏波长转换和有限波长转换技术的概念,讨论了这两种技术对网络阻塞率和波长转换增益的影响,并给出了一种用于波长和波长转换器分配的启发式算法。     

波长转换对抗毁多光纤网状WDM网络的性能影响

目前研究波长转换器对WDM网络性能影响的文献一般都是针对单光纤网络并且不考虑网络抗毁的要求。该文使用整数线性规划建模和动态系统仿真的方法,分别研究了静态业务和动态业务下,多光纤网状WDM网络在满足抗毁要求时,节点配置波长转换器对网络性能的影响,以此对使用波长转换器获得的好处给出较为全面的结论。     

干扰长度对WDM光网络阻塞性能的影响

引入WDM光网络中静态业务下一种适合于计算各种网络阻塞概率的分析模型,计算出波长连续网络和波长可转换网络的阻塞概率,采用波长转换增益定性分析了干扰长度对这两种网络阻塞性能的影响,并提出关于干扰长度的相对波长转换增益系数和相对通道长度系数的概念.Matlab仿真的结果表明,波长转换增益最大值大约为相对通道长度系数的一半;当干扰长度较大时,波长连续网络中的波长利用率较高,波长转换增益不明显.应用到网络拓扑设计时,可考虑相对通道长度系数,适当减小通道长度而增大干扰长度,从而减少波长转换器的数目,降低网络的硬件花费,提高网络的使用效率.     

WDM网络中一种支持优先级的波长分配算法

有效的算法能大大的提高波长利用率,文中研究了WDM光网络在固定路由无波长转换条件下的波长分配算法,以便充分挖掘WDM网络的带宽资源。提出了一种基于RLI并考虑通道优先级的波长分配算法,并给出了该算法的具体实现步骤。最后的仿真结果表明,与RLI算法相比该算法显著的降低了网络的阻塞率,有利于提高网络资源的利用率。     

波长变换对光网络中阻塞性能的影响

在网状光网络中建立稀疏波长变换模型,定性的计算了有和没有波长转换器网络的阻塞概率。研究了在这两种网络中通道长度对阻塞概率的影响,以及波长转换器对提高网状光网络的阻塞性能的影响。建模结果表明,对大容量的网状网,波长变换器能大大降低网络的阻塞概率,但对一个波长数目不定且业务量接近饱和的网络作用不大。     

抗毁WDM网络中支持多优先级的波长分配算法

WDM网络将在未来的骨干网中扮演重要的角色,具备抗毁能力和支持多优先级都很必要.本文提出了两种基于优先级的波长分配算法——波长编号法和分配限额法,并研究了动态业务下,抗毁WDM网络使用这两种波长分配方法后网络的阻塞率性能.研究的内容包括业务量负载、光纤链路的波长通道数和每条链路包含的光纤数对网络性能的影响.本文还给出了仿真结果.     

基于等价网络的WDM光网络多播路由与波长分配算法

针对节点功能不同光网络中的路由与波长分配问题,建立了一种具有节点功能区分的波分复用器(WDM)多波长光网络模型,根据节点功能将其分为A,B两类,并提出了波长等价弧、等价网络,以及等价多播树等概念.在此基础上基于节点功能区分的WDM光网络的特点和相应的路由和波长分配策略,提出了一种如何在此类WDM光网络中实现多播连接的快速有效的算法,分析了算法计算复杂度,指出了算法的优缺点.     

Survivability Performance Evaluation of WDM Networks with Wavelength Converters

Network Survivability is defined as the ability of a network to support the committed Quality of Services (QoS) continuously in the presence of different failure scenarios. Both availability and performance degradation of a system in presence of failure are integral components of survivability evaluation. Therefore, a composite model is presented for network survivability that includes system availability analysis to find out the cost due to system downtime, and system failure impact analysis to find out the transient performance degradation when failure occurs. A new analytical technique is presented to evaluate the excess loss due to failure (ELF) as the transient performance degradation when the system is operating in gracefully degraded states. Single and multiple link failures as well as node failures are considered. An algorithm is proposed to carry out the steady state availability analysis of a network even when the available paths between a pair of nodes are non-disjoint. A recursive generalized form of the availability algorithm when there are multiple paths available between a pair of nodes is presented. The availability model and the performance model are combined to construct a hierarchical model to evaluate the network survivability performance. Simulation results are used to validate the proposed model. A WDM network with wavelength conversion is considered as an example for this evaluation.     

Performance Analysis of WDM Optical Networks with Wavelength Usage Constraint

Due to power considerations, it is possible that not all wavelengths available in a fiber can be used at a given time. In this paper, an analytical model is proposed to evaluate the blocking performance of wavelength-routed optical networks with and without wavelength conversion where the usable wavelengths in a fiber is limited to a certain maximum number, referred to as wavelength usage constraint. The effect of the wavelength usage constraint is studied on ring and mesh-torus networks. It is shown that the analytical model closely approximates the simulation results. We also evaluate the performance of the first-fit wavelength assignment algorithm and compare its performance with the random wavelength assignment algorithm through simulation. It is observed that increasing the total number of wavelengths in a fiber is an attractive alternative to wavelength conversion when the number of usable wavelengths in a fiber is maintained the same.     

Dimensioning of the Wavelength Converters in a WDM Optical Packet Switch

This paper analyzes the performances of a WDM optical packet switch making use of fiber delay lines to resolve output packet contentions; the optical packet switch is equipped with tunable wavelength converters which can shift optical packets to any wavelength of the output link which they are directed to. As not all packets need conversion, we propose a dimensioning technique allowing to reduce the number of converters and to improve the signal quality by reducing unnecessary conversions. The obtained results show that a remarkable reduction of the number of converters is obtained with respect to that needed by other switches described in literature. Such a saving is achieved by maintaining the packet loss probability below a prefixed threshold.     

WDM Passive Optical Networks With Wavelength Sharing

A wavelength-division-multiplexing-based passive optical network (PON) is demonstrated in which wavelengths can be shared by a flexible number of users on other PONs served from the same central office. Bidirectional transmission is demonstrated through a typical PON at 2.5 Gb/s with wavelength sharing of 16.      

Performance Analysis of Distributed Protocols for Bidirectional Wavelength Reservations in Dynamic WDM Networks

In this paper, we present an analytical model to investigate the blocking performance of distributed wavelength reservation protocols in dynamic WDM networks. This model extends the approach of [S. Arakawa et al., Proceedings of ITC-CSCC, (Sado, Japan, July 1999), pp. 828--831] in two aspects: firstly we take into account the contention between the two end nodes of bidirectional links in bidirectional wavelength reservations. Secondly, we make a distinction between the effects of different kinds of network delays, which include signaling transmission delay, signaling processing delay, and optical cross-connects (OXC) operation delay. As an application, we use our method to analyze the performance of two different signaling processing schemes, namely forward after cross-connect (FAXC) and forward before cross-connect (FBXC) [M. Goyal et al., Technical Digest of OFC2002 (Anaheim, CA, Mar. 2002) pp. 489--490]. The accuracy of our analytical model is validated by discrete-event simulations. Numerical results show that the FBXC scheme can give a lower network blocking probability than FAXC, and the efficiency of FBXC becomes more significant as the OXC operation delay increases.     

Wavelength Conversion in WDM Optical Networks: Strategies and Algorithms for Limiting the Number of Converters in the Optical Cross-Connects

This paper considers the problem of wavelength conversion in optical networks using wavelength division multiplexing technique. In the previous literature, two main wavelength routing and assignment strategies have been introduced: wavelength path (WP) and virtual wavelength path (VWP), depending on whether the signal stays on the same wavelength or is converted to another during its travel throughout the network. While the former method does not require any wavelength conversion, the latter needs wavelength conversion in each optical node and, in particular, a wavelength converter per each signal handled by the node itself. From the previous literature emerged that the VWP leads to optical cross-connect (OXC) with lower dimensions compared to the ones required by the WP scheme, and that the difference between the WP and VWP schemes increases as the number of wavelengths carried by each fiber increases. In this paper a new strategy is introduced, named partial virtual wavelength path (PVWP), with the related wavelength routing and assignment algorithm, which makes limited use of wavelength conversion compared to the VWP scheme, and allows the same advantages of VWP to be attained with lower OXC dimensions. The paper reports a comparative analysis among the different strategies, considering both the cases of a network without failures and a network with the possibility of failure restoration. The main result is that the proposed PVWP strategy allows the same advantages of the VWP scheme with a strongly reduced number of wavelength converters (around 5% of the number required by VWP scheme). This figure does not vary appreciably if failure restoration is considered. The new strategy can be adopted by using an opportune OXC architecture, as illustrated in the paper, which allow a limited number of converters to be shared among all the channels as a common pool.     

波分复用全光网中的一种新的波长分配算法

研究了动态业务下，采用固定选路和备用选路的波分复用光传送网中的波长分配问题．提出了一种新的波长分配算法，新算法在已市算法的基础上进行了改进，与已有算法相比，该算法能更精确地描述波长分配对全厨状态的影响。计算机仿真表明，如果用阻塞率来衡量，该波长分配算法优于文献中已有的波长分配算法。     

WDM网络中一种考虑优先级的波长分配算法

提出了一种在多优先级的情况下,基于RCL算法来提高网络流量的波长分配新算法.文章对于具有相同优先级的光路建立请求时,采用相对容量损失的RCL法分配波长,而对于不同优先级的光路请求,通过调整门限值来保证网络尽可能地接纳更多低优先级的网络流量.仿真结果表明,当负载较高时,多接纳的网络流量已十分明显.     

Dynamic Wavelength Assignment for WDM All-Optical Tree Networks

We develop an on-line wavelength assignment (WA) algorithm for a wavelength-routed WDM tree network. The algorithm dynamically supports all$bf k$-port traffic matrices among$N$end nodes, where$bf k$denotes an integer vector$[k_1 ldots, k_N]$and end node$i, , 1leq ileq N$, can transmit at most$k_i$wavelengths and receive at most$k_i$wavelengths. Our algorithm is rearrangeably nonblocking, uses the minimum number of wavelengths, and requires at most$d^ast-1$lightpath rearrangements per new session request, where$d^ast$is the degree of the most heavily used node. We observe that the number of lightpath rearrangements per new session request does not increase as the amount of traffic$bf k$scales up by an integer factor. In addition, wavelength converters cannot reduce the number of wavelengths required to support$bf k$-port traffic in a tree network. We show how to implement our WA algorithm using a hybrid wavelength-routed/broadcast tree with only one switching node connecting several passive broadcast subtrees. Finally, using roughly twice the minimum number of wavelengths for a rearrangeably nonblocking WA algorithm, we can modify the WA algorithm to be strict-sense nonblocking.