光突发交换的交换控制策略和光缓存配置

光突发交换是面向下一代互联网的光交换模式.讨论了异步光突发交换系统的交换控制策略以及相应光缓存的优化配置策略.除了传统预约模式的交换策略,还研究了非预约和改进型预约模式,它们的性能评估由计算机仿真给出.结果表明:在条件相当情况下,改进型预约模式具有最低的丢包率.同时光缓存的配置对上述几类交换控制策略的性能都有很大影响,仿真结果指出:光缓存的粒度对系统性能具有重要影响,所讨论的几类交换模式都存在最佳时延粒度,研究结果对光交换矩阵的设计有指导意义.     

一种光突发交换网络中降低填充开销的突发组装算法

针对光突发交换（0BS）网络现有组装算法在输入业务量较小时组装填充开销较大的性能缺陷，提出了一种新的突发组装算法，称之为BPRA算法．该组装算法通过改变突发生成时间，在不增加端到端时延的条件下利用偏置时间来延长突发组装时间，从而可以有效降低填充空闲比特的数目．同时提出一个近似分析模型来计算空闲比特填充概率．分析和仿真结果表明BPRA算法可以明显降低组装填充开销．     

光突发交换网络试验平台

研究并实现了几种光突发交换网络的关键技术,包括高速的突发组装技术、突发调度技术及突发光发射及接收技术等,其中突发接收具有较高的灵敏度和小于80 ns的时钟数据恢复时间.在这些技术的基础上,建立了一个灵活且可扩展的光突发交换网络实验平台.在OBS网络平台上进行了TCP性能评估实验和多QoS业务传输实验。实验结果表明光突发网络中丢包率严重影响其上TCP的性能,但若丢包率小于0.1%,则TCP性能可得到一定程度的保证.光突发交换网络实验平台上的TCP传输实验还表明了OBS物理层带来的额外延时对上层TCP带宽有极大的影响.实验结果还验证了OBS网络中的QoS保证机制.     

光分组交换矩阵的实验研究

对光交换矩阵的基本机理进行了研究,介绍了所研制成功的光分组交换矩阵的基本结构、功能、性能和相关的实现技术,重点分析了质子交换型铌酸锂（APE：LiNbO3)光波导开关与相应的系统技术,给出了2．5Gbit/s异步转移模式（ATM）光信元交换实验研究结果。     

一种改进型光可分突发交换性能分析模型

提出了一种改进型光可分突发交换性能分析模型(Modified Optical Composite Burst Switching,MOCBS).该模型考虑了截断剩余造成的影响,通过分析截断概率、平均截断剩余长度,得到截断剩余造成的分组丢弃概率,从而更准确地估计分组丢失率.在此基础上,推导出支持区分服务的POCBS(Prioritized OCBS)性能分析公式.仿真验证了MOCBS的准确性以及POCBS分析公式的有效性.     

一种光突发交换网络中降低填充开销的突发组装算法(英文)

针对光突发交换(OBS)网络现有组装算法在输入业务量较小时组装填充开销较大的性能缺陷,提出了一种新的突发组装算法,称之为BPRA算法.该组装算法通过改变突发生成时间,在不增加端到端时延的条件下利用偏置时间来延长突发组装时间,从而可以有效降低填充空闲比特的数目.同时提出一个近似分析模型来计算空闲比特填充概率.分析和仿真结果表明BPRA算法可以明显降低组装填充开销.     

光突发交换网络核心节点中QoS策略研究

对OBS网络核心节点中所采取的多种改善QoS的策略进行了研究并提出了一种反馈证实机制以减少数据突发的丢失.其原理是在BCP传输建立端到端通路的过程中,如果在某核心节点处不能成功预约波长信道,该节点将在数据突发离开边缘节点之前返送NACK证实信息,取消边缘节点的该数据突发调度输出计划,并将其重新缓冲排队以备再次调度发送.这样可以尽量为数据突发提供可靠的波长路径,减少数据突发的丢失.     

TCP在光突发交换试验网络中性能的实验研究

在光突发交换试验网络中对TCP的性能进行了实验研究.首先研究了丢包对网络性能的影响,表明突发包丢失将导致TCP吞吐量的快速下降,并且丢包率越高,可用的TCP传输带宽越窄.然后分别详细研究了OBS试验网络中的延时损伤和报文段关联增益,以及对TCP传输性能的影响.最后综合考虑这些因素.实验结果表明,在本OBS试验网络中存在一个不随丢包率变化的最优化突发包组装时间250us使得可用TCP带宽最大化.为了最大限度地提高TCP的传输性能,组包时间应该选择等于或稍大于这个最优值.     

光突发交换网络中最长队列优先调度组装机制

提出了光突发交换网络中的最长队列优先调度（LQF-SA）机制,将边缘节点的突发组装和突发调度两种技术紧密结合在一起.在现实网络中,业务通常是非均匀分布的,即在某些源宿对之间存在更大的业务流量.仿真结果表明,LQF-SA机制可以很好地完成非均匀分布业务流的组装.即使在均匀分布业务流输入下,LQF-SA机制在突发大小分布、组装效率和突发丢失率等方面也优于已提出的轮询调度组装（RR-SA）机制.     

A maximum-efficiency-first multi-path route selection strategy for optical burst switching networks

In this paper, the impact of a path selection on other existing paths in optical burst switching (OBS) networks is studied by analyzing the contention among different traffic streams and the interaction between the route selection and traffic load balance. The results show that there exists a mutual reinforcement interaction among the traffic load of a path, the path burst loss ratio and the contention ability of the path when burst loss ratio based multi-path selection strategies are adopted, which may increase the unbalance of traffic and lead to severe congestion further. A maximum-efficiency-first multi-path selection strategy, which considers the performance of the burst flows and the impact of a path selection on existing OBS paths at the same time by a combined metric of route efficiency, is proposed to maximize the utility of the burst flows and minimize the increment of lost throughput on the path. The performance of the proposed multi-path selection strategy is evaluated through simulation. The results show that the presented strategy obviously outperforms the least burst loss ratio strategy and shortest path first strategy in terms of the burst loss ratio in the practical unbalanced background traffic, especially when the network is heavily loaded.     

Performance analysis of an integrated scheme in optical burst switching high-speed networks

A new integrated scheme based on resource-reservation and adaptive network flow routing to alleviate contention in optical burst switching networks is proposed. The objective of the proposed scheme is to reduce the overall burst loss in the network and at the same time to avoid the packet out-of-sequence arrival problem. Simulations are carried out to assess the feasibility of the proposed scheme. Its performance is compared with that of contention resolution schemes based on conventional routing. Through extensive simulations, it is shown that the proposed scheme not only provides significantly better burst loss performance than the basic equal proportion and hop-length based traffic routing algorithms, but also is void of any packet re-orderings.     

Feed-forward control of SOA gain for power equalization in optical burst switching networks

We evaluate the performance of a novel technique to equalize power variation between optical bursts using feed-forward control of semiconductor optical amplifier gain. The technique enables large dynamic range power equalization of incoming optical bursts with up to 10 dB power difference. Moreover, the technique can remove SOA-induced waveform distortion. We achieve more than 4 dB sensitivity improvement with equalized burst power.     

Performance enhancement of optical burst switched networks

Optical burst switching (OBS) is an optical switching paradigm which offers a good tradeoff between the traditional optical circuit switching (OCS) and optical packet switching (OPS) since it has the relatively easy implementation of the first and the efficient bandwidth utilization of the second. Hence, OBS is a promising technology for the next generation optical Internet. A buffer-less OBS network can be implemented using ordinary optical communication equipment without the need for either wavelength converters or optical memories. In an OBS network, burst-loss performance is a critical concern. In OBS, the data-burst transmission is delayed by an offset time (relative to its burst control packet (BCP), or header) and the burst follows its header without waiting for an acknowledgment for resource reservation. Thus, a burst may be lost at an intermediate node due to contention, which is generally resolved according to the local routing and bandwidth information. The routing table maintained in each OBS node is generally pre-computed and fixed to forward the data bursts. Such a static forwarding feature might have limited efficiency to resolve contentions. Moreover, a burst may be lost and the network may be congested when a network element (e.g., fiber link) fails. In this paper, an efficient integrated scheme based on dynamic routing and burst segmentation has been proposed to improve reliability of data transport and network load balancing in optical burst switched networks. Simulation results demonstrate that the proposed approach reduces effectively blocking probability and hence contention. Further, it provides end-to-end throughput performance also. Hence, it establishes an appropriate tradeoff between loss rate and end-to-end throughput.     

Blocking and delay performance for differential output-ports choosing probability scheme applied optical burst switching network

Differential output-ports choosing probability (DOCP) scheme is a novel traffic outputting model for core router in optical burst switching (OBS) network. In this paper, we provide an analytical model for studying the performance of traffic blocking and delay in DOCP-applied OBS network. We first evaluate blocking probabilities using DOCP and confirm it by simulation. Then, in optical buffer-equipped OBS core router, we consider the average delay time for buffered traffic and the average total traffic queueing length in core router. The knowledge of delay performance is useful for the optical buffer architecture configuration in differential traffic scenario. Several results indicate that, under the same load condition, the blocking probabilities and the delay time will change along with the changeable ports choosing probability and the ratio between different length bursts in the OBS system.     

Distributed queue optical burst switched ring—A MAC protocol with zero blocking for OBS ring networks

Optical burst switching is a switching technology that is able to provide viable switching for optical internet with the present day technology. This paper presents a new medium access protocol for wavelength division multiplexing based optical burst switched metro ring network namely distributed queue optical burst switched ring (DQOBSR). It is a slotted, distributed queue access protocol with fixed bus structure and simple wavelength assignment. The topology consists of two unidirectional, counter rotating rings and the medium access protocol is based on distributed queuing. The protocol provides zero blocking with acceptable access delays. A mathematical model is also presented to determine the mean access delay which is the important performance parameter for the protocol. The simulation and model based results are compared and found to be in good agreement.     

A novel collision free WDM optical switching network

To avoid burst collision in OBS-WDM ring network, a novel scheme called Estimated Traffic based Longest Queue First (ETLQF) has been proposed. In it, the scheduler considers both the queue length and the incoming traffic. The scheduler can adjust the queue lengths in advance to cater to the incoming traffic, targeting to reduce cell loss ratio (CLR) and improve the buffer utilization. The simulation results show that the ETLQF in OBS ring networks not only reduces CLR by 10-60%, but also achieves high throughput while saving 65% of wavelengths, when compared to conventional schemes with only destination delay.     

一种新型的突发汇聚算法研究

分析了现有的突发汇聚算法,同时针对现有汇聚算法的一些特点,提出了一种新型突发包汇聚算法,可以实现在不同网络负载的情况下,根据到达边缘节点的速率同时结合网络的性能(丢包率),动态的调整突发包汇聚的门限值,从而实现了智能组包,克服了原有算法简单,不灵活的缺点.通过仿真结果可以得知,该算法可以明显改善光突发交换网络的丢包率和时延性能,能够很好地支持实时性要求很高的数据业务.     

The cost benefits of multi-layer architecture with multi-granularity switching in optical networks

The benefits of using multi-granularity switching to reduce the number of ports, including both electronic and optical ports, have been investigated. A novel multi-layer ring architecture with multiple switches is proposed and two general equations to count the number of ports are derived. Comparisons are made between the multi-layer multi-granularity switching (MM-XC) architecture and the reported architectures: multi-switching (M-XC) and single switching (S-XC) architectures. The results show that the M-XC or S-XC architectures can be deduced from the MM-XC architecture and the latter architecture can achieve more considerable savings of the port count compared to the former.     

Performance analysis of dedicated protection and shared protection for waveband switching based on multi-granularity optical cross-connect

In order to save the switching ports and the cost of Multi-Granularity Optical Cross-Connect (MG-OXC), the waveband switching technique was proposed to groom multiple wavelength-level traffic to a few waveband tunnels to be switched by a few switching ports in MG-OXC. At the same time, protection for fibers is very important to ensure the service continuity since each wavelength carries a lot of traffic. Although existing works have addressed the waveband switching protection, most of them separately considered only the dedicated protection or shared protection in static demand scenario and did not deeply analyze and compare the dedicated protection and shared protection in dynamic demand scenario. Therefore, in this paper, we deeply study the consumptions of wavelengths in fibers and ports in MG-OXCs for Waveband Shared Protection (WSP) and Waveband Dedicated Protection (WDP) in dynamic demand scenario, and propose the port-cost calculation and update methods based on a new waveband layered auxiliary graph that is developed based on MG-OXC structure. In simulations, we compare WSP, WDP, traditional end-to-end waveband shared protection and traditional end-to-end waveband dedicated protection. Simulation results show that the shared protection has smaller port-cost, better wavelength utilization efficiency and lower blocking probability than the dedicated protection with the same waveband switching policy, and the sub-path waveband switching has bigger port-cost, better wavelength utilization efficiency and lower blocking probability than the end-to-end waveband switching with the same backup wavelength assignment policy.