WO2008061461A1 - Procédé et dispositif de protection de diffusion de groupe dans un réseau optique à multiplexage par division de longueur d'onde - Google Patents

Procédé et dispositif de protection de diffusion de groupe dans un réseau optique à multiplexage par division de longueur d'onde Download PDF

Info

Publication number
WO2008061461A1
WO2008061461A1 PCT/CN2007/070503 CN2007070503W WO2008061461A1 WO 2008061461 A1 WO2008061461 A1 WO 2008061461A1 CN 2007070503 W CN2007070503 W CN 2007070503W WO 2008061461 A1 WO2008061461 A1 WO 2008061461A1
Authority
WO
WIPO (PCT)
Prior art keywords
path
protection
working
node
source node
Prior art date
Application number
PCT/CN2007/070503
Other languages
English (en)
French (fr)
Inventor
Cai Lu
Lemin Li
Huiying Xu
Hongfang Yu
Hongbin Luo
Original Assignee
Huawei Technologies Co., Ltd.
University Of Electronic Science And Technology Of China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co., Ltd., University Of Electronic Science And Technology Of China filed Critical Huawei Technologies Co., Ltd.
Priority to CN2007800002880A priority Critical patent/CN101322343B/zh
Priority to EP07800979.2A priority patent/EP1993223B1/en
Publication of WO2008061461A1 publication Critical patent/WO2008061461A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0227Operation, administration, maintenance or provisioning [OAMP] of WDM networks, e.g. media access, routing or wavelength allocation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0227Operation, administration, maintenance or provisioning [OAMP] of WDM networks, e.g. media access, routing or wavelength allocation
    • H04J14/0238Wavelength allocation for communications one-to-many, e.g. multicasting wavelengths
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0278WDM optical network architectures
    • H04J14/0284WDM mesh architectures
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0287Protection in WDM systems
    • H04J14/0289Optical multiplex section protection
    • H04J14/0291Shared protection at the optical multiplex section (1:1, n:m)

Definitions

  • the present invention relates to WDM optical network technologies, and in particular, to multicast protection technologies in WDM optical networks. Background technique
  • WDM Widelength Division Multiplexing
  • service provision technologies which are the backbone of the future Internet (Internet) backbone
  • Internet Internet
  • multicast refers to the transmission of data from one source node to multiple destination nodes.
  • the well-known SaD (Splitter and Delivery) switching module can support the implementation of multicast.
  • the SaD switching module can divide one wavelength resource into multiple wavelength resources, or divide one wavelength resource into multiple wavelength resources.
  • the DXC (Optical Cross-Connect) that supports multicast can be constructed by using the SaD switch module.
  • the OXC can also be called MC-OXC (Multicast Capable OXC), which is mainly used to implement optical crossover.
  • devices with TaC (Tap and Continue) capability can also support multicast implementation.
  • FIG. 1 is a schematic diagram of multicast data transmission.
  • the source node s transmits data to the destination nodes dl, d2, and d3, respectively.
  • u and V may be referred to as intermediate nodes, and the edge between each two nodes is called a link, and each link has at least one link.
  • a wavelength resource, a channel composed of at least one link from a source node to any destination node is called a path.
  • s->u->dl is a path from s to dl.
  • the link may fail or fail for a variety of reasons. If the source node has only one path to the destination node, the path will also fail when the link fails, which will result in data transmission blocking.
  • the source node to the destination node have a working working path
  • the source node to the destination node There is an alternate protection path.
  • the link in the protection path is completely different from the link in the working path.
  • the source node can transmit data to the destination node through the protection path.
  • the path consisting of solid lines is the working path
  • the path consisting of the dotted lines is the protection path.
  • the working path of s to dl is s->u->dl.
  • the data is transmitted by the protection path s->v->dl.
  • the constructed work tree is as shown in Figure 2A, and the work tree includes s to dl respectively.
  • the working paths of d2 and d3 are constructed as shown in Figure 2B.
  • the protection tree includes protection paths from s to dl, d2 and d3 respectively. If the working path of s to dl fails, the data of s to dl can be Convert to the corresponding protection path transmission in the protection tree.
  • the dual-tree protection method for link separation is simple and intuitive, and the algorithm complexity is relatively low. However, the one-time construction of the working tree occupies more network resources, and the link of the protection path cannot be duplicated with the link of the working path.
  • the multicast data transmission diagram assumes that the multicast request is s to transmit data to dl and d2 respectively.
  • the cost of s->dl and s->d2 is 1, dl->d2 and d2-> respectively.
  • the cost of dl is 2, the cost of d2->u and u->d2 is 3, and the cost of dl->u and u->dl is 4, respectively.
  • the constructed working tree is as shown in the figure.
  • the protection tree since s->dl and s->d2 have already been used as working paths, the protection path of s to dl and d2 cannot be found, and the protection tree cannot be constructed. Therefore, the dual-tree protection method of link separation occupies more network resources when constructing the working tree, so that the probability of success of constructing the protection tree is low, and the probability of network congestion is large.
  • the technical problem to be solved by the embodiments of the present invention is to provide a multicast protection method and apparatus in a WDM optical network to reduce the probability of network congestion.
  • the technical solution provided by the embodiment of the present invention is as follows:
  • a multicast protection method in a WDM optical network comprising: obtaining a working path of a source node to a destination node; obtaining a protection path corresponding to the working path;
  • a multicast protection device in a WDM optical network comprising:
  • a path obtaining unit configured to obtain a working path of the source node to a destination node, and a protection path corresponding to the working path
  • the completion determining unit is configured to determine whether the working path of the source node to all the destination nodes and the corresponding protection path have been obtained, and if yes, the operation is ended, otherwise the path obtaining unit is notified that the path is not obtained in the multicast service.
  • the destination node operates.
  • the embodiment of the present invention has the following advantages: In the embodiment of the present invention, each time a working path is obtained, the protection path corresponding to the working path is obtained, so that the first working path is avoided. Working tree and occupying more network resources In the embodiment of the present invention, the shortest path is selected as the working path or the protection path, so that the selected one can be selected.
  • the cost of the working path or the protection path is the most d; in the embodiment of the present invention, the node configured with the wavelength converter and/or the optical splitter is selected as the node on the working path or the protection path as much as possible, due to the wavelength converter and/or The optical splitter has the function of converting wavelength resources or splitting light, so the subsequent working path or protection path can be selected by means of such a node. If the working path or the protection path is selected, the working path or the protection path that is not obtained is occupied as little as possible. The node can provide more room for the subsequent selection of the working path or the protection path.
  • the above technical means can further improve the probability of successfully constructing the working tree and the protection tree, and reduce the probability of network congestion;
  • the configuration is first selected to have waves.
  • the node of the long converter or the splitter acts as a node on the working path or the protection path. If the subsequent working path or protection path is selected to occupy such a node as much as possible, subsequent nodes that are not on the obtained working path or protection path Whether or not a wavelength converter or a splitter is configured may have little effect on successfully constructing a working tree or a protection tree, and therefore, embodiments of the present invention can be applied under conditions of sparse configuration.
  • FIG. 1 is a schematic diagram of multicast data transmission
  • FIG. 2A is a schematic diagram of a working tree constructed by using a link-separated dual-tree protection method for data transmission shown in FIG. 1
  • FIG. 2B is a dual-tree protection using link separation.
  • the method is a schematic diagram of a protection tree constructed by the data transmission shown in FIG. 1.
  • FIG. 3 is another schematic diagram of multicast data transmission
  • FIG. 3A is a working tree constructed by using the double-tree protection method of link separation for the data transmission shown in FIG.
  • Figure 4 is a flow chart of an embodiment of the present invention
  • 5A is a schematic diagram of a working tree constructed by using s to dl in FIG. 3
  • FIG. 5B is a schematic diagram of a protection tree constructed by using the embodiment of the present invention for data transmission shown in FIG. 3;
  • FIG. The embodiment is a working tree diagram of the data transmission construction shown in FIG. 3;
  • FIG. 6A is a network physical topology diagram;
  • FIG. 6B is an auxiliary diagram of FIG. 6A;
  • FIG. 7 is a network topology diagram for simulation according to an embodiment of the present invention;
  • Fig. 8 is a simulation effect diagram of the network blocking probability in the case where the configuration of the optical splitter and the wavelength converter are the same and the number of destination nodes is different;
  • Fig. 9 is a simulation effect diagram of the network blocking probability in the case where the number of optical splitters is different;
  • FIG. 4 is a flow chart of an embodiment of the present invention.
  • a multicast protection method in a WDM optical network includes the following steps:
  • S401 Obtain a working path from the source node to a destination node. Since it is a multicast service, there may be multiple destination nodes.
  • the working path from which the source node first obtains the destination node may be randomly selected, or may start from the destination node where the data transmitted by the source node first arrives, and obtain the source node to this.
  • the working path of the destination node There may be multiple candidate working paths from the source node to a destination node, and the candidate working path refers to the source. All paths from the node to the destination node that can act as a working path, including all paths in the source node to the destination node that have not been determined as protection paths.
  • the first shortest working path may be selected from the source node to the candidate working path of the destination node as the working path of the source node to the destination node, where the first shortest working path refers to the source node to The least costly working path among all the candidate working paths of the destination node, the cost may refer to the distance from the source node to the destination node, and may also refer to the cost of transmitting the unit traffic. If the working path of the source node to the destination node is not obtained before the working path of the source node to the destination node is obtained, a shortest path may be selected from the source node to the destination node as the source node to the destination. The working path of the node.
  • the cost of all the nodes on the obtained working path to the candidate working path of the destination node may be calculated.
  • the candidate work path with the least cost is called the second shortest work path
  • the node with the candidate work path having the least cost to the destination node on the obtained work path is called the source node of the second shortest work path
  • the source node is The combination of the working path of the source node and the second shortest working path to the second shortest working path serves as the working path of the source node to the destination node.
  • the data is transmitted in the optical network as a carrier with a wavelength resource, and each wavelength resource can only transmit data from the source node to a destination node, but if a node is configured with a wavelength converter, it can be from a wavelength resource.
  • the received data continues to be transmitted with another wavelength resource, which expands the selection of the working path from the source node to the destination node.
  • data received from one wavelength resource can be transmitted in multiple identical wavelength resources, which also expands the selection range of the working path from the source node to the destination node.
  • the source node and/or the source node of the second shortest working path may be configured with a wavelength converter and/or a splitter, or, if the source node is obtained from the working path of a destination node, at least one source node has been obtained to the other
  • the working path of the destination node can calculate the cost of the candidate working path of the node with the wavelength converter and/or the splitter on the obtained working path to the obtained working path, and then select the shortest working path. If a first shortest working path is selected as the working path of the source node to a destination node, the network resources may be occupied as little as possible, and the corresponding protection path may be obtained subsequently and subsequently obtained.
  • the working path and the protection path provide more options and thus reduce the probability of network congestion. If the nodes on the selected working path are mostly nodes configured with a wavelength converter and/or a splitter, the selection range can be expanded for the subsequent selection of the working path and the protection path, thereby further reducing the probability of network congestion.
  • S402 Obtain a protection path corresponding to the working path. In this embodiment, after obtaining one working path, the protection path corresponding to the working path is calculated, and the corresponding protection path refers to the path of the source node and the destination node having the same working path as the working path, and all links on the protection path are protected. Separated from all links on the working path. There may be multiple candidate protection paths from the source node to a destination node.
  • the candidate protection path refers to all paths that can be used as protection paths from the source node to the destination node, including the source node to the destination node is not determined to be working. All paths to the path.
  • the first shortest protection path may be selected from the source node to all the candidate protection paths of the destination node as the protection path of the source node to the destination node, where the first shortest protection path refers to the source node to The least costly protection path among all the candidate protection paths of the destination node, the cost may refer to the distance from the source node to the destination node, and may also refer to the cost of transmitting the unit traffic.
  • the cost of the candidate protection path of all the nodes on the obtained protection path to the destination node may be calculated.
  • the least costly protection path is called the second shortest protection path
  • the node with the least cost protection candidate path on the obtained protection path and the destination node is called the source node of the second shortest protection path, and the source node is to this.
  • the combination of the protection path of the source node of the second shortest protection path and this second shortest protection path serves as a protection path from the source node to the destination node.
  • the data is transmitted in the optical network as a carrier with a wavelength resource, and each wavelength resource can only transmit data from the source node to a destination node, but if a node is configured with a wavelength converter, it can be from a wavelength resource.
  • the received data continues to be transmitted with another wavelength resource, which expands the selection range of the protection path from the source node to the destination node.
  • data received from one wavelength resource can be transmitted in multiple identical wavelength resources, which also expands the selection range of the protection path from the source node to the destination node.
  • the source node and / or The source node of the second shortest protection path may be configured with a wavelength converter and/or a splitter, or, if the protection path of the source node to the destination node is obtained before the protection path of the source node to the destination node is obtained,
  • the cost of the candidate protection path for all nodes on the protected path that are configured with the wavelength converter and/or the splitter can be calculated to select the shortest protection path. If the first shortest protection path is selected as the protection path of the source node to the destination node, the network resources can be occupied as little as possible, and more options for the subsequent obtaining the working path or the protection path can be provided, thereby reducing network congestion. Probability. If the nodes on the selected protection path are mostly nodes configured with a wavelength converter and/or a splitter, the selection range may be expanded for the subsequent selection of the working path or the protection path, thereby further reducing the probability of network congestion.
  • S403 Obtain a working path of the source node to the other destination node and a corresponding protection path until all the working paths and all the protection paths are respectively constructed into one working tree and one protection tree. Since it is a multicast service, there may be multiple destination nodes. In this case, the number of working paths and protection paths obtained may be the same as the number of destination nodes, respectively. However, it must be calculated after each working path is obtained. The protection path corresponding to the working path, and then the working path of the source node to another destination node, and so on, until the working path and the protection path of the source node to all the destination nodes are obtained. In addition, each time a working path is obtained, the working path can be added to the working tree constructed by all the working paths obtained before.
  • the protection path can be added to all previously obtained protection paths.
  • the protection tree is constructed until all the working paths and all the protection paths are respectively constructed into a working tree and a protection tree, wherein the working tree and the protection tree take the source node as the root node, the destination node as the leaf node, the working path and the protection path.
  • the link on the side is the edge.
  • the protection route construction from the source node to all destination nodes is completed, if the working tree Any link on the link is invalid If the working path fails, the data transmitted on the failed working path can be converted to the corresponding protection path in the protection tree for transmission.
  • the links on all working paths in this embodiment can be completely separated from the links on all protection paths, that is, no link can be ensured to be the same, which can ensure that the working path fails.
  • the transmitted data is recovered for transmission.
  • FIG. 3 For the multicast data transmission shown in Figure 3, you can take the following steps to build the working tree and protection tree: First select the working path from s to dl. In all paths from s to dl, select the shortest path s->dl as the shortest path.
  • the nodes, d2, and dl are leaf nodes, and the links of s->d2 and d2->dl are used as edges.
  • the protection tree constructed is as shown in FIG. 5B.
  • N is a node set, which is a link set, where each link is a bidirectionally transmitted fiber
  • W a wavelength resource on each fiber Number
  • WC f The wavelength converter configuration of the node, if the node is configured with a wavelength converter, then WCj Is 1; otherwise 0;
  • LG(N, E) Auxiliary-graph, where N is the set of nodes of the auxiliary map, E is the edge set of the auxiliary map, according to the network physical topology G (N, and the configuration of the node wavelength converter WCj , the auxiliary map LG (N, the construction process is as follows: the network physical topology G (N, 7 ⁇ copy W shares, form a layered layered network; for the node configuring the wavelength converter, with the edge of the weight 0 For each node corresponding to each layer, for example, the node configures the wavelength converter, and the node corresponding to each layer is f, then the edge with the weight value of 0 is connected to ff (0 ⁇ 1 ⁇ 2 ⁇ ), as shown in the figure. 6 ⁇ and 6 ⁇ , Figure 6 ⁇ is the network physical topology, each link has two wavelengths, and nodes 1 and 2 in the network are configured with wavelength converters, then Figure 6 is its auxiliary map;
  • ⁇ s, Dj multicast service request, where s is the source node of the multicast service, D is the destination node set of the multicast service; ⁇ indicates an empty set; T w : a work tree;
  • T p protection tree
  • v w a set of nodes that can be branched on the work tree, which can include two types of nodes, one is a node with a splitter on the work tree, and the other is a branch node on the work tree;
  • V p a set of nodes that can be branched on the protection tree, which may include two types of nodes, one is a node on the protection tree configured with a splitter, and the other is a node on the protection tree;
  • N w The set of nodes on the work tree.
  • N p Protects the set of nodes on the tree.
  • D* destination node /) set of nodes that are not covered in the set
  • the shortest path between nodes I to j; C p path;
  • the cost can be the cost of distance or transmission unit traffic. 4.
  • the WDM optical network has the following special constraints:
  • Wavelength continuity constraint Under this constraint, path calculation is closely related to wavelength selection.
  • the routing calculation problem under such constraints is called RWA (Routing and Wavelength Assignment).
  • the corresponding optical tree calculation problem is called MC-RWA (Multicast RWA, Multicast Routing and Wavelength Assignment);
  • Sparse configuration means that only some nodes in the WDM optical network can act as branch nodes of the multicast tree, or only some nodes configure wavelength converters or optical splitters, which makes the optical tree
  • the computational problem cannot solve the problem of the traditional Steiner Tree (the best generated subtree that covers a subset of network node sets).
  • the above two constraints are constraints that must be considered when constructing a multicast route. Of course, when studying the multicast protection routing problem of an optical network, these constraints are inevitably considered.
  • a node configured with a wavelength converter or a splitter is used as much as possible, so that more options can be provided for subsequent path selection, so the present invention is not only It involves wavelength selection and also involves routing calculations in sparse configurations.
  • multicast protection can be divided into dynamic protection and static protection. Static protection requires a given set of multicast requests to find the route and corresponding resource configuration method that satisfies these requests, and guarantees that all affected multicast connections can be 100% in the event of any single link failure. Recovery. Dynamic protection is for a multicast request, and sufficient link resources are reserved when establishing its multicast route to ensure 100% recovery when a single link fails.
  • the embodiment of the present invention may be applied to perform route calculation of a multicast request, and then apply another embodiment of the present invention to perform route calculation of another multicast request. By analogy, until the protection route calculation for all multicast requests is completed.
  • the examples were subjected to related simulation experiments.
  • the network topology used for simulation is shown in Figure 7.
  • Each edge is a bidirectionally transmitted fiber link.
  • the value on the side is the cost of the edge.
  • the dynamic multicast service model is used in the simulation.
  • the dynamic multicast service flow arriving at the network obeys the Poisson distribution. That is, the time interval between the two front-end multicast services arriving at the network is subject to a negative exponential distribution, and the duration of each multicast service is negative. index distribution.
  • the multicast protection route of the multicast service is calculated by using the dual-tree protection method of the link separation and the method provided by the embodiment of the present invention.
  • This simulation mainly investigates the blocking probability of the network to evaluate the performance of the protection method under the same network configuration and the same service strength.
  • the first is to investigate the probability of network congestion in the case where the splitter and the wavelength converter are configured identically and the number of destination nodes is different. Among them, 50% of the nodes (that is, 7 nodes) configure the optical splitter, and 50% of the nodes (that is, 7 nodes) configure the wavelength converter.
  • the node configuring the optical splitter and the nodes configuring the wavelength converter are randomly distributed.
  • Each type of splitter and wavelength converter performs 50,000 calls with different number of destination nodes, and the call strength is 20 Erlang.
  • the average of the blocking probabilities is obtained by taking 5 different distributions, and the simulation result shown in FIG. 8 is obtained.
  • the solid line in FIG. 8 indicates a dual-tree protection method for link separation, and the broken line indicates the method provided by the embodiment of the present invention.
  • the method of the embodiment of the present invention is used under the condition that the number of destination nodes is the same.
  • the probability of network congestion is lower than the probability of network congestion of the dual tree protection method of the link separation.
  • the second is to investigate the probability of network congestion in the case of different number of splitters.
  • the network has 50% of nodes configured with wavelength converters, the number of splitters is from 0 to 14, and the wavelength converter and the splitter are randomly distributed.
  • the number of destination nodes for each multicast call is 3, and the destination nodes are randomly distributed.
  • Different numbers of optical splitters are configured to perform 50,000 calls in the random distribution of the wavelength converter and the optical splitter, and the call strength is 20 Erlang.
  • the average of the blocking probabilities is obtained in five different configurations, and the simulation results shown in FIG. 9 are obtained.
  • the solid line in FIG. 9 indicates a dual-tree protection method for link separation, and the broken line indicates the method provided by the embodiment of the present invention. As can be seen from FIG. 9, the method of the embodiment of the present invention is used under the same number of optical splitters.
  • the probability of network congestion is lower than the probability of network congestion of the dual-tree protection method with link separation. Again, the probability of network congestion in the case of different wavelength converters is considered.
  • the network has 50% node configuration splitter.
  • the number of wavelength converters is from 0 to 14, and the wavelength converter and the optical splitter are randomly distributed.
  • the number of destination nodes per multicast call is 3, and the destination nodes are randomly distributed.
  • Different numbers of optical splitters are configured to perform 50,000 calls in the random distribution of the wavelength converter and the optical splitter, and the call strength is 20 Erlang 0.
  • the average of the blocking probabilities is obtained in 5 different configurations, and the simulation result shown in FIG. 10 is obtained.
  • the broken line indicates the method provided by the embodiment of the present invention.
  • the embodiment of the present invention is used.
  • the probability of network congestion of the method is lower than the probability of network congestion of the dual tree protection method of the link separation. It can be seen from the above three simulation results that the probability of network congestion using the method of the embodiment of the present invention is significantly lower than that of the double-tree protection method of the link separation.
  • the above description is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can also make several improvements and retouchings without departing from the principles of the present invention. It is considered as the scope of protection of the present invention.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Optical Communication System (AREA)
  • Small-Scale Networks (AREA)

Description

WDM光网络中组 ^护方法及装置
本申请要求于 2006 年 11 月 21 日提交中国专利局、 申请号为 200610145781.9、 发明名称为" WDM光网络中组播保护方法"的中国专利申请 的优先权, 其全部内容通过引用结合在本申请中。
技术领域 本发明涉及 WDM光网络技术, 尤其涉及 WDM光网络中组播保护技术。 背景技术
随着 WDM ( Wavelength Division Multiplexing, 波分复用)光网络应用的 迅速普及, 作为未来 Internet (因特网 ) 骨干支撑的 WDM光网络组网技术和 业务提供技术受到了越来越多的关注。 另一方面, 随着光学技术的日益成熟和 功能越趋完善的各种光通信器件 /设备的大批涌现, 许多原来需要在业务交换 层面完成的工作被更多地移植到了光层。一个典型的例子就是近年来广受关注 的光网络组播技术, 其中, 组播是指由一个源节点向多个目的节点传输数据。 在 WDM光网络中,组播的最终实现依赖于硬件设备的技术支撑。著名的 SaD ( Splitter and Delivery, 分光传输)交换模块就可以支撑组播的实现, SaD 交换模块可以将一个波长资源分成多路同一个波长资源,也可以将一个波长资 源分成多个波长资源。 利用 SaD 交换模块还可以构造任意规模的支持组播的 OXC (光交叉连接 ), OXC也可称为 MC-OXC ( Multicast Capable OXC, 支持 组播的光交叉连接), 主要用于实现光的交叉。 另外, 具有 TaC ( Tap and Continue, 分路继续)能力的设备也可以支撑组播的实现, 具有 TaC能力的设 备可以从光信号中取出一部分功率输出到本地电端口(如果本地是组播的目的 节点之一), 而余下的功率沿光路(light path, 跨越多个 OXC设备的透明波长 通路)继续传输, 由于不需要分光器, 所以组播路径不是树状结构, 而是一条 从源点出发、 经过所有组播目的节点的光路。 利用 SaD和 TaC硬件设备, 可 以将光路的概念扩展为光树( light tree ), 其基本思想是, 通过在 OXC中配置 分光器( light splitter或 power splitter ), 构造波长级别的全光透明的树状通路, 从而实现点对多点的全光组播。 图 1为一个组播数据传输示意图。 如图 1所示, 源节点 s分别向目的节点 dl、 d2及 d3传输数据, u、 V可以称为中间节点, 每两个节点之间的边称为链 路,每条链路至少有一个波长资源, 源节点到任意一个目的节点的由至少一条 链路组成的通道称为路径, 例如, s->u->dl即为 s到 dl的路径。 在实际应用中, 由于多种原因, 链路可能会出现故障或者失效, 如果源节 点到目的节点只有一个路径, 则链路失效时, 路径也会失效, 进而导致数据传 输阻塞。 为防止这种情况的出现, 本领域技术人员釆用源节点到目的节点的双 路径方式的技术手段, 即, 源节点到目的节点存在一个正常工作的工作路径的 同时, 源节点到目的节点还存在一个备用的保护路径,保护路径中的链路与工 作路径中的链路完全不同, 当工作路径失效时, 源节点可以通过保护路径将数 据传输到目的节点。 如图 1所示, 由实线组成的路径为工作路径, 由虚线组成 的路径为保护路径, 正常情况下, s到 dl的工作路径为 s->u->dl , 当工作路径 失效时, 由保护路径 s->v->dl传输数据。 目前,存在多种技术方案可以实现为源节点到目的节点之间构建工作路径 和保护路径,链路分离的双树保护方法即为其中的一种方案。这种方案的基本 思想为, 为每个组播请求先后构建一个工作树及一个保护树, 工作树或保护树 中的每个工作路径或保护路径均为源节点到对应目的节点的最短工作路径或 保护路径, 且工作树中的链路与保护树中的链路分离。如果工作树中任意一条 链路失效, 则数据可以转换到保护树中传输。 如图 1所示, 假设组播请求为 s将要分别向 dl、 d2及 d3传输数据, 在链 路已知的前提下, 构建的工作树如图 2A所示, 工作树中分别包括 s到 dl、 d2 及 d3的工作路径, 构建的保护树如图 2B所示, 保护树中分别包括 s到 dl、 d2及 d3的保护路径, 如果 s到 dl的工作路径失效, 则 s到 dl的数据可以转 换到保护树中的对应的保护路径传输。 链路分离的双树保护方法简单直观, 算法复杂度比较低,但是一次性构建 工作树会占用较多的网络资源,并且保护路径的链路不能与工作路径的链路重 复, 这种情况下, 保护路径的选择余地非常小, 以致构建保护树的成功几率较 低, 进而导致网络阻塞的概率较大。 如图 3所示的组播数据传输示意图, 假设组播请求为 s将要分别向 dl及 d2传输数据, s->dl和 s->d2的代价分别为 1 , dl->d2和 d2-> dl的代价分别 为 2, d2->u和 u-> d2的代价分别为 3 , dl->u和 u-> dl的代价分别为 4,。 使 用链路分离的双树保护方法构建工作树时, 由于 s到 dl、 d2的代价最小的路 径(即最短工作路径)分别为 s->dl、 s->d2, 所以构建的工作树如图 3A所示, 但在构建保护树时, 由于 s->dl、 s->d2 已经作为工作路径而导致无法找到 s 到 dl、 d2的保护路径, 进而无法构建保护树。 所以,链路分离的双树保护方法在构建工作树时占用较多的网络资源, 以 致构建保护树的成功几率较低, 进而导致网络阻塞的概率较大。 发明内容
本发明实施例要解决的技术问题在于提供一种 WDM光网络中组播保护 方法及装置, 以降低网络阻塞的概率。 本发明实施例提供的技术方案如下:
.一种 WDM光网络中组播保护方法, 包括 : 获得源节点到一个目的节点的工作路径;获得所述工作路径对应的保护路 径;
重复上述步骤,直至获得源节点到所有目的节点的工作路径及对应的保护 路径为止。 一种 WDM光网络中组播保护装置, 其特征在于包括:
路径获得单元, 用于获得源节点到一个目的节点的工作路径, 以及所述工 作路径对应的保护路径;
完成判断单元,用于确定源节点到所有目的节点的工作路径及对应的保护 路径是否都已获得, 若都已获得则结束操作, 否则通知所述路径获得单元对组 播业务中的未得到路径的目的节点进行操作。
与现有技术相比, 本发明实施例具有以下优点: 在本发明实施例中,每获得一个工作路径后, 随即再获得工作路径对应的 保护路径,这样就避免了先由所有的工作路径构建工作树而占用较多的网络资 源、进而无法成功构建保护树的情况出现, 所以釆用本发明实施例可以降低网 络阻塞的概率; 在本发明实施例中,尽量先选择最短路径作为工作路径或保护路径, 这样 可以保证选择的工作路径或保护路径的代价最 d、; 在本发明实施例中, 尽量先选择配置有波长转换器和 /或分光器的节点作 为工作路径或保护路径上的节点, 由于波长转换器和 /或分光器具有转换波长 资源或分光的作用, 所以可以借助这样的节点选择后续的工作路径或保护路 径,如果在选择工作路径或保护路径时,尽量少占用未在已获得的工作路径或 保护路径上的节点,则可以为后续的工作路径或保护路径的选择提供较多的余 地,所以,釆用上述技术手段可以进一步提高成功构建工作树和保护树的几率, 降低网络阻塞的概率; 由上述可知,在本发明实施例中,尽量先选择配置有波长转换器或分光器 的节点作为工作路径或保护路径上的节点,如果后续的工作路径或保护路径的 选择尽量占用这样的节点,则后续的未在已获得的工作路径或保护路径上的节 点是否配置有波长转换器或分光器对成功构建工作树或保护树的影响会很小, 因此, 本发明的实施例可以应用在稀疏配置的条件下。 附图说明 图 1为一个组播数据传输示意图; 图 2A为使用链路分离的双树保护方法为图 1所示的数据传输构建的工作 树示意图; 图 2B为使用链路分离的双树保护方法为图 1所示的数据传输构建的保护 树示意图; 图 3为另一个组播数据传输示意图; 图 3A为使用链路分离的双树保护方法为图 3所示的数据传输构建的工作 树示意图; 图 4为本发明实施例的流程图; 图 5A为使用本发明实施例为图 3中的 s到 dl构建的工作树示意图; 图 5B为使用本发明实施例为图 3所示的数据传输构建的保护树示意图; 图 5C为使用本发明实施例为图 3所示的数据传输构建的工作树示意图; 图 6A为一个网络物理拓朴图; 图 6B为图 6A的辅助图; 图 7为本发明实施例用于仿真的网络拓朴图; 图 8为分光器和波长转换器配置相同、目的节点数量不同的情况下的网络 阻塞概率的仿真效果图; 图 9为分光器数量不同情况下的网络阻塞概率的仿真效果图; 图 10为波长转换器数量不同情况下的网络阻塞概率的仿真效果图。
具体实施方式 下面将结合附图,对本发明的最佳实施方案进行详细描述。 首先要指出的 是, 本发明中用到的术语、 字词及权利要求的含义不能仅仅限于其字面和普通 的含义去理解,还包括进而与本发明的技术相符的含义和概念, 这是因为作为 发明者, 要适当地给出术语的定义, 以便对本发明进行最恰当的描述。 因此, 本说明和附图中给出的配置, 只是本发明的首选实施方案, 而不是要列举本发 明的所有技术特性。要认识到,还有各种各样的可以取代本发明方案的同等方 案或修改方案。 首先对本发明实施例的技术方案进行说明。 图 4 为本发明实施例的流程 图。 如图 4所示, 本发明实施例提供的一种 WDM光网络中组播保护方法, 包 括如下步骤:
S401 : 获得源节点到一个目的节点的工作路径。 由于是组播业务, 所以目的节点可以有多个,先获得源节点到哪个目的节 点的工作路径可以随机选择,也可以从源节点传输的数据最先到达的目的节点 开始, 获得源节点到这个目的节点的工作路径。 源节点到一个目的节点可能存在多个侯选工作路径,侯选工作路径是指源 节点到这个目的节点的所有能够作为工作路径的路径,其包括源节点到这个目 的节点的所有路径中还未被确定为保护路径的路径。在本实施例中, 可以从源 节点到这个目的节点的所有侯选工作路径中选择第一最短工作路径作为源节 点到这个目的节点的工作路径, 其中, 第一最短工作路径是指源节点到这个目 的节点的所有侯选工作路径中代价最小的工作路径,代价可以指源节点到这个 目的节点的距离, 还可以指传输单位业务量的代价。 如果获得源节点到一个目的节点的工作路径之前还没有获得任何一个源 节点到其它目的节点的工作路径,则可以从源节点到这个目的节点的所有路径 中选择一个最短路径作为源节点到这个目的节点的工作路径。如果获得源节点 到一个目的节点的工作路径之前已经获得至少一个源节点到其它目的节点的 工作路径,则可以计算已获得的工作路径上的所有节点到这个目的节点的侯选 工作路径的代价, 代价最小的侯选工作路径称为第二最短工作路径,在已获得 的工作路径上与这个目的节点具有最小代价的侯选工作路径的节点称为第二 最短工作路径的源节点,将源节点到这个第二最短工作路径的源节点的工作路 径与第二最短工作路径的组合作为源节点到目的节点的工作路径。 另外,数据在光网络中以波长资源作为载体传输, 并且每个波长资源只能 传输源节点到一个目的节点的数据, 但是, 如果一个节点配置有波长转换器, 则可以将从一个波长资源中接收的数据以另外一个波长资源继续传输,这样就 会扩大源节点到目的节点的工作路径的选择范围。 同样,如果一个节点配置有 分光器, 则可以将从一个波长资源中接收的数据以多路相同的波长资源传输, 这样也会扩大源节点到目的节点的工作路径的选择范围。 所以, 源节点和 /或 第二最短工作路径的源节点可以配置有波长转换器和 /或分光器, 或者说, 如 果获得源节点到一个目的节点的工作路径之前已经获得至少一个源节点到其 他目的节点的工作路径,则可以计算这个目的节点到已获得的工作路径上的所 有配置有波长转换器和 /或分光器的节点的侯选工作路径的代价, 进而选择出 最短工作路径。 如果首先选择一个第一最短工作路径作为源节点到一个目的节点的工作 路径, 则可以尽量少的占用网络资源, 为后续获得对应的保护路径及后续获得 工作路径及保护路径提供更多的选择余地, 因而可以降低网络阻塞的概率。 如果选择的工作路径上的节点多数为配置有波长转换器和 /或分光器的节 点, 则可以为后续的工作路径及保护路径的选择扩大选择范围, 进一步降低网 络阻塞的概率。 S402: 获得所述工作路径对应的保护路径。 在本实施例中,每获得一个工作路径后,就计算工作路径对应的保护路径, 对应的保护路径是指与工作路径具有相同的源节点及目的节点的路径,并且保 护路径上的所有链路与工作路径上的所有链路分离。 源节点到一个目的节点可能存在多个侯选保护路径,侯选保护路径是指源 节点到这个目的节点的所有能够作为保护路径的路径,其包括源节点到这个目 的节点的未被确定为工作路径的所有路径。在本实施例中, 可以从源节点到这 个目的节点的所有侯选保护路径中选择第一最短保护路径作为源节点到这个 目的节点的保护路径, 其中, 第一最短保护路径是指源节点到这个目的节点的 所有侯选保护路径中代价最小的保护路径,代价可以指源节点到这个目的节点 的距离, 还可以指传输单位业务量的代价。 如果获得源节点到一个目的节点的保护路径之前已经获得至少一个源节 点到其他目的节点的保护路径,则可以计算已获得的保护路径上的所有节点到 这个目的节点的侯选保护路径的代价,代价最小的保护路径称为第二最短保护 路径,在已获得的保护路径上与这个目的节点具有最小代价的侯选保护路径的 节点称为第二最短保护路径的源节点 ,将源节点到这个第二最短保护路径的源 节点的保护路径与这个第二最短保护路径的组合作为源节点到目的节点的保 护路径。 另外,数据在光网络中以波长资源作为载体传输, 并且每个波长资源只能 传输源节点到一个目的节点的数据, 但是, 如果一个节点配置有波长转换器, 则可以将从一个波长资源中接收的数据以另外一个波长资源继续传输,这样就 会扩大源节点到目的节点的保护路径的选择范围。 同样,如果一个节点配置有 分光器, 则可以将从一个波长资源中接收的数据以多路相同的波长资源传输, 这样也会扩大源节点到目的节点的保护路径的选择范围。 所以, 源节点和 /或 第二最短保护路径的源节点可以配置有波长转换器和 /或分光器, 或者说, 如 果获得源节点到一个目的节点的保护路径之前已经获得至少一个源节点到其 他目的节点的保护路径,则可以计算这个目的节点到已获得的保护路径上的所 有配置有波长转换器和 /或分光器的节点的侯选保护路径的代价, 进而选择出 最短保护路径。 如果首先选择一个第一最短保护路径作为源节点到一个目的节点的保护 路径, 则可以尽量少的占用网络资源, 为后续获得工作路径或保护路径提供更 多的选择余地, 因而可以降低网络阻塞的概率。 如果选择的保护路径上的节点多数为配置有波长转换器和 /或分光器的节 点, 则可以为后续的工作路径或保护路径的选择扩大选择范围, 进一步降低网 络阻塞的概率。
S403: 获得源节点到其他目的节点的工作路径及对应的保护路径,直至将 所有的工作路径及所有的保护路径分别构建成一个工作树及一个保护树为止。 由于是组播业务, 所以目的节点可以有多个, 这种情况下, 获得的工作路 径及保护路径的数量可以分别与目的节点的数量相同,但是, 必须保证每获得 一个工作路径后, 就计算工作路径对应的保护路径,之后再获得源节点到另外 一个目的节点的工作路径, 以此类推, 直至获得源节点到所有目的节点的工作 路径和保护路径为止。 另外, 每获得一个工作路径后, 就可将工作路径加入到 之前获得的所有工作路径构建的工作树, 同理, 每获得一个保护路径后, 就可 将保护路径加入到之前获得的所有保护路径构建的保护树,直至所有的工作路 径及所有的保护路径分别构建成一个工作树及保护树, 其中, 工作树及保护树 以源节点为根节点,目的节点为叶子节点,工作路径及保护路径上的链路为边。 在将工作路径或保护路径加入到工作树或保护树中时,如果工作路径或保护路 径上的链路与已加入到工作树或保护树中的链路相同,则可以只将还没有加入 到工作树或保护树中的链路加入到工作树或保护树中。 如果获得一个包含源节点到所有目的节点的工作路径的工作树并且获得 一个包含源节点到所有目的节点的保护路径的保护树,则说明源节点到所有目 的节点的保护路由构建完成,如果工作树上的任意一条链路失效而导致所在的 工作路径失效,则可以将失效的工作路径上传输的数据转换到保护树中的对应 的保护路径进行传输。
需要说明的是,本实施例中的所有工作路径上的链路与所有保护路径上的 链路可以实现完全分离, 即可以保证没有任何一条链路相同, 这样可以保证因 工作路径失效而导致不能传输的数据恢复传输。 现在以图 3为例,对本发明实施例的方法进行说明。对于图 3所示的组播 数据传输可以釆取下述步骤构建工作树及保护树: 首先选择 s到 dl的工作路径, 在 s到 dl的所有路径中, 选择 s->dl这个 最短路径作为 s到 dl的工作路径, 以 s为根节点, dl为叶子节点, s->dl的链 路作为边, 构建的工作树如图 5A所示; 选择 s到 dl的保护路径, 由于 s->dl路径已作为工作路径, 则在剩余的 s 到 dl的路径中只有 s->d2->dl的代价最小, 所以将 s->d2->dl作为 s到 dl的 保护路径, 以 s为根节点, d2、 dl为叶子节点, s->d2、 d2->dl的链路作为边, 构建的保护树如图 5B所示;
以 dl作为起点, 选择 dl到 d2的工作路径, 由于 dl->d2的路径已作为保 护路径, 所以只有 dl->u->d2作为 dl到 d2的工作路径, 因此, s到 d2的工作 路径为 s-> dl->u->d2, 将这个工作路径加入到图 5A所示的工作树中, 构建的 新的工作树如图 5C所示, 至此, 最后构建的工作树即为图 5C所示的工作树; 选择 s到 d2的保护路径, 由于 s到 d2已存在一个保护路径 s->d2, 所以 最后构建的保护树即为图 5B所示的保护树。 为实现本发明的实施例, 可以釆取如下算法: 首先对算法中涉及的符号和术语进行定义:
G(N, L): 给定的网络物理拓朴, 其中, N为节点集, 为链路集, 其中, 每条链路为一根双向传输的光纤; W: 每根光纤上的波长资源数;
WCf. 节点 的波长转换器配置情况, 如果节点 配置波长转换器, 则 WCj 为 1 , 否则为 0;
MCji 节点 ·的分光器配置情况, 如果节点 _;配置分光器, 则 Λ ς^ Ι , 否 则为 0;
LG(N, E): 网络辅助图 ( auxiliary -graph ), 其中 N为辅助图的节点集, E 为辅助图的边集, 根据网络物理拓朴 G(N, 及节点波长转换器配置情况 WCj, 辅助图 LG(N, 构造过程如下: 将网络物理拓朴 G(N, 7 ^复制 W份, 形成一个 层的分层网络; 对于配置波长转换器的节点,用权值为 0的边连接各层对应的节点。例如, 节点 ·配置波长转换器,且 点在各层对应的节点为 f 则用权值为 0 点的边连接 ff (0<λ1≠λ2<Ψ) , 如图 6Α及 6Β所示, 图 6Α为网络物理拓朴, 每条链路有两个波长, 且网络中节点 1和节点 2配置波长转换器, 则图 6Β为 其辅助图;
{s, Dj: 组播业务请求, 其中 s为该组播业务的源节点, D为该组播业务 的目的节点集; Φ表示空集; Tw: 工作树;
Tp: 保护树;
vw: 工作树上可以分支的节点集, 可以包括两类节点, 一类是工作树上配 置有分光器的节点, 另一类是工作树上的枝叶节点;
Vp: 保护树上可以分支的节点集, 可以包括两类节点, 一类是保护树上配 置有分光器的节点, 另一类是保护树上的枝叶节点;
Nw: 工作树上的节点集。
Np: 保护树上的节点集。
D*: 目的节点/)集中没有覆盖的节点集;
Pi/. 节点 I到 j之间的最短路径; Cp: 路径;?的代价, 可以为距离或者传输单位业务量的代价。 4叚设网络釆用集中控制, 根据组播业务动态的到达或离开网络, 控制中心 维护辅助图 GN,^, 组播业务请求 , /到达网络后, 进行如下步骤:
( 1 )初始化变量, D*=D, Vw={s}, Vp={s}, Tw→, ΤΡ=Φ, Nw={s},Np={sj;
(2)如果呼叫源节点 s没有配置波长转换器, 则用权值为 0的边连接 s 点在辅助图 GN, 各层的对应节点, 否则, 转到 (3);
(3)如果/ Ζ=Φ, 算法结束, 否则, 转到 (4);
(4)计算一对最短的链路分离的最短路径对 pdm和 其中, d≡D*, m VW, n VP, 对于给定 i e/ , 计算 ί到 ^和 链路分离的一对最短路过 程如下: 计算 ί到 的最短路径 如果 ί在 7 上, 则0^ =0, 否则, 先删除 上的所有链路, 利用 Dijkstra算法计算 ί到 的最短距离 pdm 计算 ί到!^的最短路径 „, 如果 在2 上, 则0^ =0, 否则, 先删除 和 Α »上的所有链路, 利用 Dijkstra算法计算 ί到 的最短距离 Pdn',
(5)计算路径对 和 是否构成一个环, 如果构成一个环, 则会有两 种情况: 第一种情况是节点 配置有波长转换器, 第二种情况是两个路径都是 通过同一波长平面接入节点 d
( 6 )保存路径对 pd mPd n ,将 pd m和 pd n上的所有节点分别放入 Nw和 Νρ , 将 m和 上所有配置分光器的节点分别放入 ^和 ^,如果 m 构成一 个环, 则将; ¾m ;^上所有配置分光器的节点分别放入 和 (7)将 ί从 中删除, 转到 (3)。 通过上述算法, 就可以得到一个工作树及一个保护树。 上述算法借助了辅助图模型 ,这种模型可以在计算路由的同时解决波长分 配问题。但是,在网络节点和波长资源数量比较多时,辅助图的规模就比较大, 辅助图的节点数能够达到 其中 N为网络节点数, 为波长资源数。 如 果不借助辅助图模型, 则可以在网络物理拓朴上先计算路由,再进行波长分配 的方法。 如果波长分配不成功, 则放弃次路由重新计算。 与一般光网络的路由计算问题相比, WDM光网络中存在如下特殊的约束:
( 1 ) 波长连续性约束 (wavelength continuity constraint): 在该约束下, 路 径计算与波长选择密切相关,这种约束下的路由计算问题被称为 RWA( Routing and Wavelength Assignment, 路由和波长分配) 问题, 而相应的光树计算问题 称为 MC-RWA ( Multicast RWA, 组播路由和波长分配) 问题;
( 2 )分光器件稀疏配置约束: 稀疏配置是指 WDM光网络中仅有部分节 点可以充当组播树的分支节点, 或者说,仅有部分节点配置波长转换器或分光 器, 这就使得光树的计算问题无法釆用传统的基于 Steiner Tree (施泰纳树, 即 求覆盖网络节点集合的一个子集的最佳生成子树) 问题的解决方案。 上述两种约束是在构建组播路由时必须考虑的约束条件, 当然,在研究光 网络的组播保护路由问题时, 不可避免的也要考虑这些约束条件。 本发明实施例在为源节点到目的节点选择工作路径或保护路径时,尽量使 用配置有波长转换器或分光器的节点,这样可以为后续的路径选择提供更多的 选择余地, 所以本发明不但涉及到波长选择问题,还涉及到稀疏配置下的路由 计算问题。 另外,组播保护可以分为动态保护和静态保护两种。静态保护要求对给定 的一组组播请求, 求出满足这些请求的路由和相应的资源配置方法, 并且在任 一单链路失效的情况下保证所有被影响的组播连接都可以实现 100 %的恢复。 动态保护是针对一个组播请求,在建立其组播路由时预留足够的链路资源以确 保单链路失效时的 100 %恢复。 对于静态保护给定的一组组播请求, 可以先应用本发明的实施例, 进行一 个组播请求的路由计算, 再应用本发明的实施例, 进行另外一个组播请求的路 由计算, 以此类推, 直至将所有的组播请求的保护路由计算完成为止。
实施例进行了相关的仿真实验。 用于仿真的网络拓朴如图 7所示,每条边为一 根双向传输的光纤链路, 边上的数值为该条边的代价, 每条光纤链路上有 16 个波长资源。 仿真釆用动态组播业务模型, 到达网络的动态组播业务流服从泊松分布, 即, 两个前后组播业务到达网络的时间间隔服从负指数分布,每个组播业务的 持续时间按照负指数分布。组播业务到达网络后, 利用链路分离的双树保护方 法和本发明实施例提供的方法计算该组播业务的组播保护路由。如果路由算法 计算成功, 则分配网络资源, 直到呼叫结束后释放网络资源; 如果算法失败, 则呼叫阻塞。 本仿真主要是在相同网络配置和相同业务强度情况下,考察网络的阻塞概 率以评价保护方法的性能。 首先考察的是,在分光器和波长转换器配置相同、 目的节点数量不同的情 况下的网络阻塞的概率。 其中, 网络有 50 %的节点(即 7个节点)配置分光器, 50%的节点(即 7个节点)配置波长转换器。 配置分光器的节点和配置波长转换 器的节点随机分布。每一种分光器和波长转换器分别在不同目的节点数量情况 下进行 5万次呼叫, 呼叫强度为 20 Erlang, 取 5次不同分布进行阻塞概率的 平均,得到图 8所示的仿真结果。图 8中的实线表示链路分离的双树保护方法, 虚线表示本发明实施例提供的方法, 由图 8可以看出,在目的节点数量相同的 条件下,釆用本发明实施例的方法的网络阻塞的概率比链路分离的双树保护方 法的网络阻塞的概率低。 其次考察的是, 分光器数量不同情况下的网络阻塞的概率。 其中, 网络有 50%的节点配置波长转换器,分光器数量由 0到 14,波长转换器和分光器随机 分布。 每个组播呼叫的目的节点数量为 3 , 并且目的节点随机分布。 不同数量 的分光器配置在波长转换器和分光器的随机分布情况下进行 5万次呼叫 ,呼叫 强度为 20 Erlang„ 取 5次不同配置进行阻塞概率的平均, 得到图 9所示的仿 真结果。 图 9中的实线表示链路分离的双树保护方法,虚线表示本发明实施例 提供的方法, 由图 9可以看出, 在分光器数量相同的条件下, 釆用本发明实施 例的方法的网络阻塞的概率比链路分离的双树保护方法的网络阻塞的概率低。 再次考察的是, 波长转换器数量不同情况下的网络阻塞的概率。 其中, 网 络有 50%的节点配置分光器, 波长转换器的数量由 0到 14 , 波长转换器和分 光器随机分布。 每个组播呼叫的目的节点数量为 3 , 并且目的节点随机分布。 不同数量的分光器配置在波长转换器和分光器的随机分布情况下进行 5 万次 呼叫, 呼叫强度为 20 Erlang0 取 5次不同配置进行阻塞概率的平均, 得到图 10所示的仿真结果。 图 10中的实线表示链路分离的双树保护方法, 虚线表示 本发明实施例提供的方法, 由图 10可以看出, 在波长转换器数量相同的条件 下,釆用本发明实施例的方法的网络阻塞的概率比链路分离的双树保护方法的 网络阻塞的概率低。 由上述 3个仿真结果可以看出,釆用本发明实施例的方法的网络阻塞的概 率相对于链路分离的双树保护方法的网络阻塞概率明显降低。 以上所述仅是本发明的优选实施方式,应当指出,对于本技术领域的普通 技术人员来说, 在不脱离本发明原理的前提下, 还可以作出若干改进和润饰, 这些改进和润饰也应视为本发明的保护范围。

Claims

权 利 要 求
1.一种 WDM光网络中组播保护方法, 其特征在于包括: 获得源节点到一个目的节点的工作路径;获得所述工作路径对应的保护路 径;
重复上述步骤,直至获得源节点到所有目的节点的工作路径及对应的保护 路径为止。
2.如权利要求 1所述的 WDM光网络中组播保护方法, 其特征在于: 如果 之前还没有获得任何一个工作路径,则所述获得源节点到一个目的节点的工作 路径的过程为:从所述源节点到所述目的节点的侯选工作路径中选择出的第一 最短工作路径。
3.如权利要求 1所述的 WDM光网络中组播保护方法, 其特征在于: 如果 之前还没有获得任何一个保护路径,则所述获得所述工作路径对应的保护路径 的过程为:从所述源节点到所述目的节点的侯选保护路径中选择出的第一最短 保护路径。
4.如权利要求 1所述的 WDM光网络中组播保护方法, 其特征在于: 如果 已经获得至少一个工作路径,则所述获得源节点到一个目的节点的工作路径的 过程包括: 从已获得的工作路径上的所有节点到所述目的节点的侯选工作路径中选 择一个第二最短工作路径及第二最短工作路径的源节点;所述第二最短工作路 径为已获得的工作路径上的所有节点到所述目的节点代价最小的侯选工作路 径,所述第二最短工作路径的源节点是在已获得的工作路径上与所述目的节点 具有最小代价的侯选工作路径的节点; 将所述第二最短工作路径与源节点到所述第二最短工作路径的源节点的 工作路径组合作为源节点到所述目的节点的工作路径。
5.如权利要求 4所述的 WDM光网络中组播保护方法, 其特征在于: 所述 源节点和 /或第二最短工作路径的源节点配置有波长转换器和 /或分光器。
6.如权利要求 1所述的 WDM光网络中组播保护方法, 其特征在于如果已 经获得至少一个保护路径,则所述获得所述工作路径对应的保护路径的过程包 括: 从已获得的保护路径上的所有节点到所述目的节点的侯选保护路径中选 择一个第二最短保护路径及第二最短保护路径的源节点;所述第二最短保护路 径为已获得的保护路径上的所有节点到所述目的节点代价最小的侯选保护路 径,所述第二最短保护路径的源节点是在已获得的保护路径上与所述目的节点 具有最小代价的侯选保护路径的节点; 将所述第二最短保护路径与源节点到所述第二最短保护路径的源节点的 保护路径组合作为源节点到所述目的节点的保护路径。
7.如权利要求 6所述的 WDM光网络中组播保护方法, 其特征在于: 所述 源节点和 /或第二最短保护路径的源节点配置有波长转换器和 /或分光器。
8.如权利要求 1所述的 WDM光网络中组播保护方法, 其特征在于: 所述 获得所述工作路径对应的保护路径之后还包括:判断所述工作路径及对应的保 护路径是否构成一个环, 如果是, 将所述对应的保护路径作为工作路径, 将所 述工作路径作为保护路径。
9.如权利要求 1所述的 WDM光网络中组播保护方法, 其特征在于: 获得 第一个工作路径后, 进一步包括: 构建工作树, 之后每获得源节点到一个目的 节点的工作路径后, 将所获得的工作路径加入到所述工作树中。
10.如权利要求 1所述的 WDM光网络中组播保护方法, 其特征在于: 获 得第一个保护路径后, 进一步包括: 构建保护树, 获得源节点到一个目的节点 的保护路径后, 将所获得的保护路径加入到所述保护树中。
11.如权利要求 1所述的 WDM光网络中组播保护方法, 其特征在于: 所 述源节点到所有目的节点的工作路径上的链路,与源节点到所有目的节点的保 护路径上的链路分离。
12.—种 WDM光网络中组播保护装置, 其特征在于包括:
路径获得单元, 用于获得源节点到一个目的节点的工作路径, 以及所述工 作路径对应的保护路径; 完成判断单元,用于确定源节点到所有目的节点的工作路径及对应的保护 路径是否都已获得, 若都已获得则结束操作, 否则通知所述路径获得单元对组 播业务中的未得到路径的目的节点进行操作。
13.如权利要求 12所述的 WDM光网络中组播保护装置, 其特征在于: 所 述装置还包括: 环路判断单元, 用于确定所述路径获得单元获得的工作路径及 对应的保护路径是否构成环, 若是, 将所述对应的保护路径作为工作路径, 将 所述工作路径作为保护路径。
14.如权利要求 12所述的 WDM光网络中组播保护装置, 其特征在于: 所 述装置还包括: 工作树创建维护单元, 用于在所述路径获得单元获得工作路径 后, 根据所述工作路径创建工作树, 或将所述工作路径加入到所述工作树。
15.如权利要求 12或 14所述的 WDM光网络中组播保护装置, 其特征在 于: 所述装置还包括: 保护树创建维护单元, 用于在所述路径获得单元获得保 护路径后,根据所述保护路径创建保护树, 或将所述保护路径加入到所述保护 树。
16.如权利要求 12所述的 WDM光网络中组播保护装置, 其特征在于: 所 述装置位于网络的节点内, 所述节点内还包括波长转换器和 /或分光器。
PCT/CN2007/070503 2006-11-21 2007-08-16 Procédé et dispositif de protection de diffusion de groupe dans un réseau optique à multiplexage par division de longueur d'onde WO2008061461A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN2007800002880A CN101322343B (zh) 2006-11-21 2007-08-16 Wdm光网络中组播保护方法及装置
EP07800979.2A EP1993223B1 (en) 2006-11-21 2007-08-16 Method and device of group broadcast protection in wdm optical network

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CNA2006101457819A CN101192883A (zh) 2006-11-21 2006-11-21 Wdm光网络中组播保护方法
CN200610145781.9 2006-11-21

Publications (1)

Publication Number Publication Date
WO2008061461A1 true WO2008061461A1 (fr) 2008-05-29

Family

ID=39429393

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2007/070503 WO2008061461A1 (fr) 2006-11-21 2007-08-16 Procédé et dispositif de protection de diffusion de groupe dans un réseau optique à multiplexage par division de longueur d'onde

Country Status (3)

Country Link
EP (1) EP1993223B1 (zh)
CN (2) CN101192883A (zh)
WO (1) WO2008061461A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010028588A1 (zh) * 2008-09-10 2010-03-18 华为技术有限公司 组播的切换方法及系统
CN101909223A (zh) * 2009-10-27 2010-12-08 北京邮电大学 一种基于资源的wdm光网络通道保护p圈优化配置方法

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101888572B (zh) * 2009-05-11 2014-08-13 中兴通讯股份有限公司 一种wson光传送网可替代路由实现方法和系统
CN101577844B (zh) * 2009-06-05 2012-05-23 中兴通讯股份有限公司 一种波分复用网络路径搜索方法及系统
CN102104534B (zh) * 2009-12-17 2014-07-02 中兴通讯股份有限公司 组播业务保护方法及系统
CN102137026B (zh) * 2011-04-29 2013-07-24 东北大学 一种wdm光网络中的多约束多播路由方法
CN102186123B (zh) * 2011-04-29 2014-01-22 东北大学 一种wdm光网络中的基于子树的多播共享多层保护方法
CN103023665B (zh) * 2011-09-23 2015-12-16 华为技术有限公司 一种组播业务保护的方法、网络设备和系统
CN108494475B (zh) * 2018-03-09 2020-01-10 苏州大学 量子通信网络中的冗余路径的设置和工作方法
CN113472646B (zh) * 2021-05-31 2023-02-10 华为技术有限公司 一种数据传输方法、节点、网络管理器及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020163685A1 (en) * 2001-05-01 2002-11-07 Swanson Eric Arthur System and method for routing working and protect paths
CN1472968A (zh) * 2002-07-31 2004-02-04 华为技术有限公司 光网络中端到端保护路径的获取方法
CN1588890A (zh) * 2004-10-08 2005-03-02 烽火通信科技股份有限公司 Ason网络中控制平面参与保护倒换的方法
CN1773947A (zh) * 2004-11-09 2006-05-17 中兴通讯股份有限公司 智能光网络中实现光组播的方法
CN1790960A (zh) * 2005-12-06 2006-06-21 电子科技大学 一种wdm网中混合共享链路波长资源的通路保护方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE10237584B4 (de) * 2002-08-16 2005-10-06 Siemens Ag Verfahren zur Verwaltung von Ressourcen beim Aufbau eines Ersatzpfades in einem transparent schaltbaren Netzwerk

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020163685A1 (en) * 2001-05-01 2002-11-07 Swanson Eric Arthur System and method for routing working and protect paths
CN1472968A (zh) * 2002-07-31 2004-02-04 华为技术有限公司 光网络中端到端保护路径的获取方法
CN1588890A (zh) * 2004-10-08 2005-03-02 烽火通信科技股份有限公司 Ason网络中控制平面参与保护倒换的方法
CN1773947A (zh) * 2004-11-09 2006-05-17 中兴通讯股份有限公司 智能光网络中实现光组播的方法
CN1790960A (zh) * 2005-12-06 2006-06-21 电子科技大学 一种wdm网中混合共享链路波长资源的通路保护方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP1993223A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010028588A1 (zh) * 2008-09-10 2010-03-18 华为技术有限公司 组播的切换方法及系统
CN101674222B (zh) * 2008-09-10 2013-01-30 华为技术有限公司 组播的切换方法及系统
CN101909223A (zh) * 2009-10-27 2010-12-08 北京邮电大学 一种基于资源的wdm光网络通道保护p圈优化配置方法

Also Published As

Publication number Publication date
EP1993223B1 (en) 2016-05-04
EP1993223A4 (en) 2013-03-20
CN101192883A (zh) 2008-06-04
EP1993223A1 (en) 2008-11-19
CN101322343B (zh) 2011-06-08
CN101322343A (zh) 2008-12-10

Similar Documents

Publication Publication Date Title
WO2008061461A1 (fr) Procédé et dispositif de protection de diffusion de groupe dans un réseau optique à multiplexage par division de longueur d&#39;onde
EP0752795B1 (en) Link reservation in communications networks
JP2006527543A (ja) 光ネットワーク・トポロジ・データベースおよび光ネットワーク・オペレーション
WO2014005522A1 (zh) 路由计算方法及装置
Palmieri A reliability and latency-aware routing framework for 5G transport infrastructures
Khalil et al. Dynamic provisioning of low-speed unicast/multicast traffic demands in mesh-based WDM optical networks
US20080205262A1 (en) Node controller and node system
Assi et al. Integrated routing algorithms for provisioning “sub-wavelength” connections in IP-over-WDM networks
Babarczi et al. Optimal dedicated protection approach to shared risk link group failures using network coding
US7706383B2 (en) Minimum contention distributed wavelength assignment in optical transport networks
US9451342B2 (en) Method and apparatus for managing link on multi-layer networks
Chitra et al. Hidden Markov model based lightpath establishment technique for improving QoS in optical WDM networks
Lu et al. Combining electronic layer protection and pre-planned optical restoration for improved and resource-efficient reliability
Xu et al. Multicarrier-collaboration-based emergency packet transport network construction in disaster recovery
Chu et al. Lightpath rerouting in wavelength-routed WDM networks
Vijaya Saradhi et al. A framework for differentiated survivable optical virtual private networks
Zheng et al. Dynamic path restoration based on multi-initiation for GMPLS-based WDM networks
Liu et al. Dynamic call and connection admission control in automatically switched optical network for grid computing applications
Correia et al. Protection schemes for IP-over-WDM networks: Throughput and recovery time comparison
Yu et al. Connection level segment shared protection for dynamic multicast traffic grooming
Saradhi et al. iOPEN: integrated optical Ethernet network for efficient dynamically reconfigurable service provisioning
Zhang et al. Research and simulation of ASON survivability testbed
Lu et al. A distributed signaling scheme for provisioning dynamic traffic in wavelength-routed networks
Zhao et al. Dynamic lightpath establishment mechanisms based on automatically switched optical network (ASON)
Ghimire et al. Design and analysis of protocols for qos and autonomous recovery in gmpls controlled ip over wdm networks

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200780000288.0

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 07800979

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2007800979

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE