IPv4／IPv6转换网关的设计与实现
下一代移动互联网IPV内容转换网关项目方案建议书
下一代移动互联网IPV6内容转换网关项目方案建议书项目简介(一)项目背景IPv6的部署要经历一个渐进的过程。
IPv6提供很多过渡技术来实现这个渐进过程。
传统系统的坚固性被高估了。
很长时间内，IPv4仍将存在，即使一些或连网世界的其余部分已升级为IPv6。
到那时，升级系统将需要保持与IPv4系统的互操作能力。
随着时间的推移，互操作的负担将由早期的实现者承担转为由传统系统的维护者来承担。
任何情况下，同时支持IPv4和IPv6的系统都是必要的。
所以，需要一个能支持IPv4-IPv6转换的代理服务器。
(二)项目需求IPV4/IPV6 HTTP协议转换功能要求IPV6网络中的IPV6终端可以访问IPV4网络中的IPV4网页。
有如下功能需求:1．导航WEB server：（1）要求导航页面包括各热门门户网站的链接，以及搜索栏（用于用户键入导航网页上未列出的网站URL）。
（2）要求具备内容重定向功能。
要求对于已有的门户网站链接，点击该链接后指向门户网站URL，其中ipv6*是http IPV4/V6协议转换服务器群中某一台服务器的ipv6地址（例如：点击的链接，该链接指向）。
要求对于在该导航网页上用户自己键入的网址，该WEB server需自动将其指向HTTP IPV4/IPV6内容转换网关。
2．H TTP IPV4/IPV6协议转换网关：（1）要求该网关为双栈，可实现IPV4/IPV6互通。
（2）要求网关从通过导航门户改写过的http请求中的目的IPV4 URL提取出来，并向该目的服务器索取网页；（3）要求网关将索取的网页的URL以及网页内容中所包含的URL根据流量均衡的需要动态的加入网关群中的某一台网关的IPV6地址，将此URL改写成返回给用户PC；具体实现流程如下图所示：ICP Server HTTP IPV4/IPV6协议转换代理网关服务器群网络加速功能1． 内容压缩功能：要求可结合移动互联网信道特点设计压缩算法，实现移动互联网的通信的高效传输，包括图像、文本、文件、邮件、FTP 应用的压缩。
IPv4-IPv6组播过渡网关的研究与实现
IPv4-IPv6组播过渡网关的研究与实现作为下一代互联网的主要协议,IPv6受到越来越多的关注。
国内CERNET2 IPv6主干网的开通,更将IPv6的发展推至一个崭新的阶段。
但从IPv4完全过渡到IPv6仍需要相当长的时间,在这漫长的过渡期中,需要纯IPv4站点和纯IPv6站点之间能够直接通信。
目前,有很多有用的过渡技术,但绝大多数是关于单播的。
组播作为IPv6必须支持的特性之一,同样存在着过渡问题。
成功用于单播的协议转换技术,并不完全适用于组播,因为单播采用的是一对一的通信方式,而组播采用一对多的通信方式。
国际上已经有相关组织探讨了组播过渡的问题,国内也有少数科研工作者开始关注这方面的研究,但目前都未提供可用于实际网络的具体解决方案。
本论文研究了几种具有代表性的现有组播过渡技术,分析了这些技术的工作原理、优缺点和各自的适用性。
在此基础上,针对IPv4组播域和IPv6组播域中主机相互进行组播通信的问题,论文提出了一个基于协议转换技术的组播过渡模型——MTG(Multicast Transition Gateway)。
MTG是一个网关式系统,它位于IPv4组播网络和IPv6组播网络的边界。
MTG 针对常用的组播通信需求,支持多种部署方案,采用可管理可扩展的设计原则。
MTG针对组播视频会议和组播视频分发两种主要组播应用,利用灵活可靠的跨域(IPv4组播域和IPv6组播域)组播交互机制,较好地解决了一对多、多对多两种组播模型的过渡问题。
论文在基于网关技术(网络层的组播协议转换技术)的基础上,设计了简单、清洁、高效、可管理和可扩展的MTG模型。
本文根据组播地址映射是否随时间变化,提出了静态组播地址映射和动态组播地址映射的概念。
既可以实现等量组地址空间的映射,也能完成大组播地址空间到小组播地址空间的映射。
本文在深入分析Linux内核对组播支持的基础上,提出使用可装载的模块编程技术,利用Linux内核防火墙netfilter框架,在内核中网络层报文处理的入口对组播报文进行处理。
移动IPv4 v6转换网关设计方案说明书
Mobile IPv4/v6 translation gatewayZheng Xiang College of Medical Information Engineering Guangdong Pharmaceutical CollegeGuangzhou, ChinaZhengming MaCollege of Information Science and TechnologySun Yat-sen UniversityGuangzhou, ChinaAbstract—IETF has specified Mobile IPv4 and Mobile IPv6 in RFC3344 and RFC3775 respectively, but not yet discussed Mobile IPv4/v6 in any published RFC. This paper proposes a scheme to solve one of Mobile IPv4/v6 problems which Home Agent (HA) is located in IPv4 network, Correspondent Node (CN) is located in IPv6 network, while MN moves within IPv6 network. In the solution, a gateway called Mobile IPv4/v6 translation gateway (MIPv4/v6-TG) is introduced to bridge between IPv4 network and IPv6 network, which is made up of a traditional NAT-PT gateway and a Mobile IP application level gateway (MIP-ALG) built upon the NAT-PT gateway. MIP-ALG maintains a MIP table, a data structure, which is formed by entries. And it can work compatibly with RFC3344 and RFC3775.Keywords—MobileIP(MIP), NAT-PT, Mobile IP application level gateway (MIP-ALG), Mobile IPv4/v6 translation gateway (MIPv4/v6-TG), MIP TableI.INTRODUCTIONIETF specified the latest version of Mobile IPv4 [1] and Mobile IPv6[2] in RFC3344 and RFC3775 respectively. These two protocols work very well in pure IPv4 network and pure IPv6 network. However, both of them can not be directly applied to IPv4/v6 mixed networks. As the current network evolves gradually from IPv4 to IPv6, more and more mobile nodes need to roam in IPv4/v6 mixed networks. So we need to develop a transition scheme to support Mobile IP in IPv4/v6 mixed networks.As a transition scheme, the solution should be developed based on the transition mechanisms for IP. Currently, three transition ways are recommended by IETF, namely, dual stacks (RFC4213) [3], tunneling (RFC3053) [4], and NAT-PT (RFC2766) [5][6]. There have been a lot of studies on Mobile IP in IPv4/v6 mixed networks, many of which are based on NAT-PT. An Internet Draft published by IETF [7] presents a solution to a situation of Mobile IPv4/v6, in which Home Agent (HA) and Mobile Node (MN) are in IPv6 network, Correspondent Node (CN) is in IPv4 network and NAT-PT gateway is located between IPv4 network and IPv6 network. It is proposed in this draft that NAT-PT do all of MIPv6 functionalilties on behalf of CN. This draft, however, deals with only part of the issues involved in Mobile IPv4/v6. In addition, this draft does not make clear how NAT-PT gateway does the functionalities on behalf of CN.[8] studies the situation where both HA and CN are in IPv6 network and MN is in IPv4 network. A border router is placed between IPv4 network and IPv6 network to connect the two networks. After acquiring a new Care-of Address (CoA) in the IPv4 network, MN performs two registration processes, one is with the border router, and the other is with HAv6. Packets sent from CN to MN are encapsulated in new IPv6 packets which will be routed to the home network. In the home network, HAv6 intercepts these packets and again, encapsulates them in new IPv6 packets and then sends them to the border router. When these packets arrive at the border router, the border router encapsulates them in IPv4 packets and sends them to MN through a tunnel.[9] studies the situation where HA is in IPv4 network, CN is in IPv6 network and MN moves within IPv4 network.A NAT-PT gateway is used to connect the two networks. In this scheme, MN has to register its new CoA with both HA and CN. Since CN knows exactly the location of MN, it can send packets directly to MN, without passing through the home network. In this way, the triangle routing in RFC3344 can be avoided.In [10], if HA is in IPv6 network, MN always registers its new CoA with HA by BU messages. If MN moves to IPv4 network, the BU messages sent by MN will be tunneled to NAT-PT gateway which decapsulates the messages and sends them to HA. Similarly, if HA is in IPv4 network, MN always registers its new CoA with HA by register request messages. If MN moves to IPv6 network, the register request messages sent by MN will be tunneled to NAT-PT gateway which decapsulates the messages and sends them to HA.Mobile IPv4/v6 should be compatible with Mobile IPv4 and Mobile IPv6 because it is a transition scheme from Mobile IPv4 to Mobile IPv6. RFC3344 and RFC3775 are the schemes to MIPv4 and MIPv6 recommended by IETF. However, the solutions proposed in [8][9][10] do not pay enough attention to the compatibility with RFC3344 and RFC3775. This would prevent them from being chosen as a feasible transition scheme. This paper proposes a scheme to solve Mobile IPv4/v6 problems. The solution is based on Mobile IPv4/v6 Translation Gateway (MIPv4/v6-TG) which is made up of NAT-PT and a Mobile IP-Application Level Gateway (MIP-ALG). In the following sections, we will first describe the fundamental principles of our solution. Then we will introduce the data structure, MIP table, which is maintained by MIP-ALG, and describe in details how to create, use and update MIP table entries.II. T ERMINOLOGYThe terms frequently used in this paper are listed in Table 1.In addition, “v4” and “v6” are used to indicate IPv4 and IPv6, respectively. For example, HAv6 stands for HA that is located in IPv6 domain and CNAv6 the IPv6 address of CN. By the way, if an IPv4 address is marked with a “#”, it stands for an address from the address pool of NAT-PT gateway and is used to map an IPv6 address. If an IPv6 address is marked with a “*”, it means that this address is made up of a 96-bit NAT-PT prefix and an IPv4 address.TABLE I.T ERMS THAT ARE FREQUENTLY USEDTerms Meaning TermsMeaningMIP Mobile IP CNACorrespondent NodeAddressHA Home Agent BU Binding Update MN Mobile Node BA Binding Acknowledge CN Correspondent Node HoTI Home Test Init HAAHome Agent AddressCoTICare-of Test Init HoA Home Address HoT Home Test CoA Care-of Address CoTCare-of TestIII. MIP V 4/V 6-TGMIPv4/v6-TG is made up of NAT-PT and MIP-ALG. On IPv6 network side, MIPv4/v6-TG acts as one of the Mobile IPv6 entities to combine with other MIPv6 entities inside the IPv6 network to form a complete MIPv6 model described in RFC3775. Inside the IPv6 network, the registration process and communication process can be performed as specified in RFC3775. Similarly, on IPv4 network side, MIPv4/v6-TG acts as one of the Mobile IPv4 entities to combine with other MIPv4 entities inside the IPv4 network to form a complete MIPv4 model described in RFC3344. Inside the IPv4 network, the registration process and communication process can be performed as specified in RFC3344.As mentioned above, MIPv4/v6-TG will act as different MIP entities according to different combinations of the IP versions of HA, MN and CN. However, it can not decide which role it should act, if it does not know the IP versions of the three entities when MIPv4/v6-TG intercepts a MIP-related message or datagram. Therefore, MIPv4/v6-TG should keep the information of the IP versions of the three entities and the bindings that are used when it acts as a particular MIP entity. In our solution, we introduce a new data structure called MIP table to solve this problem. We use 3-bit binary numbers to indicate the location where MN, HA and CN sit. If the communication entity location in the IPv6 network we record it as 1, otherwise we record it as 0.The three MIP entities have eight kinds of IP version combinations, from 000 to 111. At any time, there may be many MIP sessions, which correspond to the same scenario and thus use the same kind of MIP table entry. MIPv4/v6-TG should be able to distinguish them so that the messagesand datagrams can be sent to the right place. Therefore, An MIP table entry should, typically, have the following fields. Type: A three-bit field indicating the IP versions of HA, MN and CN respectively. For each bit, a value of 1 indicates the MIP entity is located in IPv6 network, while a value of 0 indicates the MIP entity is located in IPv4 network. Type value varies from 000 to 111.MIPv6 Message Entrance: A 128-bit field through which a particular MIP table entry can be found and accessed when a MIPv6 message is intercepted by MIPv4/v6-TG. Usually, this field is set to the IPv6 home address.MIPv6 Datagram Entrance: A 128-bit field through which a particular entry can be found and accessed when a MIPv6 datagram is intercepted by MIPv4/v6-TG. If an intercepted packet is not a MIPv6 message, MIPv4/v6-TG will take out its destination address and use the address as an index to search the MIP table.MIPv4 Message Entrance: A 32-bit field through which a particular entry can be found and accessed when a MIPv4 message is intercepted by MIPv4/v6-TG. MIPv4 Datagram Entrance: A 32-bit field through which a particular entry can be found and accessed when a MIPv4 datagram is intercepted by MIPv4/v6-TG.Cached Bindings: Bindings of home address and care-of address that are used by MIPv4/v6-TG when it acts as a particular MIP entity. Bindings may be of IPv4 form or of IPv6 form. Entries may have no binding, one binding, or two bindings. This depends on the types of the entries.Source Port: A 16-bit field that records the source port value of Registration Request message, extended Registration Request message, or Agent Request message. All of these three messages are sent through UDP ports.Destination Port: A 16-bit field that records the destination port value of Registration Request message, extended Registration Request message, or Agent Request message.State: A 1-bit field indicating whether the entry is completed or not. There are two possible states for each entry: 1 (finished) and 0 (unfinished). An unfinished state indicates that the entry is now being created or updated. Entries with an unfinished state can not be used by MIPv4/v6-TG as a guide to process MIP datagrams.Lifetime: A 16-bit field indicating entry lifetime. The meaning of this field depends on the value of the State field.IV. A PPLICATION E XAMPLEWe will solve one of Mobile IPv4/v6 problems which Home Agent (HA) is located in IPv4 network, Correspondent Node (CN) is located in IPv6 network, while MN moves within IPv6 network. And we will describe the creation, update and usage of type 011 MIP table entries as an example in the following.A. The create of Type 011 MIP Table EntriesWhen MIPv4/v6-TG intercepts a BU message, it takes out HoAv6* from the message and uses it as an index to search the MIP table. If no matching entry is found, MIPv4/v6-TG will create a new entry, using the information carried in the BU message. If a matching entry is found,MIPv4/v6-TG will go further to see if the second bit of the Type is equal to 1 (i.e., MN is in IPv6 network). If so, MIP-ALG will just update the existing entry, without creating a new entry. Otherwise, MIP-ALG will create a new entry. The creation process of the 011 type MIP table entries isFigure 1. The creation process of the 011 type MIP table entriesThe concrete steps as follows:(1) On receiving the BU message, MIP-ALG can infer that HA is located in IPv4 network and MN is located in IPv6 network. However, MIP-ALG does not know the IP version of CN and needs other means to learn this information later.MIP-ALG creates a new entry and sets the field of Cached Bindings to HoAv6*↔CoAv6, MIPv6 Message Entrance to HoAv6*, MIPv4 Message Entrance to CoAv4#. CoAv4# is an IPv4 address taken from the NAT-PT address pool and mapped with CoAv6. The fields of HoTI/HoT, CoTI/CoT and State are all set to 0. Lifetime is set to a value in which the creation of the entry must be accomplished.(2) MIP-ALG translates the BU message into a Registration Request message with CoAv4# and HAAv4 as its source address and destination address, respectively. HoAv4 is carried in the new message. HAAv4 and HoAv4 are acquired from HAAv6* and HoAv6*, respectively. This Registration Request message will be routed to HAv4.(3) MIPv4/v6-TG intercepts the Registration Reply message replied from HAv4, takes out the destination address as an index to search the MIP table and finds this entry.(4) MIP-ALG translates the intercepted Registration Reply message into a BA message with HAAv6* and CoAv6 as its source address and destination address, respectively. HoAv6* is carried in the message. HAAv6* and HoAv6* are acquired by adding a 96-bit NAT-PT prefix to HAAv4 and HoAv4. CoAv6 is acquired by searching the NAT table, using CoAv4# as the index. HAAv4, HoAv4, and CoAv6 can be all acquired from the intercepted Registration Reply message. This BA message will be routed to MNv6.(5) MIPv4/v6-TG intercepts HoTI message, takes out HoAv6* from the intercepted message as an index to search the MIP table and finds this entry. Note that HoTI arrives through tunneling.(6) MIP-ALG can judge the IP version of CN by the destination address of HoTI. MIP-ALG sets the field of Type to 011, and then sends HoTI to CNv6.(7) MIPv4/v6-TG intercepts HoT message replied from CNv6, takes out HoAv6* from the intercepted message as an index to search the MIP table and finds this entry.(8) MIP-ALG sets the HoTI/HoT field to 1, and then sends HoT to MNv6. Note that the HoT message is sent through tunneling.(9) MIP-ALG sets the State to 1, and then sets the Lifetime to the lifetime of the binding.B.The update of Type 011MIP Table EntriesWhen MN moves within network of the same IP version and acquires a new care-of address, it will update the binding caches on HA and CN (when CN is in IPv6 network) as well as the binding caches on the related MIP table entry under some circumstances. Like the creation of MIP table entries, the update of the entry is triggered by MIP messages, such as Registration Request messages, BU messages, HoTI/HoT messages and CoTI/CoT messages. In the creation process, MIPv4/MIPv6 Message Entrance of the entry have been set. MIPv4/v6-TG can access a corresponding entry through these entrances when it intercepts a MIP message, and then updates the entry.Note that when MN moves to a network of a different IP version, the original entry (if any) becomes invalid and a new MIP table entry should be created.A Type 011 entry corresponds to a scenario where HA is located in IPv4 network, while MN and CN are located in IPv6 network. The update process of a Type 011 MIP tableFigure 2. The update process of the 011 type MIP table entriesThe concrete steps as follows:(1) MIPv4/v6-TG intercepts a BU message, takes out HoAv6* and uses it as an index to search the MIP table, and finds a related Type 011 entry.(2) MIP-ALG sets the fields of State to 0, and Lifetime to a value in which the update of this entry must be accomplished.(3) MIP-ALG takes CoAv4# from MIPv4 Message Entrance and uses it as an index to search the NAT table. A mapping of CoAv4#↔CoAv6 will be found. Then MIP-ALG updates the mapping with the new CoAv6 carried in the BU message.(4)MIP-ALG updates Cached Bindings HoAv6*↔CoAv6 with the new CoAv6 carried in the BU message.(5) MIP-ALG generates a BA message. The source address and destination address are copied from the destination address and source address of the BU message. HoAv6* is also carried in the BA message. This BA message will be routed to MNv6.(6) MIPv4/v6-TG intercepts a HoTI message, takes out HoAv6* from the intercepted message as an index to searchthe MIP table and finds this entry. This HoTI message willbe then delivered to CNv6.(7) MIPv4/v6-TG intercepts a HoT message, takes out HoAv6* from the intercepted message as an index to searchthe MIP table and finds this entry. This HoT message will be routed to MNv6 through tunneling. The beginning point and endpoint of the tunnel are HAAv6* and CoAv6. CoAv6 canbe acquired from the Cached Bindings field of the entry.(8) MIP-ALG sets the field of State to 1, and the Lifetime to the lifetime of the binding.C.The usage of Type 011MIP Table EntriesThe introduction of MIP table aims to maintain MIP sessions in IPv4/v6 mixed networks. When a datagram sentby MN or CN passes through MIPv4/v6-TG, MIPv4/v6-TGwill take out the destination address of the datagram and uses it as an index to search the MIP table. If a matching entry is found, MIPv4/v6-TG will process the datagram, based on the information recorded in the entry.A Type 011 entry corresponds to a scenario where bothMN and CN are located in IPv6 network. In this scenario,MN and CN can communicate with each other without the participation of MIPv4/v6-TG. Therefore, Type 011entrieswill not be used in the communication processes.V.C ONCLUSIONThe key to our solution is Mobile IPv4/v6 translation gateway and a MIP table, a newly introduced data structure. With the help of this gateway and the MIP table, RFC3344and RFC3775 can be reused in IPv4 network and IPv6 network respectively. In this way, the Mobile IP entities inIPv4 network and the Mobile IP entities in IPv6 network canbe transparent to each other.Compared with other solutions, our solution has three main advantages. Firstly, it can work compatibly with RFC3344 and RFC3775. This is very important in that it makes Mobile IP in IPv4/v6 mixed networks possible without any update to the existing networks. Secondly, our solution introduces MIP table. We can use the MIP table to realize the communication in IPv4/v6 mixed network easily. Thirdly, the creation, usage and update of the MIP table entries is easily too.A CKNOWLEDGMENTThis work is supported in part by The Medium and Small Enterprise Technology Innovation Fund of The Ministry of Science and Technology (No. 09C26214402150), China. R EFERENCES[1]Charles E. Perkins, Editor, “IP Mobility Support for IPv4”, RFC3344,IETF, August 2002[2]Charles E. Perkins, Editor, “Mobility Support in IPv6”, RFC3775,IETF, June 2004[3] E. Nordmark, “Basic Transition Mechanisms for IPv6 Hosts andRouters”, RFC4219, IETF, October 2005[4] A. Durand, “IPv6 Tunnel Broker”, RFC3053, IETF, January 2001[5] E. Nordmark, “Stateless IP/ICMP Translation Algorithm (SIIT)”,RFC2765, IETF, February 2000[6]G. Tsirtsis, P. Srisuresh, “Network Address Translation-ProtocolTranslation (NAT-PT)”, RFC2776, IETF, February 2000[7]Joo- Chul Lee, “ Considerations for Mobility Support in NAT-PT”,Internet Draft, IETF, June 2005[8]Changwen Liu, “Support mobile IPv6 in IPv4 domains”, VehicularTechnology Conference, 2004. VTC 2004-Spring. 2004 IEEE 59th [9]Dae Sun Kim; Choong Seon Hong; Suda, T, “A Terminal MobilityManagement Architecture for IPv4 and IPv6 Environments”, Broadband Convergence Networks, 2006. BcN 2006. The 1st International Workshop on[10]Hyun-Ho Choi; Dong-Ho Cho, “Mobility management based onmobile IP in mixed IPv4/IPv6 networks”, Vehicular Technology Conference, 2003. VTC 2003-Fall. 2003 IEEE 58th。
基于IPv6IPv4双协议栈模式的网络设计与实现
43农业信息化基于IPv6/IPv4双协议栈模式的网络设计与实现引言2012年2月，科技部、中组部、工信部批复河南省开展国家农村信息化示范省建设试点工作。
农村信息服务综合平台是示范省项目的重要环节，农村信息服务综合平台是全省农村信息资源统一发布和展示的窗口。
互联网络是综合平台的信息基础设施，向IPv6网络过渡是互联网发展的必然趋势，是解决IP地址枯竭、分配不均和骨干网路由表膨胀等阻碍互联网发展问题的关键技术。
建设农村信息服务综合平台的基础网络时，考虑兼容性前瞻性，提供IPv4和IPv6共存的网络环境，为各种应用系统和服务提供双协议支撑平台，保障农村信息服务综合平台安全可靠运行。
1 农村信息服务综合平台网络建设原则农村信息服务综合平台依托河南农业大学信息与管理学院，已经实现了IPv4网络的全覆盖，拥有用户主机近800台，是河南农业大学校园网的组成部分。
在已有网络基础上建设平台网络，建设目标如下：（1）按照CNGI示范网络驻地网建设总体需要和规定技术要求，建设IPv6网络，兼容IPv4，主机数达到600个以上。
（2）驻地网使用IPv6协议接入CNGI国家主干车银超 余 华 郑 光 孙昌霞（河南农业大学信息与管理科学学院，河南 郑州 450002）网。
（3）为接入用户提供技术支持，使用户主机灵活接入CNGI。
（4）为CNGI开展各项关键技术的研究、试验、产业化及农业信息化提供良好的试验环境和技术支撑。
（5）在应用方面，建设必要的IPv6内容资源，部署基于IPv6的DNS、WWW、FTP等服务，供用户使用。
在选择综合组网技术时，要重点考虑下列因素：（1）兼容性。
在建设IPv6网络时，不影响已有IPv4网络的运行，兼顾原有的IPv4应用和资源。
（2）扩展性。
组网技术能支持网络平滑升级到IPv6，能为将来更进一步的扩展、扩容与升级预留充分的空间。
（3）性能与安全。
在带宽、易用、安全、可靠、管理等方面应比原有网络有提升。
IPv4向IPv6过渡方案设计与实现 文献综述
IPv4向IPv6过渡方案设计与实现+文献综述4.1 总结 114.2 展望 12参考文献 12致谢 13IPv4向IPv6过渡方案设计与实现引言基于IPv4协议的互联网的发展极大地促进了社会和科技的进步，随着互联网技术的不断发展,网络用户规模设备急剧增长,对IPv4地址的需求量愈为急切，IPv4地址所能维持的时间越来越短。
XX年2月3日,国际互联网协会ICANN 官方宣布: 全球最后一部分IPv4地址分配完毕，IPv4地址空间已尽【1】。
此外，除了地址匮乏，IPv4存在网络安全隐患、不提供服务质量保证、IP地址配置复杂、缺少移动性支持等诸多问题，为了彻底解决这些问题，下一代互联网采用IPv6来代替IPv4被已成为业界共识。
【2】1.研究现状及意义1.1研究现状IPv6取代IPv4成为下一代互联网的核心已成为历史潮流，然而当前的网络设备大部分依然是基于IPv4的，不可能将它们在短时间内一下子都过渡到基于IPv6的设备，在很长的一段时期内，将是IPv6网络和IPv4网络共存的阶段。
如何平稳顺利地过渡到IPv6网络成为业界研究的热点，IETF 组织发布了多种过渡技术标准，主流方案可分为3类：双栈技术、翻译技术、隧道技术。
1.2研究意义鉴于纵横网络多年的IPv4网络规模极其庞大，IPv6过渡将采取增量部署、循序渐进的方式进行实施。
对企业来说，过渡过程中发生的任何问题都可能造成巨大的经济损失，它们更希望IPv4向IPv6过渡平稳顺利地进行。
因此，研究各种IPv6过渡技术及其相对的优缺点、应用场景是很有必要的，可以帮助运营商或企业等制定完善的过渡规划并顺利的实施。
源自/六"维:论'文;网(加7位QQ3249"1142.IPv4/IPv6过渡技术分析2.1 双栈技术双栈技术是一种最简单的直接的过渡机制，是所有过渡技术的基础；是指设备同时运行IPv4和IPv6两种协议的技术实现方式，当双协议栈节点配置IPv4和IPv6地址后，就可以在相应接口上转发IPv4和IPv6报文，双协议栈技术适合IPv4节点之间或IPv6网络节点之间的通信。
IPV6转换网关的设计与实现的开题报告
跨越IPV4/IPV6转换网关的设计与实现的开题报告一、选题背景当前，全球互联网的发展已经进入到IPv6时代，但IPv4地址的需求仍然非常庞大。
在IPv6推广过程中需要考虑IPv4的兼容性，因此需要在IPv4和IPv6之间建立一种有效的网络转换机制，以提供当前互联网应用程序所必需的跨网络能力。
IPV4/IPV6转换网关技术就是为这一目的而存在的。
IPv4和IPv6之间的转换是指处理IPv4地址和IPv6地址之间的映射问题。
目前，常见的IPv4/IPV6转换方式包括Dual-Stack技术、IPv6 over IPv4隧道技术和NAT64/DNS64转换技术。
其中，NAT64/DNS64技术是逐渐被广泛关注和采用的技术，原因在于它可以实现IPv4/IPv6网络的透明互联，保证了各种互联网应用程序的无缝衔接。
二、选题目的本文主要针对IPv4/IPv6转换中的NAT64/DNS64技术，考虑跨越IPV4/IPV6转换网关的设计与实现，以提升互联网应用程序的网络适应性和扩展性。
三、研究内容1. IPv4/IPv6转换技术的理论基础介绍IPv4/IPv6转换技术和NAT64/DNS64技术的基本原理、特点和应用场景，对跨越IPV4/IPV6转换网关的设计与实现提供理论支撑。
2. NAT64/DNS64技术的架构设计针对IPv4/IPv6转换中的NAT64/DNS64技术，研究并设计NAT64的网络架构、其实现原理、过程及其与DNS64的协作，实现网络传输的无缝转换。
3. 跨越IPv4/IPv6转换网关的实现利用Linux虚拟机实现NAT64/DNS64技术，并实现跨越IPv4/IPv6转换网关的设计与实现，在真实网络环境中进行网络测试，验证设计方案的合理性和可行性。
四、研究意义本研究将跨越IPv4/IPv6转换网关的设计与实现贯穿于整个研究过程，强调理论和实践的结合，对IPv4/IPv6转换技术及其应用实现有着重要的推动作用；同时，此项研究结果也为实现网络的跨越、网络的互通性等方面提供一定的理论指导和技术支持，具有重要的应用价值。
网络规划设计中的IPv4与IPv6过渡方案(四)
网络规划设计中的IPv4与IPv6过渡方案随着互联网的快速发展，IPv4（Internet Protocol version 4）已经接近资源耗尽的状态。
为了解决IPv4地址短缺的问题，IPv6（Internet Protocol version 6）被引入，并成为了下一代互联网协议的标准。
然而，由于现有网络基础设施主要依赖IPv4，IPv4与IPv6的共存与过渡成为了网络规划设计中的一项重要任务。
一、IPv4与IPv6的区别IPv4是目前广泛应用的互联网协议版本，它采用32位地址格式，提供约43亿个可用地址。
然而，随着全球互联网用户数量的激增，IPv4的地址资源正在日益枯竭。
IPv6利用128位地址格式，提供了可观的地址空间，其中每个用户可以获得大量的地址，从而解决了地址短缺的问题。
此外，IPv6还提供了更好的包头结构以及支持高效路由和安全性等特性。
二、IPv4与IPv6的共存方案为了让现有的IPv4网络能够逐步过渡到IPv6网络，一些共存方案被提出。
1.双栈（Dual-Stack）方案：这是最常用的方案之一，它要求网络中同时支持IPv4和IPv6协议栈。
每个主机都同时拥有一个IPv4地址和一个IPv6地址，实现了IPv4与IPv6的共存。
通过双栈路由器的存在，IPv4和IPv6之间的通信可以相互转换和交互。
2.隧道（Tunneling）方案：在此方案中，IPv6数据报通过IPv4网络进行传输。
IPv4网络扮演着隧道的角色，将IPv6数据报包装在IPv4数据包中进行传输。
这种方式可以在现有的IPv4网络上快速部署IPv6服务，但需要在隧道终点进行IPv6数据报的解封装。
3.转换（Translation）方案：转换是一种将IPv4地址和IPv6地址进行转换的方式。
它可以实现IPv4与IPv6之间的互通性，是一种比较灵活的过渡方案。
在转换过程中，可能需要进行地址格式和协议头的修改，以实现数据的转换和交换。
IPv4_IPv6转换网关·····
IPV4/IPV6转换网关的研究与设计摘要：随着计算机网络应用的飞速进步，现有的IP通信协议(IPv4协议)已展现出众多的问题，如不能适应新的网络应用、地址资源即将耗尽以及对安全性无法保证等。
IPv6是继IPv4后出现的新一代通信协议，它的出现为互联网的发展带来了新契机。
IPv6的众多优势成为取代IPv4必然的发展。
本文从IPv6协议本身出发，阐述了IPv6协议及其与IPv4协议的比较，对目前现有的几种IPv4/IPv6过渡技术进行了讨论，分析了它们的特点，设计了转换网管，并通过测试证明转换方法的可用性。
关健词：IPV4、IPV6、NAT-PT、协议栈转换Abstract: With the rapid progress of computer network applications, the existing IP communication protocol (IPv4 protocol) has shown many problems, such as the application can not adapt to the new network, the address of resources running out, and so on can not guarantee security. After following the IPv4 IPv6 is a new generation of communication protocol, it was a development of the Internet brought new opportunities. The many advantages of IPv6 to replace IPv4 as an inevitable development. This article from the IPv6 protocol itself, and expounds the IPv6 protocol and its comparison with the IPv4 protocol, several of the currently available IPv4/IPv6 transition technology is discussed, analyzed their characteristics, the design of the conversion network, and tested the availability of proof conversion method.Key words: IPV4, IPV6, NAT-PT, protocol stack conversion一、引言随着Internet的迅猛发展,现有的IP协议(IPV4协议)在应用中暴露出了出现了许多问题,如不能适应新的网络应用的需求、地址资源不足以继续适应告诉网络增长，以及对安全性难以保证等。
- 1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。
ＩＰｖ４／ＩＰｖ６转换网关的设计与实现作者：韩银锋来源：《电脑知识与技术·学术交流》2008年第23期摘要：详细介绍了IPv4IPv6转换网关的设计与实现过程；首先介绍转换网关的工作流程，其次介绍地址转换和协议转换设计，最后分析了DNS ALG设计。
关键词：IPv6；IPv4；转换网关中图分类号：TP393文献标识码：A文章编号：1009-3044(2008)23-883-02Design and Implementation of IPv4/IPv6 Translation GatewayHAN Yin-feng(Xi'an Aeronautical Polytechnic Institute, Xi'an 710089, China)Abstract:In this thesis, a translation gateway is introduced, both the design and implementation. First, the flow of translation gateway is introduced, secondly, address Translation and protocol Translation Are described, In the end of this thesis, DNS ALG is analyzed.Key words: IPv6; IPv4; Translation Gateway随着Internet的迅猛发展，现有的IP协议（IPv4协议）在应用中出现了很多问题，如地址资源即将耗尽，不能适应新的网络应用以及对安全性无法保证等。
下一代Internet协议(IPv6)不仅解决了IPv4遇到的问题，而且还给IP带来了一些新特性，它取代IPv4成为必然。
然而，现有的IPv4网路已发展了很多年，它不可能在短时间内都过渡到IPv6网络，在相当长的一段时期内，IPv6网络将和IPv4网络共存。
如何实现IPv4节点和IPv6节点之间相互通信是目前Internet急切解决的问题。
1 转换网关的工作流程本系统设计的转换网关的主要功能是通过IPv4和IPv6分组头的地址和语义转换、DNS ALG设计、TCP层校验和修改以及ICMPv4和ICMPv6的类型、代码转换，以实现IPv4节点和IPv6节点之间相互通信。
转换网关的工作流程如图1所示。
1)IPv4主机要与ipv6主机建立连接，但是不知道该ipv6主机的ipv4地址。
于是发出DNS Request问的IPv4地址，DNS request传给IPv4 DNS，若在IPv4 DNS 中找到ipv6主机的记录（IPv4是A，而IPv6的报文类型是A6或AAAA），则转到7)，否则转到2)。
2)IPv4 DNS将DNS Request转发到DNS-ALG。
3)DNS-ALG将Request中的A改为A6或AAAA，转发到IPv6网络。
4)IPv6 DNS接收到此Request，回复IPv6主机的address为3FFE:3600::140.115.78.8。
5)DNS-ALG接收到address后，在address上减去前缀，变成IPv4地址140.115.78.8。
6)DNS-ALG将A6或AAAA改为A之后继续把DNS Reply送回给IPv4 DNS再传给IPv4主机。
7)IPv4主机认为的地址为3FFE:3600::140.115.78.8，所以发出源地址为"202.115.64.5"，目的地址为140.115.78.8的IPv4分组。
8)当分组经过转换网关时，其首先查找其地址映射表中是否有源地址为“202.115.64.5”的表项，如果有则直接使用之，如果没有，从可用的地址池中分配一个IPv6地址给此Ipv4主机，假设分配的IPv6地址是3FFE:3600:B::203.79.65.7，同时在地址映射表中添加一个表项：3FFE:3600:B::203.79.65.7——202.115.64.5，从而建立起IPv4主机与IPv6主机间的地址映射。
9)转换网关根据建立的地址映射关系，进行地址和语义转换，把IPv4的分组头转换成IPv6的分组头，分组头中的源地址为“3FFE:3600:B::203.79.65.7”，目的地址为"3FFE:3600:B::140.115.78.8"，然后向IPv6网络发出分组。
反方向建立连接的过程与上述情况类似，不再详述。
从转换网关工作流程可以看出，转换网关的设计主要包括地址转换、协议转换和DNS-ALG设计。
2 地址转换设计地址转换是对IPv6地址端口对和IPv4地址端口对进行转换。
当IPv6主机要访问IPv4的网络时，必须有一个合法的IPv4地址，在该模块设计中，合法的IPv4地址端口对存放在一个地址池中，同时包括一个分配状况信息的链表。
在协议转换时，通过查询这个地址转换表，将相应的转换后的地址端口对写入新的IP数据报口。
2.1 表的设计地址转换表是地址转换的过程中最重要的一个数据结构，整个模块要实现的功能都是通过对这个转换表的操作实现。
数据转换表用于存放所有经过转换的IPv4和IPv6地址端口对的映射关系。
综合考虑查询速度和数据存储空间，决定采用了Hash表+链表的数据结构。
2.2 表的建立在Hash表建立之前，首先向地址池进行查询，查找可用的IPv6地址端口对；接着，建立IPv4和IPv6地址端口对的对应关系，并分别计算Hash值，根据Hash值，把IPv4和IPv6地址端口对的信息分别插入到Hash表头所指向的相应链表的表头；最后，要给这两个记录设置互相指向的指针，表明它们之间的映射关系。
2.3 表的搜索表的搜索主要是根据IPv4和IPv6地址端口对进行检索。
首先根据地址端口对计算出Hash 值；然后找到该值对应的链表，之后根据关键字进行匹配，顺序搜索；最后，将搜索到的记录移至Hash表头后的第一个记录。
2.4 表的修改当地址映射关系发生变化时，要对地址转换表的记录进行相应的修改。
修改的过程就是先搜索，后修改，即先根据表的搜索算法找到要修改的记录，之后再改动记录里面的内容。
2.5 表的删除随着转换网关工作的日趋频繁，地址转换表的记录会越来越多。
所以，一定要有一个表的维护策略，定期将无用或很久没用用到的记录删除，以释放空间。
3 协议转换设计IPv4-IPv6（IPv6-IPv4）转换是根据一定的转换算法，把IPv4(IPv6)的报头转化为IPv6(IPv4)的报头。
而IPv6-IPv4的转换和IPv4-IPv6转换类似，但转换方向相反。
下面重点介绍IPv4-IPv6转换。
IPv4-IPv6转换是根据一定的转换算法，把IPv4的报头转化为IPv6的报头。
1)IP协议转换算法IPv6报头各字段的内容填写规则如下：Version：6Priority：全0。
Flow Label：全0。
Payload Length：IPv4 报头中的 Total length－IPv4 报头的Header Length×4。
Next Header：填入 IPv4 报头中的 Protocol 域的值。
如果协议号是 1(ICMPv4)，则应替换成为 58(ICMPv6)。
Hop Limit：IPv4 的 TTL -1。
如果结果已经为 0，则提示“TTL Exceeded”。
Source Address 和 Destination Address：填入经转换后的IPv6地址。
如果数据报是一个分片的话，还要增加一个分片扩展头。
此时，各域的设置基本上与不分片相同，只有以下的差别：Payload Length：在上述计算结果的基础上再加 8。
Next Header：44。
分片扩展头的设置如下：Next Header：填入IPv4中的Protocol域的值。
如果协议号是1(ICMPv4)，则应替换为58(ICMPv6)。
Reserved：0Fragment Offset：用IPv4的分片偏移量替换。
MF 标志位：用IPv4的MF标志位替换。
Identification：低16位用IPv4中的标识符替换；高16位置0。
2)ICMP4-ICMP6协议转换算法在ICMPv6中对部分代码(Code)的含义进行了改变，因此必须进行相应得代码转换。
ICMP4-ICMP6协议转换算法如表1所示：4 应用层(DNS ALG)设计纯IPv4主机和纯IPv6主机之间无法直接通过IP地址进行连接通信。
但由于IPv4和IPv6具有相同的域名结构，IPv4和IPv6可以使用相同的域名空间，因此可以通过DNS-ALG进行IPv4地址与IPv6地址之间的映像，实现纯IPv4主机和纯IPv6主机之间的连接通信。
该模块主要的函数如下：1)DNS数据报转换函数这是DNS数据报转换的主函数，输入原来伪DNS数据报指针，输出转换后的DNS数据报指针。
在这个函数中，对DNS报文中各个不同的字段，调用不同的转换函数进行相应的转换。
2)查询问题转换函数查询问题转换函数用来转换DNS数据报中的查询问题部分，根据查询问题数循环修改其中的查询名、查询类型字段。
查询类通常是1，表示互联网地址，在这里不需要修改。
若查询类型是"AAAA"，就要将其改为"A";若为"A"，则改为"AAAA"。
如果查询类型是"PTR"的话，则请求获得一个IP地址对应的域名。
3)资源记录转换函数不同的字段在进行转换的时候，输入不同的指针。
与查询问题转换函数一样，根据资源记录数对资源记录部分进行循环转换。
域名和类的内容不需要改变，类型的转换与查询问题转换函数相同。
生存时间设置为Q。
转换之后，更新数据长度和数据的内容。
4)反向转换函数该函数用于将PTR类型的查询名和资源记录名进行转换。
转换过程中，首先查找地址转换表，若有待转换的IPv4/IPv6地址映射记录，则取出记录中的值作为返回值：若没有，则采取加上或去掉IP地址的前缀的方法进行地址转换。
5 小结IPv6协议是下一代Internet通信协议，它取代IPv4成为了发展的必然。
如何实现纯IPv4节点和纯IPv6节点之间相互通信是IPv6发展急需解决的关键问题。
本文对IPv4和IPv6协议之间的相互转换进行了深入的研究，基本实现了IPv4和IPv6协议之间的数据通信，达到了预期的效果。
参考文献：[1] 伍海桑.IPv6原理与实践[M].北京:人民邮电出版社,2000.[2] RFC 2766:Network Address Translation-Protocol Translation(NAT-PT)[S],2000.[3] 李津生,洪佩琳.下一代Internet网络技术[M].北京:人民邮电出版社,2001.[4] Loshin P. IPv6详解[M].机械工业出版社,2000.。