Separation Protocol of Locator and Identifier Towards IPv6

Abstract

In the current TCP/IP architecture, the IPv6 address has a dual meaning in semantics. It not only represents the topological location of the network node, but also the identity of the node, which is usually referred to as the semantic overload problem of the IP address. The semantically overloaded IP address represents the topological position of the network, and the topological position of the network generally does not move, so the device entering the new network environment needs to replace the new identity IP to adapt to the change of the topological position. The semantic overload of IP addresses is not conducive to supporting mobility and user identity authentication, resulting in tight storage space for routing equipment, lack of unified communication identification for network equipment, and difficulties in network traceability and management. In order to solve the problem of IP address semantic overload, this project focuses on the separation technology SPLIT6 of IP address identity and location.¶

1. Introduction

In the current Internet architecture, the IPv6 address carries too much semantics. The network layer protocol uses the IPv6 address as the location identifier of the user terminal, and the transport layer protocol uses the IPv6 address as the identity identifier of the user terminal. This dual identity of the IPv6 address cannot satisfy the Internet's increasing mobility and security requirements.¶

In order to solve these problems caused by the semantic overload of IPv6 addresses, separating the location information and identity information of IPv6 addresses has become an important research direction.¶

2. IPv6 address semantics problem

In the current TCP/IP architecture, the IPv6 address has a dual meaning in semantics at the same time. It not only represents the topological location of the network node, but also the identity of the node, which is usually referred to as IP address semantic overload. The semantically overloaded IP address represents the topological location, and the topological location cannot be moved, so the IP address representing the identity of the node cannot move with the movement of the user or device. The equipment entering the new network environment needs to be replaced with a new identity IP to adapt to the change of topological location. The semantic overload problem of IP addresses is not conducive to supporting mobility, affecting the scalability of core routing, reducing the effectiveness of existing security mechanisms, and restricting the development of several new technologies. Due to the semantic overload problem of IP addresses, the following problems exist in TCP/IP in actual operation: The storage space of routing equipment is tight. In order to improve and ensure the performance of the Internet, the routing table entries of the routing devices in the Internet should not be too many. If a large number of IP address prefixes that have not been aggregated are advertised to the core route, it will cause the expansion of the core routing table entry DFZ (default-free zone), the increase in the frequency of route updates and the increase in communication volume, and the slower route convergence, which will cause serious problems. Affect the performance and scalability of routing.¶

3. exist network problem

The network equipment lacks a unified communication identification. With the development of IOT (Internet Of Things) in the current Internet, the number of devices connected to the network has increased exponentially. These devices need to communicate with other devices, so a unique communication identifier that can represent this device must have. Currently, the industry does not agree to use IPv6 address as a universal communication identifier for devices. There are two reasons. One is because IP addresses have dual meanings. As the network environment changes, the device IP address will also change. Therefore, the difference between the device and the IPv6 address A one-to-one correspondence cannot be established between them; the second reason is that considering the performance and security of IOT devices, IOT devices are generally simple in design and only use the physical layer, link layer, and network layer of the network instead of the transport layer. And application layer to reduce overhead. Therefore, the IP address is generally used to identify the device, but many IOT devices are highly mobile. How to ensure that the IOT device can still use a fixed IP address to identify it when it is moving is an important problem that needs to be solved. In view of the above problems, if the coupling problem between the identification location and the identification identity can be solved, the development of IOT and the Internet of Things can be greatly promoted.¶

Network security control is difficult. The most important way of network security management and control is to trace the location and identity information of the IP address of the initiator of the network behavior. However, in the current TCP/IP architecture, the IPv6 address has a dual meaning, which is not only fixed network location information, but also unbound identity information. It is not possible to locate a specific device through the IP address, and then locate a certain person. Because with the switching of the network environment, the same IP address may correspond to different users and different devices, and the devices of the same user will also be assigned to different IP addresses as the network switches. All these have caused great troubles to the supervision of network security. Because the current network is insecure, an important reason for frequent network attacks is that the attacker's address cannot be traced to the source or it is difficult to trace the source. If each user can be assigned a fixed identity-IPv6 address in the network, then network attackers will have nowhere to hide, and network supervision will become simple. Therefore, IP semantic overload is the frequent occurrence of network attacks, and the source of the attack cannot be traced back to the root cause.¶

User identity is difficult to authenticate. Due to the dual meaning of IP, users cannot always log in to the network using a fixed IP address. Because once the user switches the network environment, he needs to change his device's IP address and network configuration to log in to the network again. The reason for this phenomenon is that the IP address assigned by the user has location attributes, so this IP address is bound to the network environment where it is located, and the IP address cannot move with the user's location. Frequent switching of IP address and network environment will bring a lot of inconvenience to users. For example, the ongoing network conference will be interrupted, the video being watched will be suspended, and the sending and receiving of emails will need to be re-authenticated with the IP address. The mobile performance of the device is poor. In the current TCP/IP architecture, because the IPv6 address has a dual meaning, it represents the network topology location of the device, and it is also the identity of the device. This leads to poor mobility of the device, and a device carrying a specific IP address cannot log in to the network after switching to another network environment. These devices need to reconfigure the network and change the IP address to log in to the network again. In the current Internet architecture, the IP address carries too much semantics. The network layer protocol uses the IP address as the location identifier of the user terminal, and the transport layer protocol uses the IP address as the identity identifier of the user terminal. This double identity of the IP address cannot Meet the increasing mobility and security needs of the Internet.¶

4. possible solution

In order to solve these problems caused by the semantic overload of IP addresses, it has become an urgent need for academia and industry to separate the location information and identity information of the IP address. In recent years, countries around the world have successively initiated a number of research projects on the separation of IP address location information and identity information. The MobilityFirst[1] project started in 2010 and was funded by the Future Internet Architecture (FIA) program of the National Science Foundation. The first phase of the FIA project started in 2010-14 and produced a new mobility-centric architecture called MobilityFirst (MF), and a prototype implementation of the protocol stack. IETF (Internet Engineering Task Force) [2] established a corresponding working group to study the separation of identity and location identification. Among them, the HIP working group advocated by Ericsson mainly studied the host identity protocol HIP (Host Identity Protocol), and proposed rfc7401[3] and rfc8002[4]. The Shim6 working group advocated by Sun company mainly researched on the IPv6 Multihoming Shim Protocol for IPv6 (Multihoming Shim Protocol for IPv6) and proposed RFC5533[5]. The RRG (Routing Research Group) working group advocated by Cisco mainly researches the Locator/Identifier Separation Protocl (Locator/Identifier Separation Protocl), and proposes RFC6830[6] and RFC8113[7]. In addition, there are TIDR (Tunneled Inter-Domain Routing) [8] and IVI [9] programs. In these researches on network systems, it is generally believed that the semantic overload of IP addresses has affected the development of network system structures. Therefore, breaking the semantic overload of IP addresses and establishing a network that separates location and identity has become an important issue to be solved in the construction of next-generation IP networks.¶

7. Acknowledgements

The authors would like to acknowledge XXX for their valuable review and comments.¶

8. References