Mobile IPv6

Mobile IPv6 is a mobility support protocol for IPv6 at the network layer. The specification was standardized at the IETF in June 2004. The standardization process was quite slow compared to the basic IPv6 specification. The initial working group draft of Mobile IPv6 was submitted in 1996, which compares favorably with the first IPv6 draft specification which was proposed to the IPng working group in 1995. The reason for the delay in the standardization of Mobile IPv6 was the need to solve security issues associated with the protocol. Mobile IPv6 enables IPv6 nodes to send or receive packets whose source address does not match the network prefix to which they are currently attached. That is, nodes have to use a type of source spoofing technique. In the early version of the specification, the protocol required the use of IPsec to ensure that the source address was valid. However, when we consider the real situation — a mobile node may communicate with many other nodes for which it does not have any identification information — using IPsec is almost impossible.

The IESG rejected the proposal from the Mobile IPv6 working group to standardize the protocol specification at that time, and insisted the working group propose a procedure to securely validate the source address of a mobile node. The Mobile IPv6 working group started a discussion to solve the problem in 2000 and finally developed a loose address ownership mechanism called the return routability procedure in 2002. The specification was accepted by the IESG and published as in 2004.

The KAME project originally used the Mobile IPv6 stack that was contributed by Ericsson. The project started to implement its own Mobile IPv6 stack in 2001 during the middle of the second term of the KAME activity. KAME implemented several versions of Mobile IPv6 to follow and validate the latest specification. The code discussed in this chapter is based on the KAME snapshot released in July 2004. At that time, the specification had already been accepted as an RFC and the code was mature.

After KAME completed the first version of their Mobile IPv6 code, they started to redesign the architecture of the mobility stack. In the new architecture most of the signal processing tasks are moved to user space, compared to the first version of Mobile IPv6 where the code was implemented in the kernel. The design is similar to the BSD Routing Socket mechanism, which separates the routing information exchange and forwarding mechanisms, with exchanging routing information in the user space and forwarding in the kernel space. There are many benefits to this design. It makes it easier to develop complicated signal processing code since developers can utilize many advanced debugging programs and techniques, while the packet processing performance is not reduced, since it is done in the kernel. Extending or replacing some of the signal processing mechanisms is also easier, which makes it possible to add support for new mobility protocols or to adapt some part of the functions to user needs. Reducing the amount of kernel modification is important when we consider merging the developed code into the original BSDs.

Source of Information : Elsevier Wireless Networking Complete 2010

Mobile IPv6