**7. Mobility considerations**

simply hash the longer EPC's into a length of 64 bits and then use the direct mapping meth‐ od again. The hashing technique used to derive identifiers was described in Section 3.4, when the CGA namespace was introduced. Another method would be to identify if there are some bits in the longer EPC's that can be removed without affecting the uniqueness

A key benefit of the proposed solution is that there is no need to change the design of exist‐ ing RFID technology with its EPC namespace conventions. The application can be installed on a computer connected to the reader, and then all objects with RFID tags that pass this reader will put the objects online and thereby giving them the ability to communicate over

the Internet as long as the tag is within range of a reader.

**Table 3.** Overview of strategies for mapping Tag ID codes to IPv6 network addresses.

use more that 64-bit for identification.

Table 3 outlines the different strategies for mapping of tag IDs to IPv6 addresses. Essential‐ ly, these divide into methods that work with tags of 64-bit identification or less and tags that

property of the tags.

126 Radio Frequency Identification from System to Applications

One of the largest challenges for a dynamic, networked system lies within the mobility sup‐ port of the network. In the case described here, we consider a system of fixed readers that are connected in a common network infrastructure. Mobility arises when tags are moved be‐ tween readers. Readers will be wired or wireless and they will have different communica‐ tion ranges according to their MAC technology. Moreover, they will forward the read tag IDs to the server through the common network infrastructure.

When a tag moves from one reader to another, the network prefix will change but the host suffix/interface ID will still match the tag's EPC. The tag will in effect change its network address every time it passes a new reader. Hence, the challenge is to effectively keep track of tags when the address changes this rapidly.

There are basically two distinct ways to solve the mobility problem. One is a centralized ap‐ proach, such as mobile IPv6 [30], where a central server, i.e., the home agent, is used to keep track of the mobile hosts that move around in the world. The mobile IPv6 architecture relies on the concept of a home agent and a care-of address. The method is based on some soft‐ ware on the network layer that can send messages to the home agent making sure that the home agent is holding an updated address list at all times. Initially, traffic destined to the mobile host is routed to the home network and subsequently tunneled to the foreign net‐ work that the host is visiting. Fortunately, IPv6 supports mechanisms to circumvent the tri‐ angular routing problem that arises in this setup [30].

Dominikus et al. [14], proposed to use mobile IPv6 to handle the mobility of IPv6-enabled tags. In their approach, the care-of address refers to the subnet of the RFID reader, where the tag is currently present. Whilst the care-of address is a globally unique address assigned to the host, i.e., the tag visiting a foreign network, the home agent address is specific to the en‐ terprise using the issued tags.

Alternatively, mobility support can be obtained in a more distributed way by separating lo‐ cation and identity information. This can be achieved by using the HIP approach [22]. In this approach, there is a need to compute the routable IPv6 address from the given non-routable HIT the host has been given.

HIP allows consenting hosts to securely establish and maintain shared IP-layer state, allow‐ ing separation of the identifier and locator roles of IP addresses, thereby enabling continuity of communications across IP address changes. A consequence of such a decoupling is that new solutions to network-layer mobility and host multi-homing are possible [22].
