Radio frequency identification readers are likely to soon be commonly outfitted with TCP/IP stacks, as well as Ethernet, Power-over-Ethernet and 802.11 wireless LAN connections. This will be needed to help enterprises eventually scale their RFID deployments, so they can support multiple applications using a single infrastructure.
When this happens, as well as network connections, RFID kit will need management software and standards to control the readers and applications centrally.
Eventually, RFID projects will scale
RFID projects are in their early stages, hampered by doubts: for instanceWal-Mart's project is widely felt to be too ambitious, and security issues are cited by observers such as Scott Bradner. This is expected to change as the technology matures.
Historically, RFID implementations in a given organisation have been relatively small and devoted to a single application. So it has been reasonable to run, say, a serial line between a reader (or two or three) and the host supporting that application.
However, to support multiple RFID applications within an organisation or a single application across multiple organisations, the enterprise network will likely soon begin playing a more significant role than it has in the past. WANs, for example, might be required to tie multiple local RFID reader networks at different sites to a central tracking application.
RFID tags are tiny identification mechanisms affixed to goods and equipment (and even living beings) that communicate with readers across the unlicensed ultra high frequency (UHF) band.
Air interface protocols that allow readers and tags from different manufacturers to interoperate (such as EPCglobal Gen 2) have been specified to handle the wireless communications between readers and tags. On the network side, data needs to get from the RFID readers to an application or database to be useful. So network-side protocol standards work is afoot.
RFID control needs standards
EPCglobal has taken a stab at an initial Reader Protocol (RP1.0) based on the capabilities in the current generation of RFID readers.
Meanwhile, the Internet Engineering Task Force might be getting into the act. It is exploring the formation of an IETF Working Group to standardize IP-based tag data operations and management for networks of RFID readers. This activity revolves around the Simple Lightweight RFID Reader Protocol (SLRRP) - yep, pronounced "slurp" - which would interface RFID readers with upstream devices on an enterprise IP network.
In addition to building a streamlined, efficient protocol for use in dense reader environments, a potential SLRRP Working Group would hope to devise a protocol that supports a diversity of application requirements, reader architectures, and both existing and new air protocols developed by any standards body.
The SLAPP protocol, proposed to solve the somewhat similar problem of linking wireless access points, is being proposed as a basis for SLRRP.
The two ideas take a similar approach to linking diverse technologies. SLRRP aims to support optional RFID tag extensions implemented by different manufacturers, while SLAPP tries to allow wireless LAN vendors to make their own control methods.