Just like kids at school, RFID tags are divided into classes. These classes were developed primarily by two industry standards organizations-EPCglobal (a division of GS1) and the ISO-in order to classify and unify function and capabilities of RFID tags. EPCglobal has done its best to promote the use of EPC.
EPCglobal and other standardization organizations will be discussed in more detail in Chapter 9, "Standards and Regulations," but EPC classes and related tag capabilities belong in this chapter because they are so specific to tags.
Time to blow the dust off the RFID history book. In early 2000, two RFID companies started to produce RFID tags. One of them was Matrics (later bought by Symbol Technologies) and the other was Alien Technology. Matrics came up with a tag that was read-only; therefore, the data had to be encoded at the point of manufacturing and could not be rewritten by the user. The design came from some very smart engineers from the National Security Agency (NSA). These tags belonged to EPC Class 0. Matrics tags came with a dual dipole antenna starting at a size of 4" × 4" and performed very well.
Alien offered a tag that was write-once read-many (WORM) , which means that the tag's data or number could be written by the user. This tag became the basis for EPC Class 1. Alien tags came with a single dipole antenna in a squiggly format and the name "squiggle" tag. Although the standard for EPC Class 1 specifies that this tag is WORM, in reality it could be rewritten several times. Matrics saw the shortcomings of their read-only tag and introduced a WORM tag, which was unofficially called Class 0+. This tag was also fully rewritable, but was not compatible with other vendor equipment; therefore, it could not be rewritten by any reader other than Matrics. Class 0+ never became an EPC standard.
In the meantime, EPCglobal developed a new standard based on the capabilities of both EPC Class 0 and 1 tags, enhanced with many new features and functions. These tags became EPC Class 1 Generation 2 (a.k.a. Gen 2). This standard was submitted to the ISO for ratification under ISO standard 18000. After the ISO approval, Generation 2 will become one unified international air-interface protocol as well as the data standard for UHF RFID tags, and eventually for HF tags as well.
EPC Class 1 Generation 2 tags offer many new and enhanced features, such as the following:
Faster Data Transfer The Gen 2 protocol supports a data-transfer rate from tag to reader of up to 640 kilobits per second (Kbps). In contrast, Generation 1 Class 0 could transfer up to 80 Kbps and Class 1 up to 140 Kbps. Gen 2 also supports much faster writing, approximately 800 bits per second.
Higher Read Rates In the United States, read rates could reach (under ideal conditions) up to 1,500 tags per second; in Europe, because of regulatory restrictions, the rate could be up to 600 tags per second. For comparison, Generation 1 tags were capable of reads about 10 times lower.
Better Access Control and Security Gen 2 tags provide better access control and security because of 32-bit Access and Kill passwords. Gen 1 used 8-and 24-bit passwords.
Capability to Generate and Respond with Random Numbers Based on a Value Q Given by the Interrogator (Q Algorithm) This not only makes the Gen 2 tags more secure, but also gives the reader the ability to inventory tags even if they have the same EPC.
Proven Air Interface Gen 2 tags use FM0 or Miller subcarrier encoding for backscatter. FM0 is fast but more sensitive to interference, whereas Miller subcarrier encoding fits the tag responses in between the channels used by the reader. This latter method is slightly slower but is more resistant to interference.
Memory Divided into Four Memory Banks Bank 0: Reserved memory-carries 32-bit Access and 32-bit Kill passwords
Bank 1: EPC memory-includes EPC number of a flexible size (16 to 469 bits) and various protocol controls
Bank 2: Tag identification (TID) memory-consists of information about the tag itself, such as tag ID
Bank 3: User memory-provides a space for user-defined data
Reduced Ghost Reads Gen 2 tags and readers use several methods to avoid ghost reads. Tags are programmed to respond with a certain delay in a small uncertainty window. If the tag responds outside this window, the reader ignores it. Or, the tag can send a preamble (unique wave). After this preamble is validated by the reader, the communication can begin. The third method involves the reader checking a data stream coming from the tag for a valid EPC format.
Sessions Each tag can operate in four sessions . This is useful when the tags are inventoried by more readers. Each reader or group of readers is assigned a different session and they do not interfere with other readers when interrogating tags.
AB Symmetry AB Symmetry provides the tags with a capability to be in a state A or state B. A reader changes the state after it reads the tag. This method replaces the Gen 1 method, which put tags to sleep after they were read to avoid interference with interrogation of other tags. (Sometimes they never woke up, and sometimes the waking up took valuable time.)
Smaller Chips Most of the Gen 2 chips are smaller in size compared to Gen 1 chips, which allows for smaller power consumption and faster data processing due to smaller gates on the chips. Smaller size also means that more chips can be produced from a single wafer, with the idea of reducing costs.
Improved Dense Reader Mode The frequency band is divided into several smaller bands, which allows for the readers and tags to have their own section. Tags and readers are not permitted to share a band, which prevents them from interfering with each other. Readers are required to stay strictly within their band to avoid leaking their energy into the tag section.
The Gen 2 protocol was built to leave open options for additional EPC classes. The tags with these functions already exist; however, they are not standardized by EPCglobal yet:
Class 2 would include tags that would be fully rewritable, carry a large memory, and support encryption.
Class 3 would include semi-passive tags with possible sensors.
Class 4 would standardize active tags with the ability to communicate with other tags.
Finally, Class 5 would provide guidelines for tags based on Class 4 but with added reader capability. These tags would function as readers as well as tags.