How Does Dense Reader Mode Work, and How Do RFID Systems Reduce Tag Collision?

I have been confused about RFID dense reader mode and am unable to find the information I seek. In dense reader mode, readers hop between channels to avoid collisions at the tag side. The tags see query messages from the readers that are separated by channels, but passive RFID tags utilize an ASK demodulator, which consists of an envelope detector and a comparator. Such a receiver front-end at the tag side isn't capable of differentiating different channels (frequencies), which means even the channels are separated and collisions still exist. Am I correct about the above?

—Xingda,

———

Xingda,

I am not an RF engineer, so I am not capable of providing a highly technical explanation of dense reader mode. But I will try to explain how it works. Readers transmit within different channels inside the frequency band allotted for RFID readers (902 MHz to 928 MHz in the United States). In the U.S., there are 50 channels in this band. In the European band, which is smaller, there are 10 channels.

In dense reader mode, a reader listens to the channel it plans to use before emitting a signal. If it hears another reader using that channel, it switches to another one. This prevents readers from interfering with one another. There are two basic ways in which RFID systems can reduce tag collision. One is deterministic (the tree-walking algorithm). The other is probabilistic (the ALOHA anti-collision algorithm).

With the deterministic algorithm, a reader asks tags to respond if their serial number begins with, say, a one. If two tags respond, it asks them to respond if the first number of their serial number is one and the second is two. It keeps asking the tags to respond until all tags have been identified. Imagine a teacher asking all students in a classroom to stand if their last name starts with an A. If five students stand up, the teacher might say to remain standing only if a child's last name starts with an A and the second letter is B. If two students remain standing, the teacher may ask for students with the first two letters A an B and a third letter to remain standing. This is known as the tree-walking algorithm.

With the probabilistic method, tags respond to reader commands at random intervals. If a collision occurs, the tags that experienced the collision will respond again to the reader after a random period of time. If there are only a few tags in the field, it is likely that the tags will respond at different times. If there are many, it is possible that two or more tags will collide again. In that case, they will wait and respond at another random interval. This keeps happening until there are no more tag collisions. You can search for "slotted ALOHA algorithm" on the Internet to learn about how such a system works.

I hope this answers your questions.

—Mark Roberti, Founder and Editor, RFID Journal