Temporarily disable RFID tags?

Hi, all,

Here is my situation: I have a reader and three tags. Three tags are all within the read range of the reader. I need the reader to read only one tag at a time. That is to say: when one tag is being read, the other two needs to be disabled temporarily. Then in the next cycle, another tag needs to be turned on so it can be read and the other two is turned off, and so forth.

Is it something possible to achieve? One solution I can think of might be: use active tags which requires power supply to be turned on, and then use digital controls to turn on and off the tags one by one.

i wonder if there is anything other solutions so that I don't need to add circuit to control the tags, instead, can I achieve the goal by doing something on the reader's side?

Thank you very much for the help!

I don't think there is a solution to this. Passive tags will all respond if they are in range and their data will collide and be unreadable. In fact it is possible that none of them will respond at all if the loading on the readers signal is so great it drops below usable level in the tags. Even active tags would have to be manually switched on and off. The only way I can see to make this unusual situation work is to use tags that operate on different frequencies. I'm working on a 125Khz passive system right now and even with two tags in proximity it becomes very unreliable and sometimes on tag will respond, sometimes the other.

Can you explain the scenario that allows three tags to be so close together please.

Brian.

betwixt said:

I don't think there is a solution to this. Passive tags will all respond if they are in range and their data will collide and be unreadable. In fact it is possible that none of them will respond at all if the loading on the readers signal is so great it drops below usable level in the tags. Even active tags would have to be manually switched on and off. The only way I can see to make this unusual situation work is to use tags that operate on different frequencies. I'm working on a 125Khz passive system right now and even with two tags in proximity it becomes very unreliable and sometimes on tag will respond, sometimes the other.

Can you explain the scenario that allows three tags to be so close together please.

Brian.

Click to expand...

Hi, Brain,

Thank you for the discussion and sharing your experience.

In fact, as far as I know, as long as the reader and tags support EPC Gen 2 protocol (which has anti-collision feature), reading multiple tags within the read range is not a problem. My problem is another story of course.

My scenario is that I am going to mount multiple tags on an object, and using RFID to estimate the tag locations so that I can tell the orientation of the object. The object is not huge so it is very likely that all tags will be within the read range of the reader.

I'm not aware of a RFID protocol that doesn't provide an anti-collision algorithm. In so far disabling of tags isn't the problem.

Your latest posts suggest that the problem is instead identification of individual tags. In this case, reducing the "operation volume" seems to be the only option, in other word design an antenna/coupler that can be brought in a position where it accesses only one tag. Or use absorbing/shielding layers to cover the other tags.

FvM said:

I'm not aware of a RFID protocol that doesn't provide an anti-collision algorithm. In so far disabling of tags isn't the problem.

Your latest posts suggest that the problem is instead identification of individual tags. In this case, reducing the "operation volume" seems to be the only option, in other word design an antenna/coupler that can be brought in a position where it accesses only one tag. Or use absorbing/shielding layers to cover the other tags.

Click to expand...

Thank you, FvM.

You are right that the problem is identifying individual tags. The solution you mentioned, if my understanding is correct, requires manually adjustment of the antenna position? That might work, too. I hope to find some solution which is more "automatic", which looks like might be infeasible..

Or use multiple antennas in different orientations and select one at a time. The problem then is how to switch them, the antenna coils, at least on 125KHz systems can carry voltage peaks of over 100V on them. I'm not sure about 15.36MHz and higher frequency tags, I've never used them. If you use polling so the RFID field is intermittent, it should be possible to switch them during the 'off' period without harm.

Is it possible to use an optical system instead of RFID? It would make the problem far easier to solve.

Brian.

It would be helpful to know the involved RFID standard.

At UHF you could use directional antennas. Also reader and tag antennas with linear polarisation and using the polarisation for the identification.

betwixt said:

Or use multiple antennas in different orientations and select one at a time. The problem then is how to switch them, the antenna coils, at least on 125KHz systems can carry voltage peaks of over 100V on them. I'm not sure about 15.36MHz and higher frequency tags, I've never used them. If you use polling so the RFID field is intermittent, it should be possible to switch them during the 'off' period without harm.

Is it possible to use an optical system instead of RFID? It would make the problem far easier to solve.

Brian.

Click to expand...

Thanks Brain.

Why is optical system helping? I am not familiar with that field. Could you please share more insight? Thank you

I was thinking of bar codes or QR codes instead of RFID. The nice thing with optics is you can focus the beam and target individual locations!

Brian.

betwixt said:

I was thinking of bar codes or QR codes instead of RFID. The nice thing with optics is you can focus the beam and target individual locations!

Brian.

Click to expand...

That makes sense. However, for RFID reader boards, I do see some which provide phase information of the signal sent from the tag, I plan to use that to find the angle of arrival of the wave. I am not aware if any optical reader give me the information of angle of arrival...

- - - Updated - - -

flanello said:

At UHF you could use directional antennas. Also reader and tag antennas with linear polarisation and using the polarisation for the identification.

Click to expand...

yeah, that might be a solution. Direct the power beam of the antenna. I imagine that requires very narrow beam if my tags are only inches apart from each other. And also, if I don't have the location information of the tag, the scanning of the beam probably will take quite some time.

- - - Updated - - -

Guys, I found something interesting: http://www.activewaveinc.com/pdf/MiniTagv10.pdf

Paragraph 4, it says "the tag can also be configured to automatically wake up at pre-defined intervals, transmit its informaiton to the system, then go back to sleep to conserve batter time"

I imagine there is some sort of microcontroller on the tag side which does the counting thing so that the tag can be set to sleep for a pre-defined time interval. This could be useful? But of course, it requires the design of tags to be more complicated.

Can you have a tag that's got enough smarts to "speak
when spoken to"? I.e. a preamble / adddress that each
tag "sniffs" and only responds when addressed? Then you
can go to a polling scheme.

I must confess, I don't understand part of the discussion.

Anticollsion features are implemented by present RFID standards. As a simple subset, they provide selection of a tag with known ID ("speak when spoken to"). If tag IDs are unknown, the anticollsion algorithm allows to enumerate all tag IDs in the actual operation volume and access these tags individually.

As confirmed by the OP, the problem is to learn the ID of tags that are initially identified only by a location.

I don't see how this problem could be solved by additional protocol features. It requires either directional antennas or tag hardware features, e.g. an activity LED that can be sensed by a camera.

Image processing as suggested in post #10 would be an alternative approach.

Thank you all for the inspiring ideas! To avoid any confusion, here is what I wanted in one sentence:

Only one tag will respond at a time although there are multiple tags within the read range of the reader. Tags should take their turns to respond one by one

I agree that the modification should focus on the hardware side by either making the reader antenna directional or control the tags to turn on one by one.

grit_fire said:

Hi, all,

Here is my situation: I have a reader and three tags. Three tags are all within the read range of the reader. I need the reader to read only one tag at a time. That is to say: when one tag is being read, the other two needs to be disabled temporarily. Then in the next cycle, another tag needs to be turned on so it can be read and the other two is turned off, and so forth.

Is it something possible to achieve? One solution I can think of might be: use active tags which requires power supply to be turned on, and then use digital controls to turn on and off the tags one by one.

i wonder if there is anything other solutions so that I don't need to add circuit to control the tags, instead, can I achieve the goal by doing something on the reader's side?

Thank you very much for the help!

Click to expand...

I think you may try to successively send the same ACK commands to the one tag that you just identified, in this case the one tag you just identified will be the only one which could be identified in one inventory round.

I am not an expert but i did some work on HF (13.56MHz)

How big is your antenna?? What frequency are you using to implement ? How are the Passive tags aligned??

I think the above questions are important for figuring out how to proceed. If you know the alignment and frequency you work on then there would be an option to prioritize the passive tag in the code and identify them.

I fear, rf1008 posted an answer to the original question of this thread that effectively ignored the specific discussion that followed.

That's also the case with your latest post. I'll close this thread to avoid further unrelated posts. If anyone of the original contributors want it to be reopened, please tell.