How to Design 1-3/4 Inch Square PCB Antennas for SM130 RFID Reader Chessboard Project

I just bought the SM130 RFID reader which requires an external antenna. I also bought the external antenna shield that contains extras e.g. xbee, etc. What I need is to make 64 antennas for a chessboard using only one reader module. I have devised a way to make this concept work based on some info from an essay, but I think it is not enough to support my idea. So is there any advice can give me and most important is how to deal with it. 


At a world, Can someone help in the process of making a 1-3/4 inch square pcb antenna for the sm130?


Help is appreciated.



Best regards.

Helay

By the way, I have a small question, what I have read is about Integrated Laminate Substrates and Passive Devices, is there anyone can recommend me more material or articles about RFID. Thanks a lot.

There is a bit of info about designing 13.56 MHz RFID antennas herehttps://electronics.stackexchange.com/questions/377706/on-board-13-56mhz-rfid-antenna-design
Not a lot of detail but it may give you some pointers.There is a 55x55mm PCB antenna here for $2.50 each:https://shop.sonmicro.com/PA1356-1
This company also does some RFID products including an evaluation kit with antennas:https://shop.sonmicro.com/SM1250-EK

If you are going to have 64 antennas in near vicinity of each other, you will likely have problems with interaction between the antennas. You will also need an antenna switch to select which antenna you want to use.

I'm envisaging a system with 64 reader antennas, only one of which is switched to the reader electronics at any one time.  So the reader would"Scan" the board one square at a time.  So interference between antennas would not be a problem.I can however see a problem when two pieces are on adjacent squares, or an empty square has a piece on an adjacent square.  you may get false responses.  This would have to be controlled with transmit power level in the reader, perhaps?

@davi_d,Yes, the false response rate would have to be addressed but what I was thinking was that any metal in the near field of an antenna can disrupt the tuning so you need to be aware of the presence of other antennas when you design them.
The best way to avoid the problem would be to make the antenna sufficiently directional that it wouldn't matter. This would have the beneficial side effect that it would also reduce the false response rate.In this case you are trying to fit an antenna into dimensions that are a small fraction of a wavelength which implies the use of a multi-turn coil. A multi-turn coil will be directional but I wouldn't know by how much without modeling it.

Hi Elizabeth.  The impression I get from what I have read is that RFID / NFC is more an inductive coupling technique than pure RF transmission - I recall one thing had about a 10:1 inductance range for the coil.I don't know if readers have an SSI (Signal Strength Indication) - if so it should be a simple matter to only take account of responses that have an SSI corresponding to a piece right on the square.

@davi_d,You're probably right that it's more inductive coupling than pure RF. If so, then all you need is a coil etched on the PCB that fits inside the square. 
On the other hand, switching the antennas in the array without spreading the RF around could get interesting.

All - this user seems to reference Kynix a lot and never replies to any advice given on posts.I suspect the object is advertising.

Thank you for all guys advice, but I think I get a little useful information. I am really busy recently, and I get so many ideas from the online friends, I will arrange them in my free time.