7

I'm a web developer - so IoT is not my speciality at all - and I've been asked to find the cheapest and most efficient way (in this order of priority) to build a gizmo for a sport event (can't be more specific). This is how it should work :

  1. A Competitor wear a wristband carrying his unique ID.
  2. At one place there is a terminal which will scan the wristband once in contact, so organizers will know at what time the competitor arrived to this terminal via a web app.
  3. The Competitor must stay 3secs at the terminal and can't just extend arms forward, they must be at the terminal.
  4. The Competitor is acknowledged that his wristband has been successfully scanned and can now move to the next terminal. And so on

So my question is, what should I use for the wristband and the terminal knowing that the bracelets are throwaways ?

EDIT - More details :

  • Competitors can't have their phone nor any device with them during the event.
  • There will be between 40 and 50 terminals max
  • I've been asked for the cheapest solution but I don't have a min/max cost and I'm not limited by dev time (must be reasonable though)
7
  • 1
    @SeanHoulihane What do you mean by "off-the-shelf" ? I'm aware that there is no cheap solution 100% matching my needs, that's why I want to know what such project would take and if it's possible to assemble it myself based on an arduino/raspberry/[any microcontroller] solution. – AdrienXL May 10 '17 at 15:33
  • 1
    @AdrienXL I think Sean's asking if it might be easier and cheaper to buy a ready-made solution such at these—£95 for 100 RFID wristbands, and then you'd only need to develop a reader of some description. Many modern phones have RFID reading capabilities, which might be good enough for your use case. Would something like this be acceptable, or do you have a specific price/complexity constraint? – Aurora0001 May 10 '17 at 16:18
  • 1
    @Aurora0001 oh sorry If I missunderstood Sean. I've been googling around and I've concluded that those kind of wristbands is indeed the cheapest solution. However, competitors can't have their phone with them and I need the "terminals" to be the RFID reader and only them. The best solution I've found so far is an RFID reader + 433mhz emeter on a Arduino card + a computer with a receiver. But I'm affraid of the cost and the time it will take if I have to assemble the arduinos myself (I need 50 terminals). – AdrienXL May 10 '17 at 16:55
  • 1
    13.54 MHz seems to be a pretty good standard rfid to target. there are readers like this one to be had for super cheap that can interface with whatever MCU (arduino, esp8266, teensy, etc...) or micro computer (ras-pi, chip, etc..) you can get your hands on and program. Security and prevention of cheating is up to you and your skillz – Aaron May 10 '17 at 17:13
  • 1
    @SeanHoulihane Indeed, i'm trying to make a proof of concept first but close enough to reality regarding costs and time. If the project goes to the end, it will be only use by my client and won't be available for sale. – AdrienXL May 10 '17 at 18:04
7

How far away from your base computer do the terminals need to be? Does it need to be a relatively real-time system or can the check-ins be cached for a few seconds?

If you could get away with the range of wifi, and the potential latency of an mqtt message (a good protocol if you need QOS) I think an esp8266 microcontroller with one of the these RFID readers would be a nearly ideal setup.

(I personally have a couple of wemos D1 mini's *note this is not the cheapest they can be found, but I try not to promote knock off's)

I've primarily used the NodeMCU firmware, but there's no baked in library for pn532 RFID chips, so you'd have to read/write i2c/spi registers manually. Adafruit has a library for the Arduino IDE, but it only works with i2c (seems under-tested / under-developed for the esp8266)

One of the benefits of a setup like this is that you could quite easily make these battery powered with a usb battery bank (watch out because some turn off if they don't sense enough current draw).

If I were to build these with parts from aliexpress (super cheap) this would be my shopping list:

Then for deployment, you'd need some sort of decent wifi access point that can handle a bunch of lightweight connections (some have a cap on # of connections) and probably a laptop running the mqtt host and your web app server.

5
  • Any view on RFID over NFC? I don't see much driving the choice other than component availability... – Sean Houlihane May 10 '17 at 18:32
  • @SeanHoulihane comment section led me to spec a system around these wristbands. – Aaron May 10 '17 at 18:33
  • 2
    The definition of nfc vs rfid is somewhat blurred... you really just have to pick a target protocol (in this case MIFARE 1K). Also component availability is not something that can ever be ignored. I would say it's actually one of the most important factors. – Aaron May 10 '17 at 18:37
  • 2
    This is awesome @Aaron (even though you have underestimate tears a lot :P). Thank you ! All the area will be covered by wifi and no need for real real-time so it seems that your answer feets my needs perfectly. (And Thank you Sean too ! ) – AdrienXL May 10 '17 at 19:13
  • 1
    @AdrienXL hopefully you can prove yourself to this client of yours and charge lots for your tears xD – Aaron May 10 '17 at 19:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.