Low-Power RF: which one do I choose?

Wireless & IoT
1

I need help choosing an appropriate wireless module/protocol to use for my design. I’m designing a wireless system to take samples from a signal at 30 ksps with 8-10 bit resolution. The samples need to be wirelessly transmitted to a PC for processing and logging. This system will be worn by rats, and need to be small, lightweight, easy to implement, and very low-power. I most likely will only collect 500 ms of data every 1-2 seconds.

Questions:

  1. Which wireless protocol and module would suite my needs best?

  2. If I need to add a second channel (60 ksps total), would the above suggestion be able to handle it?

,Regards

John

2

You can use this chart to help decide if XBees will sufice for what you need to be able to do.

http://www.parallax.com/tabid/829/Default.aspx

If not, is wifi an option?

If will enable you to be able to transfer much more data much quicker.

(http://www.sparkfun.com/products/10050)

3

Your project is pretty tough, due to the size requirement. I worked with 200 gram rats and they are pretty small to attach a radio-telemetry system to…

I would suggest using a Nordic wireless ANT transceiver, as they are low power and small size, along with your choice of microcontroller.

Closest system I have seen is the Nike Footpod that interfaces to an iPod, also called a Nike+iPod device. You will see the coin cell battery is most of it and will also be a size limit for your project.

It has been reverse-engineered here:

http://dmitry.co/index.php?p=.%2F04.Tho … lus%20iPod

Sparkfun does not have “low power” Nordic modules as they add energy wasting voltage regulators. I buy the ones from mdfly for $7…

4

jram112:
I’m designing a wireless system to take samples from a signal at 30 ksps with 8-10 bit

,Regards

John

Suggest you calculate the data rate needed for 30,000 samples per second at 8 or 10 bits per sample. You'll see it's a higher data rate than inexpensive wireless can handle. So you should look at reducing the sample rate or sending delta datums or doing some preliminary signal processing before transmitting.

Or, you’ll wind up with 802.11 type wireless which is OK if your battery budget can take it.

5

The nRF24L01+ appears to be ideal for that application, with a suitable MCU such as a small PIC. The units can be made very small, and the battery life can be good. You could do the prototyping with the cheap nRF24L01+ modules available on Ebay, and design a PCB.

6

what kind of data does a rat have that needs to be sampled @30Kbyte/sec?

7

redwire:
Sparkfun does not have “low power” Nordic modules as they add energy wasting voltage regulators. I buy the ones from mdfly for $7…

Since you indicated that you use the mdfly’s modules, do you know how well they perform in a plastic enclosure as far as communicating over some distance with one another? Say maybe no more than 25 feet apart. What about line of sight?

Alternatively, do you know if one can connect an external antenna to these modules? I like the SFE’s modules because they come with the RP-SMA connector already on-board.

8

kirash, for these (mdfly) modules http://www.mdfly.com/index.php?main_pag … ucts_id=81 I get over 100 feet in plastic enclosures with 2 walls between. Good enough for me.

To get the most of the nRF24L01+'s I would suggest:

  1. You must have clean power for the nRF24L01+, I use an LC filter so MCU noise does not wreck sensitivity. (not an issue for SparkFun modules with built-in voltage reg).

  2. Be using a 2.4GHz channel not occupied by other equipment (wireless routers).

  3. Make sure you have transmit power set to max.

  4. Have the antennas oriented correctly. I’m using the pcb trace antenna which seems to radiate in a dipole, dead spots on the pcb back-side.

I haven’t tried any external SMA connector antenna modules. I like the ones with a built-in extra RF power amp.