# ultra low power mini wireless transmitter / receiver?

I'm looking for suggestions for a wireless transmitter & receiver combination. I'm basically just trying to bridge several buttons over a wireless signal(s) (i.e. send a 1 bit signal) in as small a package as possible:

• Very short range (needs to be 99% reliable at ~3 ft away, no line-of-sight).
• Low power, especially for the transmitter. I'd like to be able to send ~10,000 pulses off of a watch battery
• Very small transmitter (needs to fit inside something around 1 in^3, though I have a fair bit of flexibility on placement)
• Would like to prevent accidental cross-talk

Any ideas appreciated. If interested in the reasoning, I'm trying to build a wireless Di2 system (electronic bicycle shifting)

-
@vicatcu - Thanks. I've been looking around and have also found things like sparkfun.com/categories/79?sort_by=price_asc&per_page=50, but I'm really just trying to figure out how to transmit a 1-bit signal off of as small a setup as possible –  zyklus Feb 1 '12 at 20:34

A 433MHz ASK transmitter/receiver pair should do very well at the ranges you are talking about. I have used them quite successfully in my Wicked Node and Wicked Receiver products, at hundreds of feet, so you should easily be able to use them to make it work over such a small distance. The transmitter can be quite low power because you can basically draws no current when not transmitting. The receiver on the other hand, needs to always be on and continuously outputs "data" that you have to sift through to extract actual transmissions from the noise. You could use the same exact product set to do what you want here but it will be a little bit longer than 1 inch, but you might be able to hack it into a 1 cubic inch with some creativity.

-
So one of my main concerns is trying to figure out how to get around the noise. Since I'm only trying to transfer a single high signal, I'm trying to avoid having a micro-controller on the transmitter side to send a "real" transmission (as opposed to a single bit which would be mimicked with interference constantly), but from everything I'm reading there's so much noise that attempting to do this without encoding the signal might be futile -- wondering if you concur? –  zyklus Feb 1 '12 at 21:54
never mind, I'm seeing that the ASK receiver handles this all :) Any specs on the continuous power draw? –  zyklus Feb 1 '12 at 22:01
One other question, do you know if it's possible/easy to connect multiple transmitters to one receiver? Given that I think this is a rotating signal, I'd need a receiver that could pair multiple transmitters, correct? –  zyklus Feb 1 '12 at 22:12
An ASK receiver cannot handle multiple transmitters transmitting on the same frequency at the same time. You have to do something like a policy of essentially time division multiplexing, or operate on separate frequencies... –  vicatcu Feb 2 '12 at 2:25
The same frequency is fine. I am more curious as to whether or not it would be able to pair with 4 transmitters at the same time –  zyklus Feb 2 '12 at 16:58

The big players are Bluetooth LE (Low Energy) or Zigbee, but there are some new protocols that beat them easily.

You could take a look at TELRAN protocol. At the max speed (50 kbps) on a 1V coin cell it uses 3.3mA peak. There is also the ANT protocol, it has higher bandwidth (1mbps) and claims to operate for 3 years on a coin cell, so I don't know the exact peak power consumption. It's backed by very large companies (Sony phones have it build-in for example), so I guess it could be a safer bet than TELRAN in the end.

-

I'm working on the Jennic JN5148 module, which is a 32-bit microcontroller with embedded 2.4 GHz transceiver, it's about 1'x0.5'x0.1', so with the battery it will surely fit in your specs (I've seen a very tiny case with it). Plus, it consumes about 15mA * 3V transmitting, and your system requires a very short duration, say 5 ms.

2.4 GHz has the advantage that you require a very small antenna (there are also PCB patches) and you have enough bandwith to do very short transmissions, thus saving power.

In general, you can improve drastically the reliability using acknowledges, but this doubles (at least) the number of signals you are transmitting, at the expense of power. For a simple link like this you could use the plain 802.15.4 physical and MAC protocol, that is well suited for point-to-point transmission and simple networks.

Finally, if your application is allows it, it's much more efficient if you can buffer the data over some cycles (let's say 1 byte instead of 1 bit) because the metadata (header and maybe parity) scales much better with a bigger payload.

## Update

I've ran some tests on a sensor node equipping this module: I can confirm (very accurate measurements) that the transmission cycle of this module consumes about $63 \mu C$, which means $63 \mu A$ for 1 second, or 17,5 nAh; if you have a coin cell battery of 1mAh (which is very small), you can do approximately 57 millions of transmissions. And this is for a 64bit message, you can shrink it more sending only one bit.

Ah, it has a range of about 30m in a closed space with walls and many many devices in the same frequency.

-
What's the receiving current like? I've played around some with Nordic's RF stuff, but the receiver consumes a lot of current even when no data is arriving. –  supercat Feb 2 '12 at 15:22
@supercat it's rated for 15mA transmitting and 17.5mA receiving; and yes, it's often used as always on; but I think that using smartly the timing, you can make the receiver to turn on just before the incoming transmission. But that requires a good design of the software and a wise use of timers –  clabacchio Feb 2 '12 at 15:29
@supercat: Thats typical for an RX current. You need to develop some scheme to not always be listening if this is a concern. –  Yann Ramin Feb 2 '12 at 20:51

What you are asking for does not exist and can not exist. Forget about it. No RF communication is 100% reliable.

-
I'm happy with >99% :) –  zyklus Feb 1 '12 at 20:34

If you want wireless di2 then you will need to be able to transmit four different codes or on four different channels (unless you are also making it sequential, when you just need two channels). False positives are also going to be frustrating (and potentially painful) so transmitting a single bit is likely to be prone to errors.

When I thought about this I had a quick look at bluetooth low energy but didn't find anything suitable for me. If you want to go overboard then you could use an arduino mini with normal bluetooth but it may draw a bit too much current and be expensive (if you can't join all four inputs into a single transmitter).

I also thought about coded garage door openers. The remotes can be quite small (fit on a key ring). You can get one receiver to work with a number of different remotes but I don't know if one receiver is able to tell you which of the four remote buttons has been pressed.

At the moment, I am just using wires for my ultegra di2 TT bars hack but I'll keep popping back in case you get an answer.

-
Are you the person with the carbon tt bar ends? I have a solution i am working on. It involves reverse engineering the phase shifting shimano is using and getting a custom PCB capable of pretending to be each of the shifters as well as the battery and control unit. The wireless i actually have working already through basic ASK wireless. I plan to start really working on decoding the signals tomorrow –  zyklus Feb 21 '12 at 9:06
No, I am the one who made them by breaking an old DVD case. Playing with carbon, like Dave does, takes more time than I have available. Good luck decoding the signals. I'd be interested in hearing how you get on. –  Mark Feb 21 '12 at 20:32

RF Solutions' RF800/RF600 series could be used as Encoders and Decoders for this type of project. They enable you use the same transmitter in conjuction with multiple inputs/switches.

Checkout digikey and ask for RF800E/D or RF600E/D (E = Encoder, D = Decoder) - I bought a couple of those and they are pretty easy to implement.

-
What kind of specs do we get with whose modules? –  Kortuk Sep 29 '12 at 13:25

## protected by Community♦Oct 18 '13 at 16:20

Thank you for your interest in this question. Because it has attracted low-quality answers, posting an answer now requires 10 reputation on this site.