Remote instrumentation via wireless link?

Thread Starter

Lesaid

Joined Apr 16, 2008
9
Hi

I'm working on a project involving an outdoor solar powered device that I want to interrogate remotely from a PC, both to download control data and retrieve instrumentation. A wired connection is not possible - but a wireless link of maybe 20-30 metres would do the job.

I have a prototyped device which has a single data line presenting an analogue voltage representing a measurement - six address lines allow the desired measure to be selected (using analog multiplexors to switch), and will need to add some sort of AtoD conversion before transmission - 8-bit should be adequate.

I am unsure as to the best approach to provide the remote wireless link. I have been wondering about ...

(a) buying a wireless LAN card, which will talk to an IP wireless network and present an RS232 interface to the electronics (a bit nervous about the current load from this though - might need to trigger it with a separate low current receiver rather than leaving it permanently active). And I've never worked with the electronics side of RS232 either though from reading the web, it doesn't look too hard.

(b) using two matched pairs of AM or FM transmitter/receivers on the 433 Mhz bands, that effectively links a single data-line in each direction, with an arrangement of shift registers and logic to awaken the transmitter and send a sequence of all metrics over the link when triggered

(c) same as (b) but using UARTS for the encoding / decoding

(d) using a PIC processor to provide/control the link, either totally in software, or with some help from external logic. The PIC would be powered up when the receiver received a trigger rather than running all the time.

I can envisage how all of these would be possible - however, while I'm comfortable with basic electronics, TTL/CMOS logic, op-amps and such, I've never worked with serial links or UARTS before. Neither have I used a PIC before, though I'm fluent with programming in assembler and C/C++ so confident that I could learn to program a PIC easily enough.

But, I would like to pick the best approach at the first attempt and am unsure. Can anyone out there offer some ideas as to how I should choose between them (e.g. any pitfalls I should be aware of), or suggest any other alternatives.

By the way, this device is solar powered, and working off a float-charged 6V lead acid battery - only a small panel so on a tight energy budget.

Any thoughts/ideas?
 

Thread Starter

Lesaid

Joined Apr 16, 2008
9
Thank you taking the time to read my post and sending that link. I've taken a look - but those devices are I think, over-engineered for what I'm looking for - and fairly expensive (particularly as I think I would need two TxRx pairs, switched alternately. It looks like they would also be very expensive on current - I was hoping to find something that would be under 10mA - if possible, under 1mA when quiescent, listening for a 'wake-up' trigger, rather than being up around 100mA or more. Doesn't matter if the transmitter is thirsty (within reason) - it would only be active for a few seconds at a time or less, when triggered.

I've also got only a 6V supply - I'm trying to avoid the solar panel changes that would be required to raise this - though I'm already using a little voltage doubler based on a CMoS 555 chip, to provide a regulated supply above 6V at a few mA. Could do something similar with a transmitter and a more powerful driver to a doubler if necessary.

I have been looking at the sort of devices produced by www.RFSolutions.co.uk - and wondering about the simplest ways of encoding the data over such a link. I only need a range of 20-30 metres.

The IP wireless LAN solution is attractive as it avoids any need for electronics at the PC side, and the range from the 802.11b LAN is ample - but again, it is thirsty - would need to be triggered by a more frugal wireless receiver !

One thing I'm unsure of is the best way of encoding/decoding the data - on the face of it, the UART seems the obvious way - but I've never worked with them before, and wonder if they really are as simple to use as they seem. I suspect that I'd find it hard to diagnose the problem if a design using them didn't work !

Regards

lesaid
 

brumac57

Joined Apr 10, 2008
18
Hi there,
I've done a bit of work on short range RF links that send low volumes of data using battery powered micros

try the TLP434A (and matching receiver) from Laipac lots of places sell them, low cost and low power.
A 1/4 wave antenna can be a piece of wire. They can easily cover the distances you mention.

You can hook them directly to a pin on the micro (and power them from a pin). I use AVRs from Atmel and think they are better/easier than PICs (but I'm biased).
The AVRs have several low power sleep modes.


You can program the AVR using open source C tools WinAVR, GCC, PonyProg ...

The area you will have to be careful is the encoding.
You can't just squirt data out the UART to the Tx and expect it to arrive at the Rx!

The Rx UART will see a continous stream of noise (left over from the Big Bang I believe!). You will have to make your data stand out from the noise.

You will need to read up on encoding schemes, the most common is Manchester Encoding (ME). Then you will have to packetize your data and insert error checking like a CRC or check sum.

As your Tx will only be talking and not listening I suggest sending the packets numerous times because the RX has no way of saying "please send again"

There are C source examples on the Web for this sort of stuff.

Good luck !

Cheers Bruce
 

brumac57

Joined Apr 10, 2008
18
I just reread your first post and see that you are considering being able to communicate back and forth.

In which case you can request a retransmit if a CRC fails.

If you want to skirt around the ME CRC stuff you could use an RXD-315-KH (for example) from Linx Technologies.

These contain the wireless stuff and do all the encoding for you.

The only thing is that they are designed for keypads so you would have to use 8 lines from the micro to set a bit pattern on the input lines on the Linx (that matches you data byte). when you assert the TX line that exact same pattern will appear on the RX unit outputs.

A bit fiddly but they have very reliable transmissions.

One more general thing - don't use the same antenna to transmit and receive, use seperate antennas

Cheers
 
Top