Choice of frequency and protocol on RF links?

I am designing an RF link that is supposed to operate between an embedded unit and a central control/collection unit (i.e. a PC with the link hardware attached). The purpose of the link is to pass some data (in form of strings), and then some control/session parameters. Here are the main features:

- Relatively low data speeds required

- Range of up to 20-30 meters (LOS, but occasional obstructions are likely)

- Needs to handle multiple devices (i.e. units) in same environment (i.e. hitting the same controller)

The RF link component/module will interface with some MC in the embedded module.

I am wondering what choice of frequency and protocol to select:

- 433 MHz (good old AM/FM??!)

- 2.4GHz WiFi (802.11)

- 2.4GHz Zigbee

Main considerations are:

- Cost of component/module

- Ease of design/integration

I am aware that all the above options can and do work, but am looking for the best fit vis-a-vis the above main criteria.

Suggestions and comments? Thanks!

Reply to
ElderUberGeek
Loading thread data ...

In what geographic location do you expect to operate this equipment? Do you expect to write or to buy the baseband encoder/decoder firmware?

Unless designing this stuff is your bread and butter, I'd suggest buying it all off the shelf. There are 900MHz and 2.4GHz solutions that might work for you... an easy off-the-shelfer is MaxStream (but not exactly cheap).

Reply to
larwe
900MHz will not work, that's why I wrote 433MHz.... (ISM also). Well I do not *want* to design the stuff, but do not mind the NRE if resulting BOM is very low... ;)

larwe wrote:

Reply to
ElderUberGeek

The key questions are if the worldwide acceptance is required and if you are willing to apply for approvals yourself. If your quantities are small, then the best way would be using pre-approved 2.4GHz modules like MaxStream or Cirronet. If you are looking for the mass product, then it could make sense to do the RF part yourself.

Here

How much is "low"?

Should not be a problem.

Usually, this is the upper protocol issue.

Very cheap and simple, however not universally accepted.

Overkill.

May be a good solution. However it is not very cheap and not very simple either. Same comments apply to BlueTooth.

Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

formatting link

Reply to
Vladimir Vassilevsky

Thanks Vladimir.

I also forgot to mention that the link needs to be secure..... I wonder, does this eliminate the 433MHz option?!?

Vladimir Vassilevsky wrote:

Reply to
ElderUberGeek

Technically speaking, the security issue is unrelated. The security can be implemented by the software at the upper level protocol.

However a particular country specific regulations may not allow using encryption and such. Also all security related issues are subject to the import and export regulations. You may need a special license for that.

Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

formatting link

Reply to
Vladimir Vassilevsky

WHAT COUNTRY?

Reply to
larwe

A country where 433 and 2.4 are ISM... :-)

larwe wrote:

Reply to
ElderUberGeek

Doesn't help at all.

You cannot - repeat - CANNOT achieve a solution that is completely applicable worldwide, particularly if encryption is required. If you're going to get any more useful answers you need to be more forthcoming with application details.

Reply to
larwe

Hey, sorry, I didn't think it matterd that much... Firstly the country it Norway. Secondly it is not supposed to be a "global" solution (i.e. acceptable worldwide as you say).

My way of thinking, at the moment, is that probably a Zigbee chip/module would be a good way to go since they are (supposed to be?) robust, prices are coming down, and there are plenty of chips/modules/kits. And security is taken care of also. With a "pure" RF 433 solution we would have to tinker more and that it a pain.... Anyway those are my conclusions till now....

larwe wrote:

Reply to
ElderUberGeek

It matters immensely. In the US, it's forbidden to use encryption in the unlicensed bands, for instance. One of my assigned tasks (in theory) is to create a harmonized list of design requirements for ERP, duty cycle, etc. etc. across all the markets my employer services, and every time I pick up this particular to-do, I get daunted and go back to some easier job, like (say) convincing Palestine and Israel to just get along.

Then you have at least one other SRD [non-ISM] band to play with, I believe (I'd have to check to be certain) you have the 868MHz region available. But I agree with you that 433 is probably the right choice here.

For the range you're talking about, ERP requirements would be real low. And I believe Norway in common with many other Euro countries has the nice rule that you get a certain amount of on-time per hour (based on band and ERP) and you can split that up however you want; spend it all at the start of the hour, or spread it across the whole hour. Managing collision is an interesting problem though.

_Maybe_. I'm yet to be convinced that Zigbee will survive industrial/commercial applications of the type we do.

Reply to
larwe

Not for high encryption. 40 bits (WiFi) or lower is fine. As long as the government have enough computing power to break it.

Reply to
linnix

Check Part 15. It's illegal to use encryption on FRS links, amateur radio links, SRD links, ...

Since I don't work in the 2.4GHz bands much (yet) I don't know what special loophole is there for WEP/WPA.

Reply to
larwe

Not quite. The use of encryption is controlled, not forbidden. I.e. you need to obtain a license to do that.

As long as you can prove that you are not working for Iran and such and that you need encryption, say, for the multimedia content protection, you can have the strong security.

Part 15 does not say anything about the encryption.

This is out of the scope of part 15.

Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

formatting link

Reply to
Vladimir Vassilevsky

I'd have to be at work with my notes to argue with you...

Reply to
larwe

As far as the amateur radio bit goes encrytion is no illeagal, as such i.e. You can use ax25 to trasfer a pgp encoded message as anyone who can rx and decode ax25 can see who the signal is from and who it is going to and they can see the contents of the transmission. The contents of the pgp encoded message is iralivant as far as the RF side of things is concerned.

Reply to
This Computer

Do you need authentication or confidentiality? In the latter case the authorities will make less objections.

Wim

Reply to
Wim Ton

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.