Any long distance NLOS wirelss systems available

[Sam Cheung]

Any proposal/solution for a wireless transmission of data with the following conditions:
1. A production plant is 10km from its headquarters need to send its plant real time operational data to its headquarters
2. There is a small hill in between 200m high
3. All proposed outdoor equipment will be subject to -40degC ambient temperature
4. Only 2.4GHz system products will be considered and complys with FCC Rules

Thanks for any suggestions.

Sam

[ccrum]

You could use 4 802.11 APs acting in bridge mode using some high gain dishes. Put one at each location, then two back to back on the hill you are trying to clear. It could be done this way for about $1200 with Smartbridges AirPoint Pros. Smartbridges.com for the APs

[Sam Cheung]

Thank you for the suggestion. I have further questions as follows:

1. The plant operating data is mainly one way direction to the HQs via the small hill tower (to be built with about 10m high for installing the back to back APs). Should I set all the 4 APs to be bridges or one AP and one client?

2. How should the 2 back to back APs on the hill be connected, by one coaxial cable?

3. How to determine what dBi (13dBi, 18dBi, 23dBi etc.) is sufficient for 10km wireless transmission? Is there a general guideline?

4. SmartBridges is rolling out its airPointPro Total AP which integrated a 13dBi antenna, will 4 of them are capable of transmitting wireless 10km apart without using further antennas?

Thanks for clarifications.

Sam

[ccrum]

The directionality of the data flow is not important as these are two way devices. But to answer your question the AP's would be set up like this:

Plant1 AP in Bridge Mode shooting to hill. AP facing this should be in Access mode. If you use the Air Point Pros, you will then use a cat-5 crossover cable to connect the two back to back AP ethernet ports together. Run the AP shooting to plant 2 in Bridge mode and the AP at Plant2 in access mode. In reality, I think you can operate them all in brdge mode, but haing never done this I can't be sure.

I would use as high gain dish antennas as you can find for this. Air Point has a 24dbi dish listed that is certified with their equipment. This will of course require a bit of aiming precision, but it shouldn't be too hard. I wouln't use the 13dBi pannel that comes with the package. It is too broad a beam.

Hope that helps.

[Sam Cheung]

I can understand that the narrower the beam angle the farther distance the signal can be transmitted and the more difficult to align. Is there any calculation that how far can this 24dBi antenna transmit? How to determine how far capability of a particular antenna can do. Would you recommend a book on selection of antenna?

[ccrum]

Without getting into a lot of deep RF theory...the distance a signal can travel is related to many factors. The gain of the antenna can be seen as a focusing of the energy into a specific region. In this way it can increase the field strength over a given area. However, the amount of power transmited does not always mean that you will have a good signal at the other end. You could have a received signal level of -50 dBm but have crappy throughput because of interference. For those of us in communications, we like to talk about reliability instead of power becasue reliability or more importantly, reliability in the presence of interference will determine if you are going to have good communication.

Under idal conditions i.e. free space path loss, it would be easy to detemine the path loss and therefor the distance over which we could reliably communicate. However, these conditions rarely apply. To determine how far a signal will go, you need to either:

1. Determine what reliability level you think is adequate (most people choose 99.999 for point to point links) and calculate a link budget that will tell you what kind of received power you will need to make the link happen

or

2. Caluculate reliability "on the fly" to figure out if you will achive communication.

To set up a link budget you need to know a lot of factors. Lots of things can effect high frequency links. Moisture in the air, the type and elevation of the ground clutter, the terrain between the links, the exernal noise density, the noise figures and/or sensitivities of the recievers, etc.

So the answer to your question is not as easy as plugging a few numbers into a formula. That being said, there are average distances given about how far an 802.11b link will travel. These are pretty good numbers to follow. I personally would not take a point to point 2.4 gig link over 15-20 miles wihtout more power available. I'm sure ther are some who have gone farther, but their links probaly are not 99.999 percent reliable.

As for choosing the right antenna, I have personally designed over 1000 cell sites and I still don't always choose the right antenna the first time. There is no definitive book on how to do this. Antenna selection depends a lot on what you want the system to do. Antennas, if chosen correctly can act as spatial filters both minimizing the amount of interference you cause and mitigate the amount of interfernce you receive. In my opinion, it's not something that should be taken lightly. Antennas are perhaps the most important part of your network. I would reccomend spending a little bit of cash up front and having a professional RF Engineer help you with this part of your system if you are unsure about it. It will most likely save you a lot of headaches and maybe even some money down the road.

[Sam Cheung]

Thanks for the reply. Per your advice, as I am not confident about the antenna, I will it to the experts. Thanks again.

[wirelessboy]

dear ccrum
one of the longest posts i have ever seen, good work. great to see someone explain RF so much.

dear sam

forget the experts for now, just install an ANDREW 23 DBi SEMI-Parabolic antenna, and sleep peacefully!, and do tell us what was the final solution.

bye

[Sam Cheung]

Thanks for the advice. Our customer has deferred the whole project installation in which wireless is a minute but very important part of the project. Will keep you guys posted the outcome. Thanks again.