Robust, High-Uptime, High-Reliability, Low-To-Moderate Throughput Wireless Solutions?

So, I realize right off the bat that I've probably made everyone laugh with my list of requirements for this solution, but I figure it's actually worth a shot to ask.

Without going into too much detail, here is what I'm looking for:

I have a client with an "industrial" type site -- very spread out with a total fenceline measured in miles. Expectation is not necessarily to monitor the entire perimeter, but rather to capture certain hot spots such as intersections, gates, etc. There is a high-possibility of interference due to the nature of the site, which will hopefully be discovered here soon upon further investigation. Typical distances for each point-to-point can range anywhere from as little as 150' to up to 3000'. Direct line of sight will not be a problem.

Most likely looking at single camera deployments at each location, which will almost for sure necessitate a point-to-point relationship. Power will obviously be available at the camera location, and the assumption for now is that all signals will be received at a location with LAN connectivity. Targeting the highest resolution possible, but likely in the 2MP/1080p range. High framerates will be required due to roads and intersections, which will increase throughput requirements.

Based upon research, I am planning to utilize either an Axis Zipstream camera (eg P-1365-E) or a Bosch NBN-932V due to lower bandwidth capabilities, although this may be negated somewhat due to the high framerate requirements.

Obviously there is probably not a wireless solution on the planet that will provide 100% uptime. But, we are looking for something as close as we can get. I'm not saying that money is no object, but the money is not as important as getting a product that will work.

Thoughts? Recommendations? Additional questions for clarification?

Appreciate all of your help in advance!

Login to read this IPVM discussion.
Why do I need to log in?
IPVM conducts unique testing and research funded by member's payments enabling us to offer the most independent, accurate and in-depth information.

********** **** * **** ** ****** ******** ********. **** **** great ********** *** *********** ** ************ **** ************. **** *** very **** ******** & ******. **** ****** ******* ******** & no ******* ****** ***** *** ***** ** *** ******** *****.

*********. **** ** * ******* ***** ***** ** **********. ***, I **** ** **** *****.

*** **** **** ***** *** ******* ***** ** *******? ** seems **** **** *****, ***** ******* ******* ** *** ************** to *** ****** *********.

**** *** **** * ******** ****** ** *** ****? ** sounds **** *** **** ** ******** ***** *** *** *** likely ** **** *** ******** **** ******** ******** ********, *** worth ******** ** *******.

**** *** ** * ******** ***** ** ** ********** ** engage * ******* ******** *******, *** ***** *** ******* *** send ********* *** ** **** **** *** ** *** ******/*****, and **** **** *** ****** **** **** ** ***** *** way.

** *** **** **** ********* ** *** -- ***** ***.

*****, ** ********* *****, *** ******* ***** ** ******* ** very ********* ********* **** ***** **** ** *********** ********* ** run *****. **** * *** **** ***** ** ** ******, that's ** ********* ** *** ******** **** ** ****** -- the **** ** ******* ***** ** **** ** ***** ********* would ** ** *** **** ** ********* *** ****** *'* guessing. ****, ** **** *****, ***** *** ******* **** ***** just **** ** **** ** ********** **********.

** **** *** **** * ******** ****** ** **** *****. That **** ** *** **** ****. **** * ******** ** interference, * *** ************ ********* ** ********/** ************, ** * could *** ** ***** * ***********.

*** ******* ******** ** * ******* ******** ******* -- *** that *** *** ******* ****** *** ****** *** ******** ****. I ** **** **** ********* ** ******** * ************ *** a ****** *** *********** **********, *** * *** ****** *** everyone's *********** ** *** **** ******** ******** ** *** *** who * ****** ******. **** ** *** ** **** **** I **** **** **** ****, ********* ******* * **** ******** refused ** ** ** ** ** ***. :)

***** *** *** *** **** ****!

** *** **** '************' ** *** ***** ** ********* ********* or ** **********?