AP placement and performance: Wall vs. Ceiling mount?

Unanswered Question
rnigam Sat, 07/14/2007 - 07:03

The RF performace of an AP depends upon a lot many factors like enviromental conditions, the surrounding objects and the adjacent APs. Also it would also depends upon the location of clients. Kindly be more specific.

Hi,

I understand the need for enviromental specifics but I'm looking for implementation standards I can communicate to installation teams -- what elements of the enviroment would tilt the decision in favor of wall versus ceiling?

Our deployments are typically in hallways with drop-ceilings and drywall interior construction. Does the height of the antennas at say 7 feet vs 10 feet materially impact coverage given standard dipole beamwidths?

What about ceiling tile replacement enclosures with punch out holes for the antennas - does their metal construction affect signal propagation?

Does having APs on the wall create poor reflections, as opposed to more central placement on the ceiling?

Of course there are other factors to consider - ease of access vs. security and aesthetic preference.

What I'm after is a rule that can say "ceiling mount when you can, wall mount if you must" (or vice versa).

Thanks,

--Bruce Johnson

scottmac Sat, 07/14/2007 - 13:25

Well, it's kinda like the general rule of mountain climbing: "Don't fall off the mountian, especially from really high up"

Or what I learned as a pilot: The First Rule of Crashing Properly: " Hit the softest possible object(s) at the slowest possible speed" (and it's always preferable to be able to re-use the airplane).

The first rule of wireless is "line of sight" (and for distance, some Freznel considerations).

If you get best LOS from the ceiling, use it. If you have better LOS from the wall, use it.

Some antennas are best used from the ceiling, like an omni ... because you can't get 360 degree lateral coverage from the wall.

Antennas like patches (which, by the way, in most cases, would be the best choice for a hallway) are designed for wall mount. There isn't any purpose for sending a concentrated beam of rf into the floor, right??

Thare no tricks for the installation team ... they should put the stuff where the site survey determined it should go.

The stuff is decided by the design folks (omni, patch, secotr, panel ...) according to the needs of the area and application purpose.

The closest thing to a rule of thumb is that there's no rule of thumb. Even a big, drywall box with some desks can be a challenge ... EVERY SITE SHOULD GET A COMPREHENSIVE SITE SURVEY (repeat it again after me ....)

By the time it gets to installation, the APs, the antennas, the cabling, the power, the environmentals (temp, humidity, etc) ... all of that is determined (by design, according to the customer budget, as figured from the SITE SURVEY.

All the install folks get to choose is when to start, and how long to work.

There are no (safe) rules of thumb, all sites require a survey. {repeat}

Good Luck

Scott

rob.huffman Sun, 07/15/2007 - 06:17

Hi Scott/Bruce,

Awesome answer! Full of great ideas and fantastic humor :) 5 points all day long!

FWIW - I have always preferred wall mounting AP's, but, that being said, our split is still about 50/50 due to all the factors involved.

Take care,

Rob

Thanks Scott,

You must be watching the Tour de France :)

I like LOS as a determining factor - its the single variable I am looking for.

Surveys can involve relatively arbitrary AP placement sometimes. This can easily lead to self-fulfilling survey prophecy.

We deploy in hospitals, where hallways are the only practical placement location. This has worked well for coverage in adjacent rooms.

We are also concerned with the additional impact of location services, however, which demands diverse (internal and peripheral) quadrant coverage (getting antennas in peripheral corner patient rooms seems inevitable). This is contrary to standards for data/voice service (linear deploy with dipole antennas).

When LOS and positioning diversity are required, what's a surveyor to do? Zig-zag like the mountain stages of the TDF?

Does anyone have a best-practice methodology for multi-floor site surveys with all of these factors in mind?

--Bruce

scottmac Sun, 07/15/2007 - 13:06

I understand and believe hospitals have to be one of the tougher installs, you have some work cut out for you, no doubt.

It might be worthwhile for you to check into someone like Andrew Corp for your antenna stuff.

What I'm thinking of specifically would be something like a "Radiax" signal distribution system.

Radiax is like a "hard line" type of coaxial cabling (similar to the hardline that cable TV uses for copper distribution through the neighborhoods). The main difference is that "Heliax" and coax are meant to keep the signal in, Radiax is meant to "leak" the signal out.

Radiax is used in long tunnels, for example, to provide radio and cellular coverage in the tunnel.

Radiax is also used in internal paging systems ... one long run through the elevator shaft and all floors are covered (for paging).

Perhaps you can partner with or contract Andrew to provide your antenna and signal distribution.

If you use a signal distribution system, keep in mind that it *must* be re-certified to federal compliance, since it was not certified as OEM. I believe Andrew is qualified and licensed to provide that federal certification ... talk to them.

The other advantage to Radiax, if it fits this implementation, is that (all according to how it's implemented) should permit lower power levels, better "mix" control for the triangulation systems, and likely better security and access control.

Aside from that, all I can think to do is set it up in a Lab, pilot it out at a customer location (maybe provide a credit to the customer for the time taken to pilot)

It's just one of those things that you can't really fudge. It's going to work properly, or it won't. With the vast variation in customer environments, there's only one way to guarantee success ... and that's to scope it out in the actual environment.

Tour de France?? Bicycles?? You got time to watch a bunch of grownups peddling around in the (presumed-to-be-foreign) countryside?? (On TV????) What's next, Golf?? Poker?? PRofessional Jogging?? When will the emadness end .... ;=)

Good Luck

Scott

I heard the leaky coax will "smudge" location values, unless you go with fixed antenna taps ala InnerWireless. I agree though, the location stuff, like optimal AP placements, needs to be determined "on-location." It seems to me someone out there should have published a detailed methodology for conducting multi-floor site surveys by now. I know a lot of companies that provide survey services may not be interested in this getting out, or maybe they're afraid to be asked what their methodology actually is.

Thanks for your always enlightened observations Scott.

I'd much rather be in that foreign countryside sipping burgundy watching those pedalers get to the top, after hiking the mountains first of course. A few of those guys forgot your first climbing rule today and had to take themselves out of the race.

I feel like them when I have all these rapid deployments, and management that wants cookie-cutter standards that can scale for all engineers and work perfectly for every application and location.

I think the best thing to do would be to outsource the surveys, do all the building footprints and then roll them out as requested.

Thanks again,

--Bruce Johnson

Actions

This Discussion

 

 

Trending Topics: Other Wireless Mobility

client could not be authenticated
Network Analysis Module (NAM) Products
Cisco 6500 nam
reason 440 driver failure
Cisco password cracker
Cisco Wireless mode