Site Survey - TX Power Level

[derobbacher]

Hello Cisco WLAN experts,

this big hospital has invested already in additional Cisco-WLAN-APs which shall be placed in addition to existing WLAN APs

to offer a powerful and redundant "Medical Grade WLAN" to our patients and clinical personnel.

We would like to use Apple IPhones for Skype WLAN Voice telephony and would like to use

Apple IPads for a new electronic Patient Information to allow doctors to show live Videos for example.

But the new Finance Chief Officer is not convinced about our nice strategy and is demanding a Site Survey on beforehand

and after a small test Installation before allowing further Investments into a big rollout.

Therefore I need some recommendations how to "tune" the existing Cisco-WLAN-AP-installation to show the right results

afterwards, that will free the money for the implementation finally. 

Does it make sense to statically configure TX Power Level on the WLC during the Site Survey ?

To which Power Level should I adapt the sending power in 2.4- and 5 GHz-Band ?

FIXED to 1 (100mw), 2 (50mw), or 3 (25mw) ?

I haven't found any info about the TX Power Levels of Apple-devices by the way.

I would assume that I have to use an identical or lower TX-value on the APs.

Is there anybody out there who knows these details ?

Thank You for Your tipps

Greetings

Wini

[Leo Laohoo]

What I would like to know is this:  What is the WORST wireless client the hospital have?  2.4 Ghz-only WiFi RFID tag?  WiFi phones that only operate in 2.4 Ghz?  Maybe a WiFi phone that only uses 8 x 5.0 Ghz channel and requires fixed Tx and fixed channel assignment?

[Scott Fella]

Just to add to Leo and your comments, a site survey and small POC is the right path before spending money. If your design and implementation which you presented to leadership is solid, you would not even be worried about the POC and or using static TX power or any other settings to influence your POC. A proposal of doing something new, should account for fixing the old design and issues while also improving the coverage, roaming, security and user experience for example. Experienced engineers whom have worked in that vertical and others know what they need to design and implement for and would never POC to falsely get the data they want to show leadership. That will fail when it gets deployed and then how will you answer them when they are not happy.

[derobbacher]

Hello Leo and Scott,

thank You very much for Your comments.

What do You mean with POC ? POC= Proof Of Concept  or Plan Of Correction ?

In parts of the Hospital, WLAN Phones from Unify Siemens, a copy of a Ascom WLAN Phone, are in place with 23dBm.

Only useable on 2,4GHz with the existing small number of WLAN APs. On 5GHz roaming problems occur.

Target is to migrate to 5 GHz and use Apple devices for Voice for example Skype for Business.

A site Survey done 2 years before by externals came out with the result to increase the numbers of WLAN APs by factor 2.5

to accomplish this target.

In general, shall we statically fix the TX-Power during the Post-Survey ?

If Yes, to which value for voice-WLAN-coverage in 5 GHz ?

Also a question to the result of a passive Site Survey, which I have never done yet.

For example the Ekahau tool.

Is is possible to simulate the results afterward within that tools?

Let's assume, the Post Site Survey has been done using existing Cisco WLAN APs at Power Level 1 or the Level just calculated by RRM.

Is it then possible to simulate the results of the Site Survey using only 50% of the TX-Power or only 25% of the original to show the outcome ?

(Compareable to Cisco Prime Map, where I can simulate different heatmap coverages by choosing a different RSSI cutoff value) 

Thank You for Your comments

Wini

[Scott Fella]

Proof of concept. A two year old survey is kind of old as things have changed, access point models, controller models and code, etc. when you survey, you typically survey on the low TX level to allow rrm to increase if needed but also to lower if needed. You set the TX power to a level in the middle of you believe rrm will set it there, this is how you get a realistic result point in time. That is the key, the survey or PoC is point in time. Ekahau allows to adjust, but it’s predictive and only if you setup the project properly. Prime heat maps is also predictive and doesn’t really show what your coverage is, it’s just an overview.
My question to you is, you had a site survey performed two years ago, why don’t you use that to show your leadership? If that didn’t provide what you or they are expecting, then was the site survey not done to your requirements? It’s easy to show green and that you have great coverage and performance from a site survey, but when everything is in place, you better make sure you account for everything from your survey to make sure all the requirements are still met. If not, you will be fixing the environment after and for a long time.

[launchnow]

RE: Does it make sense to statically configure TX Power Level on the WLC during the Site Survey ?

Consider above ONLY. Think about it. After running RRM it will leverage TPC to do all that's possible to obtain cells with ~-70 at the boundaries.  There might be a bad design in that an AP should have been placed but was not or even a down AP. RRM tries to make up for it.

Putting all APs on the same channel at 1/2 the max power for the channel seems to make sense.

Comments?