3602i site survey Tx power settings

[tsakoulias]

Hi all,

We are planning to conduct RF site surveys on 10 sites in the UK to  deliver a high density WiFi network (including Voice services) supporting about 20 users per AP radio.

We will use 3602i APs for those networks and the same AP model will be used for the site surveys.

Since we will provide BYOD services too, we had to choose one of the least capable devices that are going to connect to our network as a reference device and check its specs.  iPhone 5 is using a dual band radio with the following specifications:

Band          Tx Power          Antenna Gain          EIRP (dBm)          EIRP mW

2.4 GHz     16dbm               -1.4dbi                    14.6                    about 32mW

UNII1          14dbm               0.14dbi                    14.14                about 25mW

UNII2           13.5dbm            -1.66dbi                    12dbm               16mW

UNII2-e          12dbm               -0.83dbi                   11.2dbm          about 13mW

These are the available Tx power levels for the 3602i AP:

2,4 GHz and 5 Ghz

Tx Power level 1: 17dBm

Tx Power level 2: 14dB

Tx Power level 3: 11dBm

Tx Power level 4: 8dBm

Tx Powe level 5: 5dBm

Tx Power level 6: 2dBm

Tx Power level 7: -1dBm

Tx Power levle 8: -4dBm

As far as i can understand, these are not EIRP values but just the Tx power of the AP radio.

The 3602i is using a 2dbi antenna on the 2.4Ghz and 4bi on the 5Ghz bands. So the EIRP values for the 2 bands will be as follows:

2.4GHz

L1 --> 19dBm --> 79mW

L2--> 16dBm --> 40mW

L3--> 13dBm --> 20mW

L4--> 10dBm --> 10mW

L5--> 7dBm --> 5mW

L6--> 4dBm --> 2.5mW

L7--> 1dBm--> 1.25mW

L8--> -2dBm --> 0.63mW

5GHz

L1--> 21dBm --> 128mW

L2--> 18dBm --> 64mW

L3--> 15dBm --> 32mW

L4 --> 12dBm--> 16mW

L5--> 9dBm --> 8mW

L6--> 6dBm --> 4mW

L7--> 3dBm --> 2mW

L8--> 0dBm --> 1mW

I have been searching various forums on the Internet and the recommended site survey TxPower for high density or Voice networks is about 12.5 - 25mW. Is this the EIRP or the AP TxPower?

Since the iPhone can transmit up to 32mW on the 2.4GHz and 13mW on the 5GHz band, what is the recommended EIRP on each band to conduct the site survey? Since the 5GHz signals don't travel as far as the 2.4GHz signals, Is it correct to use the same EIRP values on both bands?

Would anybody recommend using 5mW EIRP for the site surveys?

Regards,

Theo

[Scott Fella]

The lowest I would go on a site survey is 25mW. This is to ensure proper voice coverage which of course validates data coverage. Now since this is high density, you can still have more overlap than the minimum 20%. For example, I will still survey at 25mW in a school environment that wants an access point in each room. Survey TX power and the final config TX power doesn't have to be the same, but will most likely be higher than what you might set the TX power or what RRM sets the TX power to.

Sent from Cisco Technical Support iPhone App

[tsakoulias]

Do we consider the 25mW as the radio transmit power or EIRP?

In the example above, should we select L3 with 20mW EIRP for the 2.4Ghz and L4 with 16mW EIRP for 5GHz?  

[Scott Fella]

With Cisco it's the radio transmit not EIRP. You don't have to worry about the total EIRP unless you are doing outdoor. Doesn't make very much difference in the indoor.

Sent from Cisco Technical Support iPhone App

[tsakoulias]

But if we select 25mW for the AP TxPower and we are using an internal antenna gain of 3dbi, EIRP will be 50mW.

The survey results will be completely different.

If we go with 25mW Txpower, should we use TxPower Level 2: 14dbm =25mW for both bands?

[Scott Fella]

It's up to you, but I never survey below 25mW radio power. Like I mentioned, you just want to make sure if coverage, the final install will be different. You will not be hard setting the power on the access point but maybe tweaking the cell size.

The 2.4ghz is 2dbi and the 5ghz is 5dbi. If you also look at the voice deployment guide, Cisco states that you should survey at 25mW TX power. Again, I have not seen any issue even with sparse ap deployments and AP's set to TX of 100mW and my iphone or iPad not working. Survey is to provide coverage and you will have a lot if overlap anyways. You can survey at 14mW if you want, but don't freak out when the customer says that's a lot of access points and then want to remove some to meet his budget.

Sent from Cisco Technical Support iPhone App

[tsakoulias]

In the case of the iPhone, 25mW would be ok on the AP side for the 2,4GHz.

This device is using 13mW on the 5GHz.

Does this mean that we will have mis matched power settings between the AP and the device which will cause the device to be able to hear the AP's transmitions but not the other way around ?

[Scott Fella]

You are over thinking it.... just because there is a slight EIRP difference doesn't mean it will not work.  Your iphone works when using a public free wifi, most of those are 100mW.  This is just how RF is, its best to try to match the power, but that is very unrealistic.  Like I said, I would never survey below 25mW.... you are going to doing high density, so your main concern should be density... how many devices per access point.  This will vary from an office area, to a stadium, in which you will have more clients per AP in a stadium than in an office.  If you are surveying for the weakest device and its byod, then you better research the older iphones and survey to that if you want.

Thanks,

Scott

Help out other by using the rating system and marking answered questions as "Answered"

[ERAJAB]

Hi Theo!

Congratulations on a thorough job, you have probably already made your survey, but I would just like to comment:

You are perfectly right that the Radiated Power (EIRP) = Tx Power + Antenna Gain.

Your goal should always be to enable also the weakest device to form a symmetrical link (same EIRP possible from AP to Device and from Device to AP) otherwise you risk having one-way connections as worst case.

Normally you simplify and only look at the Tx Power figures because the device antenna specifics are unknown,

or assume that both the AP and Device have a similar antenna gain and can be disregarded.

In your case, you have managed to get figures for the weakest device, and the antenna gain are by no means the same as for the AP. Hence, I would use the EIRP figures.

For the 2.4 GHz band your device EIRP was 32 mW, the AP should then be set to nearest lower level:

Power Level 3--> 13dBm --> 20mW.

For 5 GHz it depends on the size/complexity of the site, if it is small enough to allow it, Cisco recommends using the UNII-1 band only because of no 802.11h impact there, but you only get 4 channels. If necessary, add UNII-2 channels and get 4 more. UNII-2e gives you 11 more channels, but not all devices support them, and the device you use have

a lower output power there.

Say you decide to go for 8 useable 5 GHz channels (UNII-1 and UNII-2), the device EIRP is 16 mW, AP setting:

Power Level 4 --> 12dBm--> 16mW.

NOTE!

This is the settings to use during Site Survey, and a target value. What power level your AP will get from RRM depends on the number of APs and how well they hear each other, which might be an entirely different issue

In that case, you might have to create RF Profiles and set the minimum/maximum allowed power levels under the RRM tab to get your APs to stay near this target value.

All the Best!

Jan Brannstrom

[George Stefanick]

Not sure if it was mentioned. 5GHz UNII 1 has a different tx power than UNII2/E and UNII3.

http://www.my80211.com/home/2009/10/30/what-you-need-to-know-about-tx-power-and-80211a-5ghz-on-a-ci.html

Sent from Cisco Technical Support iPad App

[ERAJAB]

Correct, there are different allowed tx power levels for the different 5 GHz bands, in this case we were dimensioning a symmetrical link based on what the weakest client device was able to use uplink for a specific band, and I used the figures given by Theo for his device in the UNII-2 band just as an example.