Monday, April 18, 2011

Cisco WCS calibrating location tracking for wireless clients

The Wireless Control System Configuration Guide goes over how to manage RF Calibration Models, but it does not however describe how long the process takes, or what exactly it entails. I will endeavor to describe the process according to how I've calibrated RF models.  I do not know if how I'm doing this is correct, this has been a matter of trial and error. You don't get the opportunity to calibrate RF deployments too often, and the number one reason for that is most likely how long it takes to complete the calibration.


I haven't had much luck using the linear calibration model, so I use the point calibration model instead. I configure my wireless card to operate as an 802.11a client for one set of point calibrations throughout the facility, then I configure it to operate as an 802.11b/g client (only) for the second pass at the calibration process.


I don't stop calibrating the floor area until I have covered the floor area with data points from one corner of the floor to the other. I don't know if this is necessary given the paragraph above, but the data collected across the floor area appears as "complete" to a customer reviewing the RF calibration.

Recently I did a full calibration of a 34,000 square foot facility. The deployment consisted of 11 3500i series CleanAir access points.  The time to calibrate from beginning to end was approximately 4 hours.  Two hours to calibrate for the 5GHz frequency, and two hours to make a second pass to calibrate for the 2.4GHz frequency. Each point calibration location sampling took at least two minutes to complete.

Neither of the design/configuration guides tells you exactly what you're supposed to do with the laptop when you're using the point collection model, unless you're really supposed to pirouette while holding the laptop. I tried to follow this example for the first calibration I did - it just ended up making me dizzy. Now I stand in one place and change the laptop orientation while changing the direction I'm facing. I've found that if I hold the laptop in the same orientation the data point collection fails quite often.

I made a video showing all the stages of the calibration process to help explain how the process is done, and the method I've used for collecting data points.


There are several important bits of information spread throughout the WCS 7.0 Configuration Guide and the Wi-Fi Location-Based Services 4.1 Design Guide.  I will list them below.

"Due to an open caveat1 concerning the use of dual-band calibration clients and performing a location calibration data collection on both bands simultaneously, it is recommended that calibration data collection be performed for each band individually at this time. When using a dual-band client, use either of the following alternatives:
1.Perform the calibration data collection using a single laptop equipped with a Cisco Aironet 802.11a/b/g Wireless CardBus Adapter (AIR-CB21AG) on each band individually. For example, proceed to disable the 5 GHz band and complete the data collection using the 2.4 GHz band only. Then, disable the 2.4 GHz band and enable the 5 GHz band, and proceed to repeat the data collection using the 5 GHz band only.
2.Perform the calibration using two people and two laptops. Each laptop should have a Cisco AIR-CB21AG and be associated to the infrastructure using a different band. The two calibration operators may operate independently; there is no need for them to visit each data point together. In this way, a complete calibration data collection can be performed across both bands in half the amount of time as option #1 above."

and

"Temporarily disable Dynamic Transmit Power Control (DTPC) prior to conducting calibration data collection. DTPC must be disabled separately for each band using either the controller GUI, the controller CLI or WCS for each controller whose registered access points are expected to participate in calibration data collection. After calibration data collection has been performed, DTPC should be re-enabled for normal production operation.

Ensure that the WLAN to which your calibration client will associate is configured to support Aironet Information Elements (Aironet IE). Doing so will enable the use of unicast radio resource measurement requests during calibration data collection for more efficient operation."

According to the WCS Configuration guide: "Only Intel and Cisco adapters have been tested. Make sure the Enable Cisco Compatible Extensions and Enable Radio Management Support are enabled in the Cisco Compatible Extension Options."

Also of note from the WCS Configuration guide: "The calibration status bar indicates data collection for the calibration as done, after roughly 50 distinct locations and 150 measurements have been gathered. For every location point saved in the calibration process, more than one data point is gathered. The progress of the calibration process is indicated by two status bars above the legend, one for 802.11b/g/n and one for 802.11a/n."


6 comments:

  1. I used a normal Office Chair with wheels when trying this at the office. But it still looks funny.

    ReplyDelete
  2. I've conducted around 50 calibrations to this day and here's what i've learned :

    1) using the Cisco ABG wifi card reduces the collection time and anyway CCX is required , so make sure your wifi has CCX supported in the latest version.

    2) there's no point to alter the computer position . the results will be almost the same if not exactly the same. i would still get to under 5 meters accuracy using this method.

    3) sometimes "out of the box" floor models will get you better results. specially for small areas (and yes you will still have to complete the hit count to 100% even when the area is fully covered after 20 hits)

    4) avoid scanning for rogues for the calibration period.. you can disable the rogue detection per AP .

    5) don't expect the WCS to give you the exact location of non static rogues/clients devices.

    6) and the most important one - try to find some one else to do it for you .. it's PAIN and you look stupid .. (i would tell people i'm checking for RF radiation to entertain myself .. it's sooo boring at times!!)

    ReplyDelete
  3. Did you see any benefits of doing WCS calibration? if there are any?

    ReplyDelete
  4. Calibrating the RF properties of a floor and applying that as the RF profile is supposed to make location tracking more accurate according to the Location Design Guide. I haven't done any before/after testing, so I cannot attest to the level of additional accuracy that is achieved.

    ReplyDelete
  5. Note to myself, the amount of square footage that can be calibrated per hour is 8500 square feet per hour. This is based on calibrating the infrastructure for the 2.4GHz radio separate from the 5GHz radio.

    ReplyDelete