Wireless Access

last person joined: yesterday 

Access network design for branch, remote, outdoor, and campus locations with HPE Aruba Networking access points and mobility controllers.
Expand all | Collapse all

VisualRF vs Real World

This thread has been viewed 3 times
  • 1.  VisualRF vs Real World

    Posted Jul 30, 2014 12:46 PM

    How do you guys handle what Airwave and VisualRF are telling you about your equipment and placement, and what your clients/customers are reporting?

     

    I use Airwave and VisualRF to plan out and monitor my locations, but recently I've gotten some complaints.

     

    Both locations were recently redone to provide better coverage, as well as supplemented to provide dual-band coverage (AP-105).

     

    Airwave is showing full coverage for both locations, but complaints are that there is "no coverage" is certain areas.

     

    For now I'll just focus on one location which is a 45 x 80 main building, with a 48 x 43 addition.  I use the assumpution that the walls between building and addition are concrete.   The location is "L" shaped and is covered by 3 APs (1 AP-93, and 2 AP-105).  The AP-93 is in the corner intersection, with the AP-105 being at the "tips" of the "L."

     

    The 48 x 43 addition is covered by an AP-105.   Specific complaints are that 1 office and 1 cubicle space have no coverage.  The AP is centrally located in the space, and the spaces that are complaining are approx 15-20 ft away from the access point.  My understanding is that an AP-105 should have no problem covering a 2100 sq ft space.

     

    With nothing else to go on other than "it's dead space, no coverage, fix it now," how do I justify making a change.....and what type of change?  Add another AP?  Move the existing?  If move, based on what?  Airwave/VisualRF is already telling me that I have full coverage.

     

    Airwave/VisualRF map is attached.

    Office floorplan is attached - reported problem areas circled in red.



  • 2.  RE: VisualRF vs Real World

    EMPLOYEE
    Posted Jul 30, 2014 01:04 PM

    Firstly, looks like you forgot to attach the files.

     

    VisualRF is pretty good and in my experience is the most accurate of the various tools available, including compared to the very expensive Airmagnet planner.

     

    That said, it is predictive in nature and is no substitute for an on site analysis using an AP on a tripod.

     

    You said that the AP93 is on the corner of the L shape. That generally should be avoided due to potential 'hidden node' issues, though that is not a hard and fast rule as well.

     

    There are many other things that can influence wireless performance and signal strength is only one element of that. 



  • 3.  RE: VisualRF vs Real World

    Posted Jul 30, 2014 03:22 PM

    Hmmmm, I swear I attached.  Let me try again....



  • 4.  RE: VisualRF vs Real World

    EMPLOYEE
    Posted Jul 31, 2014 02:55 AM

    where is the office and cubicle space they are complaining about no coverage?



  • 5.  RE: VisualRF vs Real World

    Posted Aug 01, 2014 12:27 PM

    Okay, looks like the files attached.

     

    I realize there can be many factors affecting wireless, unfortunately do a site survey for each location isn't really an option - 180+ locations (furthest locations are 4-5 hrs in any direction) + 1 guy doing wireless and other network related items = do the best you can with what you have.  :)

     

    Locations have had wireless for 12+ years - going from a "cover the open, cubicle areas" mentality, to now finally a "provide full coverage" objective.

     

    Airwave and VisualRF is my best option for planning deployments.   I know that deploying too many APs can cause just as many problems as not providing enough coverage.

     

     



  • 6.  RE: VisualRF vs Real World

    Posted Aug 01, 2014 12:34 PM

    Sorry.   The problem areas are the red circles on the floor plan jpg.

     

     

    I just find it very strange that this location has completely ignored 2 requests for client information (either user or device type).

     

    They have an AP-93 that was left behind (miscommunication by the cabling vendor) that is in the affected area.  It was disconnected, and I believe unprovisioned, but the office has specifically asked for this to be turned back on to see if it fixes the problem.

     

    It is actually located further away from the problem areas - it's basically where the "addition" label is on the floor plan.

     

    I've given them an additional license and told them to let me know when they hook it back up.  I don't imagine it will do anything but cause more problems, but that is what they were pushing for. 

     

    And again I requested client information so that I can actually try to do some troubleshooting......just more fun with working with these invisible, and magical, network connections.  :)    

     

    Sorry, occassionally I get frustrated by my clients....



  • 7.  RE: VisualRF vs Real World

    EMPLOYEE
    Posted Aug 01, 2014 02:03 PM

    I personally find it highly improbable that the two red cicled areas are having signal or connectivity problems being less than 20-25ft away from the AP in their area. If that AP is NOT connected then it's certainly possible they could have issues based on distance, walls, line of sight, etc. 

     

    Geting the real client info, what SNR they are seeing, etc will be the next steps though. But if there is NOT an AP deployed in the nearby red circle, that might need to be fixed to resolve the issue.



  • 8.  RE: VisualRF vs Real World

    Posted Aug 01, 2014 05:26 PM

    Thanks.  I also find it highly improbable. 

     

    I've checked the controller and I have 4 clients using that AP at any given time, with a max of 8.

     

    I've gone so far as to reboot APs while clients are attached and watched them connect to the next closet access point....so I'm pretty sure coverage isn't an issue.

     

    We just added APs and repositioned back in Nov 2013 to provide full coverage (previously we never cared if we covered the offices as they are hardwired), and I haven't heard a single complaint until the office sent an email to an executive about their poor wireless coverage in the office that "has been going on for ages...."

     

    I've let them know that I won't be able to provide any further help until they are willing to provide me with client detail.



  • 9.  RE: VisualRF vs Real World

    Posted Aug 01, 2014 07:14 PM

    How many clients are there, and what bandwidths are they consuming?

    For that matter, what channel/speed are they using?

     

    We found that a legacy-b-device could tie up an AP93. Consider the situation where a b-device hooks on to the 93 and an a-device has to look to the next AP over.

     

    You should be able to follow the users associations and reported signal capability/experience from the controller.

     

    Also, to you point on surveys: We did a predictive for our first building (nearest to the office) and after installation did a not-so-scientific walkthrough, but collected enough data to improve our modelling assumptions for VisualRF, then planned the next site (also nearby) etc.

    After the third location we had our assumptions tuned well enough to hit the user experience pretty squarely.

     

    We've deployed 70 or so locations so far on the way to 100 and are doing purely predictive after the 4th. Occasionally we are surprised by a metal air-duct where the drawings don't show one, or a metal shelf system where we expected a sheet-rock wall, but mostly we're hands-off and no visit required now.

     



  • 10.  RE: VisualRF vs Real World

    Posted Aug 05, 2014 05:04 PM

     

    AP in the problem area is AP-105 - 2.4GHz is channel 11, 5GHz is channel 48.  3 clients using 18.04 Kbps.  dBm of -67, -69, and -56

     

    AP at the intersection of the L is AP-93 - 2.4GHz channel 1.  0 clients on this AP.

     

    AP at top of the L is AP-105 - 2.4GHz channel 6, 5GHz channel 153.  1 client on this AP, using 0.006 Kbps. dBm of -78

     

    Of course I'm still waiting on client information on the users that are having the problem.

     

     

    As far as the survey goes, I'm pretty much in the same boat.  When we moved into our new building it was recommended that we put about 30-40 APs per floor - building is approx 410' x 125'  and 3 floors.   That was the recommendation from Cisco.  We basically ignored that and used the ever so scientific method of drawing circles on the floor plan.....using common sense when it came to elevator shafts and stairways.   We have a total of 27 and manage to cover all required areas.

     

    I did have inSSIDer and  Ekahu heatmapper that I would use if I had to make a site visit, but lost that in my upgrade to Win8.

     

    I've got 180+ locations, with about 100 of those in Airwave and having been adjusted for coverage using Airwave suggestions and previous real world experience.   To date I've NEVER had any of these problems related to our equipment.   We haven't done any firmware upgrades in at least 5 months, so I can't understand the recent complaints.  

     

    I'm basically at my wit's end of dealing with users that like to complain, yet refuse to provide any details.  I'm currently working on an AC migration, so I can't drive 250 miles to a site to sit down with a user and find out the problem is with their equipment (really, isn't it always the user's equipment?)

     

    Unfortunately we don't have license for WIPS or RFProtect to give me any better insight into what is happening at my locations, but we've enjoyed 10-12 years of relatively problem-free wireless access.

     

    Sorry for the frustration....this is 3rd office I've had to deal with this type of issue - all involve users sitting no more than 20 feet from an AP, and all involve users that refuse to provide any information or do anything other than complain to upper management that "nothing works....at all!," and all involve APs that appear to be working fine for other users in the area.   :)

     

     

     



  • 11.  RE: VisualRF vs Real World

    Posted Aug 05, 2014 06:33 PM

    If you've got an Android tablet, you can run the Aruba Utilities App and it makes an excellent WiFi diagnostic tool - leveraging the controller and Airwave for heatmap/channel/signal correlation.

     

    I've occasionally sent a remote technician into a store or office with a tablet to do reconnaissance.



  • 12.  RE: VisualRF vs Real World

    Posted Aug 06, 2014 05:28 PM

    Thanks, I'll take a look.   It's been a while since I checked the Play Store for Aruba apps - I believe they were still experimental the next time I checked.

     

     

    Anyways, turns out the user with the problem had 2 wireless cards in the desktop - an external USB along with an internal card.   They were taking turns connecting to our captive portal network and after authentication the other card would take over and cause all sorts of problems as that connection was not authenticated

     

    As with most of these problems, the symptom of "absolutely no coverage in the office" was actually "it says I have full signal, but no connection." 

     

    Funny how easy it is to find these things when you have a little bit of information....like user ID.