Please help me interpret noise in the Santa Cruz mountains (roughly-75dBm across the 2.4GHz spectrum

Discussion in 'Wireless Internet' started by Vinny P., Jul 30, 2012.

  1. Vinny P.

    Vinny P. Guest

    I just switched to a new WISP in the Santa Cruz mountains and was told to install the Ubiquiti Nanobridge M2-18 Outdoor MIMO 2.4GHz 18DBI.

    This new UBNT NB-2G18 comes with "AirView", which is apparently
    a new spectrum analyzer feature.

    Looking to debug why my "Capacity" & "Quality" results are lower
    than a neighbor's with the same equipment and only slighly shorter
    LOS, here are my AirView result from a moment ago:
    http://www2.picturepush.com/photo/a/8862580/640/8862580.png
    Note: The WISP is on a Rocket M2 + Rocketdish on channel 5 about
    3.2 miles away, LOS.

    My neighbor, 0.2 miles closer to the access point, has almost double
    the quality & capacity numbers, and 3dB better signal strength ...
    using the same equipment.

    Here are my current readings:
    http://www4.picturepush.com/photo/a/8862607/640/8862607.png

    My questions?
    Q1: Help interpreting noise in Santa Cruz would be appreciated ('specially by Jeff!).
    Q2: What can I do to improve capacity & quality?
    Q3: Why is there that much noise at channel 12 anyway?
     
    Vinny P., Jul 30, 2012
    #1
    1. Advertisements

  2. Vinny P.

    Vinny P. Guest

    Q1: Help interpreting noise in Santa Cruz would be appreciated ('specially by Jeff!).
    I belatedly realized the link I gave was lousy quality so here are the full-size images:
    http://www2.picturepush.com/photo/a/8862580/img/8862580.png
    http://www4.picturepush.com/photo/a/8862607/img/8862607.png

    How would you interpret these results with respect to noise in the Santa Cruz mountains above Silicon Valley?
     
    Vinny P., Jul 30, 2012
    #2
    1. Advertisements

  3. Vinny P.

    Char Jackson Guest

    A couple of things look concerning. Wireless Mode is shown as Station
    WDS. WDS, really? Is there no Client mode? The second thing is the use
    of channel 5, which is a strange channel to use since it's not one of
    the 3 completely non-interfering channels, (1, 6, 11).

    You asked about noise, but all I see is a noise floor of -98dBm, which
    looks excellent to me. You asked about noise on channel 12, which I
    assume was a typo.

    Regarding improving your reception, have you made sure the antenna is
    properly aimed, both horizontally and vertically?
     
    Char Jackson, Jul 31, 2012
    #3
  4. Vinny P.

    Vinny P. Guest

    Hmmm... the WISP told me to set it up that way.

    The only options are:
    a) Station
    b) Station WDS
    c) Access Point
    d) Access Point WDS

    Personally, I don't even know what WDS stands for.
    But googling, I find this thread:
    http://forum.ubnt.com/showthread.php?t=53297
    Which says:
    "WDS is always recommended if possible.
    Sometimes it's not possible with other vendors,
    since WDS is not astandard protocol. ...
    Enabling WDS provides a true "transparent" layer 2 bridge."
    There's nothing 'called' client mode in the UBNT AirOS gui. :(
    I understand. That's the choice of my WISP.

    Seems to me that channel 11 is better taking only the noise level here in
    the Santa Cruz mountains into account.
    Thanks. I didn't know WHAT to look for!

    Looking again at the spectrum analysis from the radio:
    http://www2.picturepush.com/photo/a/8862580/img/8862580.png
    I see you're looking at the current (or average) noise (green line),
    while I was looking at the peak noise (blue line).

    There is about 20 dBm difference - so that's huge!

    So, looking at the average noise (green), I see what you mean.

    Noise is about -100 dBm at channel 1, peaks at -85 dBm at channel 2,
    drops back to about -98 dBm for channel 3 through 8 (roughly), then it
    peaks again at channel 9 to about -79 dBm, and then drops down to about
    -100 dBm for the remaining channels 10, 11, and (yes), 12 before finally
    dropping to about -108 dBm after that.

    That's the kind of information I was looking for!
    I didn't know what was a good or bad number, nor what to look at.

    Well, um, er.... there is a channel 12 on the spectrum analysis; so I
    'was' talking about channel 12. Dunno if I can actually 'use' that
    channel, but, it's there on the graph. (I have the export version of the
    nanobridge.)
    I can visually see the access point, which is about 3 to 4 miles away, so
    I first aimed the antenna by eye - and then aimed it using the built-in
    signal strength aiming software which comes with Ubiquiti AirOS.

    So, I 'think' it's aimed right! :)
     
    Vinny P., Jul 31, 2012
    #4
  5. Vinny P.

    Char Jackson Guest

    Then you may have to do it that way.
    Here's some more info on WDS:
    <http://en.wikipedia.org/wiki/Wireless_distribution_system>

    The reason I don't like it is because it's an operational method that
    retransmits some of what it receives, sort of like a repeater, and I'm
    under the impression that doing so severely impacts throughput.
    I'm not too familiar with the Ubiquity UI, but I think Station is the
    equivalent of what I called Client mode. Of course, if they want you
    to use WDS, you're probably stuck doing so.
    Actually, I got the -98dBm from the UI snapshot that you provided, not
    from the spectrum analysis screen.
    To me, a noise floor of -98dBm looks very good, but it's only half of
    the equation. Equally important is your signal strength and how it
    compares to that noise level. (Disclaimer: I don't know what I'm
    talking about.)
    You won't be using channel 12 in the USA, so whatever noise is
    occurring out there is of little or no concern.
    That's how I would have done it, as well. Eyeball it first, acquire
    the connection, then while watching the real time signal strength
    meter slowly swing it side to side until you're sure you've maximized
    the signal strength. Note that the antenna may have one or more side
    lobes that could potentially work for you, as well, depending on
    whether you're staring at a huge source of interference when the
    antenna is aimed directly at the WISP. Finally, if the antenna is
    capable, adjust the antenna's vertical aim. Don't just eyeball that
    axis, since some antennas have built-in downtilt that may not be
    obvious. I don't know if that applies to the Nanobridge.

    I'm at or beyond the edge of my knowledge, so hopefully someone who
    actually knows will chime in.
     
    Char Jackson, Jul 31, 2012
    #5
  6. Y'er noise level is about -95dBm. If everything else were perfect,
    that would be a really good noise level. However, I think you have
    other problems. There are only 2 signals visible on Ch 2 and Ch 9.
    From your lofty location, you should be picking up all kinds of junk,
    even with the directional antenna. Point the Nanostation at SCZ, San
    Jose, or Loma Prieta and see if it can hear more signals. If not, fix
    your radio, antenna, coax, etc.

    Post a picture of your derangement. Most of the problems I find are
    not equipment problems. They're installation problems.

    Because there's someone transmitting on Ch 12. -70dBm is not very
    strong. -30dBm would be a strong signal. A better question to ask is
    why don't you hear more signals?
    Yeah, much better. Thanks.
    Dunno. You have an 18dBi antenna, which should have a beamwidth of
    about 10 degrees. You'll hear plenty of junk if the antenna is
    pointed in the right direction, but none if it's to the side. If your
    Nanostation is pointed at the Hilltop (or is it SurfnetC) access
    point, you don't have anywhere near enough usable signal. Something
    is wrong with your setup.
     
    Jeff Liebermann, Jul 31, 2012
    #6
  7. WDS will work, but has a bit of overhead.
    Same as client bridge mode.
    Same as transparent bridge mode. MAC addresses are preserved through
    the bridge. If you connect an ethernet switch (not a router) to your
    Nanostation ethernet port, and then connect more than one computah to
    the switch, the MAC address of each machine will appear on the
    ethernet port of the ISP access point. With just station (client
    bridge) mode, all the traffic looks like it's coming from the MAC
    address of the client bridge radio. A major advantage of this mode is
    the ability to talk between two wireless routers.
    Ordinary access point.
    Ordinary access point plus store and forward WDS repeater. Ordinary
    user machines (such as laptops, PDA's, and cell phones) can connect to
    the access point, which then repeats the traffic to the central access
    point. Other WDS access points can also pass traffic through the same
    path.
     
    Jeff Liebermann, Jul 31, 2012
    #7
  8. Vinny P.

    Vinny P. Guest

    Good idea. It's easiest to point to Loma Prieta or San Jose. I'm picking
    up an antenna mount from an old Dish TV antenna because the mount I have
    isn't sturdy enough for Santa Cruz winds - so when I set that up, I'll
    test the AirView for more signals like you suggest.
    Here's a picture of the mount on the wall of the house near the roofline.
    I tried to drill through the wall but ran into the hardest steel plate
    you can imagine. Dulled two drill bits before I realized it was
    impenetrable.
    http://www4.picturepush.com/photo/a/8864332/img/8864332.jpg

    You can see my first set of drill holes in this picture (to the right).
    What is it that is inside the wall, I'll never know.
    When I replace the antenna mast, I'll run that test for you!
    I think the spectrum analysis images of noise are really useful in
    general to anyone who has WiFi!

    You know the mountains well. Rude Dave at Surfnet & Loren at Hilltop are
    out here as well, but my signals are from Mike at Ridgewireless.net who
    installed an AP recently to serve the mountain community.

    These are the known ISP providers for the mountains that I know of (do
    you know of any others)?
    * Hughes Ku satellite
    * Viasat Exede Ka satellite
    * Surfnet WISP
    * Hilltop WISP
    * RidgeWireless WISP
    * Etheric WISP
    * Cooplabs (aka Superspeedy) WISP
     
    Vinny P., Jul 31, 2012
    #8
  9. Vinny P.

    Vinny P. Guest

    The antenna is on a cantilevered mounting arm screwed into the outside
    wall of the house under the roofline.
    http://www4.picturepush.com/photo/a/8864332/img/8864332.jpg

    Note: I tried four times to drill deeper holes but there is a solid steel
    plate in that wall (apparently) that nothing can possibly dent. What is
    it, I don't know! ???

    While the mounting arm can't be tilted - the antenna bracket has many
    degrees of tilt and rotation.

    The problem is the tilt is hard to judge by eye so I did it by airOS as
    shown in the screenshot below:
    http://www3.picturepush.com/photo/a/8864421/img/8864421.png

    BTW, I never understood transmit CCQ. Are these numbers OK?
     
    Vinny P., Jul 31, 2012
    #9
  10. Vinny P.

    Char Jackson Guest

    I would think that using the electronic alignment tool is better than
    eyeballing it. That L-shaped mounting bracket looks more flimsy than I
    was expecting, BTW. I think you said you were planning to replace it,
    which sounds good to me.
    Sorry, no idea.
     
    Char Jackson, Jul 31, 2012
    #10
  11. Vinny P.

    Vinny P. Guest

    The nanobridge, for about $70, comes with everything except cat5 cable &
    the mounting arm.

    It comes with the antenna, the mounting hardware, the power over Ethernet
    supply (POE), and the transceiver.

    I'm replacing the mounting arm with one from an old Dish TV antenna.

    My only tactical problem is there is 'something' hard as steel in the
    walls an inch under the stucco that I just can't drill through - so I'm
    going to have to re-locate the antenna if I can't figure out how to get
    through it.
     
    Vinny P., Jul 31, 2012
    #11
  12. Looking at the data sheet, it appears that you're looking at about a
    10 degree beamwidth:
    <http://www.ubnt.com/downloads/datasheets/nanobridgem/nbm_ds_web.pdf>
    Take a protractor and a piece of paper and draw 10 degree angle.
    That's your alignment accuracy. Trying to bore sight aim that antenna
    is not going to work. You'll have to rock it back and forth to find
    the maximum signal point. From the screen shots you've posted, I
    don't think your antenna is aimed at their AP.

    Looks like you have it mounted vertically polarized. Is that what
    RidgeWireless is using?
    Probably a nail stopper protecting the area where Romex crossed over a
    2x4.
    <http://www.strongtie.com/products/connectors/ns-nsp-pspnz.asp>
    Consider yourself lucky that you didn't continue as you would
    eventually have drilled through a power line. Next time you drill,
    use a stud finder to find the stud, but also use an AC voltage
    detector to make sure you're not drilling into a power line.

    Please remember that you have only one life to give for your
    connectivity.
    Ummm... you're on AirMax ISP firmware 5.3.5. Version 5.5 is out. Ask
    your ISP before upgrading:
    Client Connection Quality. Basically, it's the ratio of how an ideal
    radio would be expected to act, divided by what you're really seeing.
    The Ubiquity definition is kinda vague. This is from Mikrotik and
    hopefully should apply to Ubiquiti.

    Client Connection Quality (CCQ) is a value in percent that shows
    how effective the bandwidth is used regarding the theoretically
    maximum available bandwidth. CCQ is weighted average of values
    Tmin/Treal, that get calculated for every transmitted frame,
    where Tmin is time it would take to transmit given frame at
    highest rate with no retries and Treal is time it took to
    transmit frame in real life (taking into account necessary
    retries it took to transmit frame and transmit rate).
     
    Jeff Liebermann, Jul 31, 2012
    #12
  13. Similar, but not the same:
    <http://www1.picturepush.com/photo/a/7248774/img/7248774.gif>
    <http://www2.picturepush.com/photo/a/7253460/img/7253460.gif>
    The antenna gains are roughly the same (18dBi verus 19dBi). The
    radios are the same boards but different power levels (100mw
    Nanostation M2 versus 600mw overkill Bullet M2). You should get
    roughly the same performance and signal levels. The dish will
    probably be more difficult to aim than the panel antenna due to
    slightly narrower beamwidth.
     
    Jeff Liebermann, Jul 31, 2012
    #13
  14. That should work. Make sure it's screwed down tight and quite rigid.
    Stop! You're drilling through a nail stopper. If you keep going,
    you'll end up going through a power cable.
    <http://www.strongtie.com/products/connectors/ns-nsp-pspnz.asp>
     
    Jeff Liebermann, Jul 31, 2012
    #14
  15. Vinny P.

    willshak Guest

    Vinny P. wrote the following on 7/31/2012 3:31 AM (ET):
    You probably hit a metal framing hanger (or framing metal hanger) in
    that first area.
     
    willshak, Jul 31, 2012
    #15
  16. Vinny P.

    Vinny P. Guest

    Whatever it is, it's big because the two bracket holes are a good four
    inches apart, and I drilled four of them, about six or seven inches
    apart, so the impenetrable object is (at a minimum) four inches by six
    inches large.
     
    Vinny P., Aug 1, 2012
    #16
  17. Vinny P.

    Vinny P. Guest

    Funny you should ask that.

    I wasn't sure which way to mount it so I simply followed the instructions
    that came with the radio.

    I do know the WISP has both polarities because he's the one that told me
    I was only using half the bandwidth with my Ubiquiti Airgrid - so gaining
    the other polarity is the whole reason for me buying the nanobridge.
     
    Vinny P., Aug 1, 2012
    #17
  18. Vinny P.

    Vinny P. Guest

    :)
     
    Vinny P., Aug 1, 2012
    #18
  19. Vinny P.

    Char Jackson Guest

    Do you have a wall safe, by chance?

    Jeff suggested it might be a nail stopper, but that's not it. Those
    things are much too small to fit your description and they're
    installed on the inside edge of the wall framing, not the outside
    where you're drilling.

    It's not framing hangers, either. Those are also small and located at
    the ends of the framing members, so you wouldn't have hit 4 of them.
    Besides, hangers are almost never used on vertical walls.

    Random thought, are you in a designated earthquake zone? Some builders
    do all kinds of things under that umbrella that would be weird and
    unheard of elsewhere.
     
    Char Jackson, Aug 1, 2012
    #19
  20. Vinny P.

    Vinny P. Guest

    I'm very near the Loma Prieta fault line so maybe that's it.
     
    Vinny P., Aug 1, 2012
    #20
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.