Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Migrated to Confluence 4.0

METCRAXII Communications Test Site Survey June 2013

Overview:

A site survey was conducted from 9-11, June-2013 in order to determine whether wireless operations could be conducted successfully within the crater for the field project in October.    Intermittent Wifi connectivity during the first METCRAX field project in 2006 limited ISFS’s ability to reliably recover data between the crater floor and the base operations trailer located ~1km outside, raising concerns for the second campaign.

For METCRAXII wireless operations will consist of both Wifi and Bluetooth telemetry.   Wifi will provide the primary link between a single “Flux-PAM” DSM in the crater bottom communicating through an access-point/repeater which will also bridge a “Rim-C” profiling DSM on the south ridge to the base operations trailer outside.   Alternatively a fiber-optic cable could link Flux-PAM with Rim-C, and Wifi bridging to the base trailer only.   Within the crator two Bluetooth access points on the Flux-PAM DSM will serve 6-1hz barometers and 2-20hz sonic anemometers located partially up the side-walls.

The test was conducted within the crater to determine the viability of the wifi link between the bottom and south rim router as well as 3 simulated bluetooth data streams from representative locations where the sonic/barometer sensors will be located.

Summary of Results:

The Wifi/Bluetooth test worked well enough so that reliable, simultaneous operations can be expected.  In part this is due to a significantly reduced data requirement for METCRAXII vs the original experiment.

The fiber-optic cable option between the Rim-C to the floor Flux-PAM is an acceptable alternative to Wifi.   Brad Andes, manager of the Meteor Crater Operations, said that we could spool a line from the south-rim to the bottom and hiking the route is ok with them.   The route is steep near the rim and the cable might need to be slung with a rope to provide support.   The distance estimate is >=2500’ without measuring it.   This is based on 550’ vertical depth plus ½ of 4000’ rim diameter per Brad.   We should plan on 3000-ft for adequate spare cabling and should avoid segmented joints especially on the steep upper portion of the crater.

Traversing the slope between barometer sites along the lower side-wall is difficult.   This is because the wash areas are covered by tumble weeds and footing between rocks may be hidden.

Test Setup:

Wifi:  An Alico AP24 was setup as an access point and was located on the south rim of the crater (its configuration is listed at the end of this blog entry).  Placement on the rim appeared optimal, on a 5’ tripod at the edge of the slope with excellent view of the crater floor.  The antenna was powered through its a/c p.o.e. injector run off an inverter solar charged battery.  No noticeable external signals in the area based on WiSpy meter.

On the crater floor an EtherantIIILR was placed at the Flux-PAM site located per Dave Whiteman; roughly 100m nnw from the fenced area.   It was powered from a ‘12.7vdc’ mote battery box.   With clear skies the solar charged battery voltage floated well above 13volts.   Although not the ideal 15v, it was more than adequate.  The Eant was placed on the ground rather than on a mast and visually aligned with the rim AP24.   The tilt of the eant was roughly set.

Gordon configured the Acer laptop pc with Linux and his DSM software.   For the wifi tests, the Acer was connected via ethernet to the Eant.   Static addressing was used for the pc and radios.   Simple pings and ftp/get file transfers were run observing the throughput.   The Eant web page served as a measure of the signal quality and strength.   The ‘test’ file for ftp was ~7.6MBytes.  For reference these were results in the office (there also is variation in these):

Bench: PC to AP24 via ethernet

ftp/get 7.6MB ~= 2sec (3700Mbps)



ping 64B ~= .75ms


Bench: PC to Eant (solid/close connection – best performance)

ftp/get 7.6MB ~=12.7sec (610kbps)

Eant SNR=50, ‘small distance’


ftp/get 7.6MB ~=15sec (566kbps)

Good solid link expectation


ping 64B ~=2.4ms


Bluetooth:  For the bluetooth tests the Acer was connected via usb to a WT41 configured as an access point.   The dsm ‘data-stats’ and ‘rserial’ routines were used to communicate with three motes placed at sampling sites along the side-wall selected by Dave Whiteman.   Performance was monitored using linux hcitool.   Data throughput was compared with ideal data rate settings, and were varied between 1-30hz.  Motes were programmed to send 16byte ascii test messages to mimic sonic operations as well as normal mote power montoring values.   Message rates were altered via operator commands issued from the laptop.  Two of the motes were connected to a WT41 btradio with an omni antenna and set to maximum power (“set bt power 19 19 19”).   These WT41’s were demonstrated to easily reach 800m during the SCP project so cross-crater signals were expected.   The third mote was configured with a BlueRadios BT-XB-C40 radio.   These radios are class 1, with ranges up to 100m/330ft, which is well below the distance between the Flux-Pam and crater side-wall locations needed for the barometers.   However the desire was to determine whether a high-gain directional yaggi antenna (9db) would reliably overcome their range restriction (9db = 8x power) while narrowing the beam to avoid mutual interference.   This mote’s radio was initially programmed with reduced power at 0db, versus its maximum power of +10dbm.

Conditions: the weather was very hot and dry as a heat wave was being experienced in the region.   Humidities were around 10% and temperatures were around 100degF (a bit higher in the crater) with full sun on Monday and some scattered clouds on the 2nd day.   Winds varied from calm to quite windy and variable especially on the rim and within the crater.   These conditions probably stressed the equipment more than what can be expected during either September setup or October operations.   RF noise does increase with temperature, reducing performane so this represented good timing for the survey.

Initial Crater Wifi Tests

In the crater, an rf-survey with the WiSpy meter showed no interfering signals.   Initial power up of Eant-J had near immediate connectivity to AP24 per the leds indicating good connectivity.   Initial power setting of the AP24 was 13dbm (range setting is up to 28) and Eant was on ‘small-distance’ between access points.   The ftp/get tests varied between 16.9-24.4 seconds for the initial try.

Wifi: PC-EantJ to AP24

ftp/get 7.6MB multiple tests =16.9-24.4sec, data-rates 320--455 kbps

AP24 tx-power=13, Eant SNR~41-44db, ‘small distance’


ping 64B ~= 2.4ms


Boost AP24 tx-power

ftp/get 7.6MB tests = 14.6-14.9sec, data-rates 520kbps

AP24 tx-power=20, Eant SNR~47-48db, ‘small distance’


ping -s 2000B ~= 10ms


Results were very encouraging noting some variability in the link.   Solid connectivity during tests even with reduced power settings and trying different settings in AP/Eant including Eant ‘micro-distance’ to ‘large-distance’ (only 1 ap), and AP24 settings from 13 to 24dbm.   The higher AP24 power settings were best and having the ‘small-distance’ appeared most optimal on the Eant.  I made the mistake of issuing a ‘scan’ command over the wireless link to the AP24, preventing command termination without going to the rim, thus ending the tests.   Test duration ~1.5hours on the air.

Second Crater Wifi with BlueTooth Test

Initial testing involved the bluetooth clients connected to their access point located at the Flux-Pam site.   The first mote, with WT41 was placed and powered up along the trail at the height of where the barometer string will be.   This is about >=300m away from where the access point was powered up at the Flux-Pam site.  Communications between the Acer and this mote were established quickly and commands were easily received by the remote.   It was set to 30hz message rate.   The other 2 motes were powered up at the Flux-Pam site, connecting quickly without disruption of the first unit.   Tim Lim then walked across the floor with them to the first sonic location up the side wall marked by Dave.   Communications with the Acer remained good and all messages were received and commandability solid, including with the reduced power btradio mote which had a significantly lower rssi.   The btradio mote was then traversed along the sidewall between the other two motes, roughly 200m and 400m between them.   The rssi signals were: -7, -4 and -18 to the btradio mote.   The motes were located roughly 90-degrees apart between the end units.   The access point was power cycled and connectivity with the 3 motes was quickly restored.   The DSM data stats indicated good message recovery at ~30 messages/sec from all 3.

After the bluetooth links were working well, the Wifi was turned on between Eant-J and the Rim-C site.   Once again, initial power up of Eant-J had near immediate connectivity to AP24 on the Rim.   The AP24  power setting was initially 20dbm, and Eant was on ‘small-distance’ between access points.   The ftp/get tests varied and the Eant SNR reported values were significantly different than the first day, at ~35-37db.  The bluetooth link was turned off to determine whether it was causing the reduced link quality.  The Eant was also power cycled a couple of times.

Wifi: PC-EantJ to AP24

ftp/get 7.6MB multiple tests =25.7-34sec, data-rates 270--300kbps

AP24 tx-power=20, Eant SNR~35-37db, ‘small distance’, motes running


ping --s 2000B ~= 20ms


Boost AP24 tx-power

ftp/get 7.6MB tests = 20-25sec, data-rates 308-376kbps

AP24 tx-power=28, Eant SNR~38-39db, ‘small distance’, motes running

Turn Off Motes

ftp/get 7.6MB tests = 15-16sec, data-rates 484-507kbps

AP24 tx-power=28, Eant SNR~40-42db, ‘small distance’; note power cycled and examined Eant prior also

Turn Motes Back On

ftp/get 7.6MB tests = 14.4-15.9sec, data-rates 481-530kbps

AP24 tx-power=28, Eant SNR~40-42db, ‘small distance’.
Note: quick linkup of BTmotes with rssi’s=-7,-8,-21
DataStats rates 30,29,29: good over 2min

Results were again very encouraging even noting some variability in the wifi links.  One concern had been whether or not reflections/fading caused by the crater structure itself would degrade the wireless links.   In addition were concerns whether mutual interference between bluetooth and wifi nodes would be problematic.   A spectrum analyzer would have been an ideal tool to use for looking at the signal/noise environment at each location, but one was not available for use. 

The bluetooth links remained solid throughout all tests and impressively during transport of the radios across the crater floor.   This was also true even having a very weak power setting (0db vs 10db max) on the btradio mote while using the yaggi antenna.  With full power setting the BTradios with yaggi antennas can provide a reliable link.   The WT41 motes also provide good links as expected, although physical implementation with the motes required a second ‘white-box’ in addition to the mote box.   However the pointing a

The wifi link at first appeared to be effected by the bluetooth radios, however the degradation was ultimately attributed to poor antenna alignment of the Eant with the Rim-C AP24.    Several tests were again performed with different AP and Eant settings with good results.   It appears that the wifi will have some variability but it definitely depends upon having a good antenna alignment and proper power settings.  Note that the bluetooth and wifi antennas at the Flux-PAM site were less than ideal, whereas the AP24 location was very good.

During the field experiment the AP24 bridge/repeater will need to be raised roughly 20m above the tower base in order to get good visibility over the crater rim and above the low ridge between it and the base trailer location.





AP24 Configuration:

# jan/01/2000 00:03:42 by RouterOS 2.9.27
# software id = IH8C-3TT
#
/ interface wireless
set wlan1-Int-Ant name="wlan1-Int-Ant" mtu=1500 mac-address=00:15:6D:20:01:90 \
    arp=enabled disable-running-check=no radio-name="AP24-IntAnt" \
    mode=ap-bridge ssid="ISFS" area="" frequency-mode=regulatory-domain \
    country="united states" antenna-gain=0 frequency=2442 band=2.4ghz-b/g \
    scan-list=default rate-set=default \
    supported-rates-b=1Mbps,2Mbps,5.5Mbps,11Mbps \
    supported-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps,24Mbps,36Mbps,48Mbps,54Mbps \
    basic-rates-b=1Mbps basic-rates-a/g=6Mbps max-station-count=50 \
    ack-timeout=dynamic tx-power=18 tx-power-mode=card-rates \
    noise-floor-threshold=default periodic-calibration=default \
    periodic-calibration-interval=60 burst-time=disabled dfs-mode=radar-detect \
    antenna-mode=ant-b wds-mode=disabled wds-default-bridge=none \
    wds-default-cost=100 wds-cost-range=50-150 wds-ignore-ssid=no \
    update-stats-interval=disabled default-authentication=no \
    default-forwarding=yes default-ap-tx-limit=0 default-client-tx-limit=0 \
    proprietary-extensions=pre-2.9.25 hide-ssid=no security-profile=isffwep \
    disconnect-timeout=3s on-fail-retry-time=100ms preamble-mode=both \
    compression=no allow-sharedkey=no comment="" disabled=no
/ interface wireless nstreme
set wlan1-Int-Ant enable-nstreme=no enable-polling=yes framer-policy=none \
    framer-limit=3200
/ interface wireless manual-tx-power-table
set wlan1-Int-Ant manual-tx-powers=1Mbps:25,2Mbps:25,5.5Mbps:25,11Mbps:25,6Mbps\
    :25,9Mbps:25,12Mbps:25,18Mbps:25,24Mbps:25,36Mbps:24,48Mbps:22,54Mbps:21
/ interface wireless security-profiles
set default name="default" mode=none authentication-types="" \
    unicast-ciphers="" group-ciphers="" wpa-pre-shared-key="" \
    wpa2-pre-shared-key="" eap-methods=passthrough tls-mode=no-certificates \
    tls-certificate=none static-algo-0=none static-key-0="" static-algo-1=none \
    static-key-1="" static-algo-2=none static-key-2="" static-algo-3=none \
    static-key-3="" static-transmit-key=key-0 static-sta-private-algo=none \
    static-sta-private-key="" radius-mac-authentication=no group-key-update=5m
add name="isffwep" mode=static-keys-required authentication-types="" \
    unicast-ciphers="" group-ciphers="" wpa-pre-shared-key="" \
    wpa2-pre-shared-key="" eap-methods=passthrough tls-mode=no-certificates \
    tls-certificate=none static-algo-0=104bit-wep \
    static-key-0="656f6c33303334393738383032" static-algo-1=none \
    static-key-1="" static-algo-2=none static-key-2="" static-algo-3=none \
    static-key-3="" static-transmit-key=key-0 static-sta-private-algo=none \
    static-sta-private-key="" radius-mac-authentication=no group-key-update=5m
/ interface wireless align
set frame-size=300 active-mode=yes receive-all=no \
    audio-monitor=00:00:00:00:00:00 filter-mac=00:00:00:00:00:00 ssid-all=no \
    frames-per-second=25 audio-min=-100 audio-max=-20
/ interface wireless connect-list
add interface=wlan1-Int-Ant connect=yes mac-address=00:00:00:00:00:00 \
    ssid="ISFS" min-signal-strength=-95 area-prefix="" \
    security-profile=isffwep comment="" disabled=no
/ interface wireless access-list
add mac-address=00:20:F6:05:1E:C7 interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:25:8B interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:26:D9 interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:16:CD interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:1E:D5 interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:24:5A interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:24:7F interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:24:85 interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:24:56 interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:20:F6:05:24:4F interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:15:6D:20:01:9A interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:15:6D:05:24:7F interface=wlan1-Int-Ant authentication=yes \
    forwarding=yes ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:06:11:80:31:42 interface=wlan1-Int-Ant authentication=yes \
    forwarding=no ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
add mac-address=00:15:6D:10:29:B2 interface=wlan1-Int-Ant authentication=yes \
    forwarding=yes ap-tx-limit=0 client-tx-limit=0 private-algo=none \
    private-key="" comment="" disabled=no
add mac-address=00:0C:42:04:7B:F9 interface=wlan1-Int-Ant authentication=yes \
    forwarding=yes ap-tx-limit=0 client-tx-limit=0 private-algo=104bit-wep \
    private-key="656f6c33303334393738383032" comment="" disabled=no
/ interface wireless snooper
set multiple-channels=yes channel-time=200ms receive-errors=no
/ interface wireless sniffer
set multiple-channels=no channel-time=200ms only-headers=no receive-errors=no \
    memory-limit=10 file-name="" file-limit=10 streaming-enabled=no \
    streaming-server=0.0.0.0 streaming-max-rate=0