[amsat-bb] Re: [aprssig] 12/9600 baud Mobile Sat Msg tests (D700)

[Bob Bruninga <bruninga@xxxxxxxxxxxxx>]

On Mon, 24 Jul 2000, Ken Koster wrote:

> How large were the packets being sent and what Txdelay were you using?
> Packet length can have a major impact on performance so it would be nice to
> know what you were using.  If the packets were all short like typical aprs
> packets I'd really like to see the results for 256 and 576 byte packets.

Very good question!  Yes, they were all Mic-E position/status packets,
thus very short.  Longer packets would require another dB or so.
I loaned out ALL of my text books to Chas yesterday, so this is from
memory, but the plot of BER vs S/N is quite steep, so each one dB of
additional signal gets  you a factor of 10 improvement in BER.

So I would think that an additional 1 dB should ccorrellate with the more
typical longer tererstrial WinAPRS packets...  Again, we were searching
for the minimum satllite link for short packets...

de WB4Apr, bOB
 

> On Mon, 24 Jul 2000, Bob Bruninga wrote:
> > Today we did the same DATA performance tests on two D700 data radios as we
> > reported last week for the D7.  The results were very similar with 1200
> > baud being about 7 dB better than 9600.  Thus, we conclude that for
> > reliable delivery of short text messaging to mobile and handheld satellite
> > users, that a 1200 baud 2m downlink is by far the most effective from a
> > power budget standpoint.  Going to 9600 baud requires 5 times more power
> > and only gains about a doubling of throughput due to TXD...
> > 
> > TEST CONDITIONS:  Two D700's were separated by about 1/4 mile with 30 
> > to 40 dB of attenuators on both ends connected to mobile whips.  Radios
> > were operated at the lowest power setting.  Attenuators were adjusted to
> > find the 50% and 100% performance rate at 1200 and 9600 baud on the
> > Smeter.  Then we used a HP Sig gen to calibrate these readings to dBm.
> > S meter readings shown are the number of LCD segments lit.
> >  
> > BAND  FREQ    DATA RATE   50%          100%      Comments
> > ----  ------- ---------  -------     -------     ----------------
> >  A    145 MHz 1200 baud  S0 -123dBm  S0 -120dBm
> >  B    145 MHz 1200 Baud  S0 -121dBm  S0 -118dBm  About 2 dB worse
> >  A    145 MHz 9600 baud  S2 -116dBm  S3 -113dBm  About 7 dB worse
> > 
> >  B    435 MHz 1200 baud  S0 -122dBm  S0 -119dBm
> >  B    435 MHz 9600 baud  S3 -115dBm  S3 -111dBm  About 7 dB worse
> > 
> > PACSATS:  The above tests are between two identical radios.  My PACSAT
> > experience has shown that the 9600 baud PACSAT downlinks often decode with
> > even less signal, often at S0!  So they do seem to have optimized signals.
> > 
> > de WB4APR, Bob
> > 
> > ------ Following is a FORWARDED copy of D7 test from LAST week --------
> > 
> > For evaluating the optimum configuration for Handheld Satellite Messaging,
> > today, we did another practical test we have long wanted to do.  W4HFZ and
> > I compared 1200 and 9600 baud performance using the SAME radios under the
> > same channel under controlled conditions.  The results 'in this test' do
> > show about a 7 dB difference between 1200 baud AFSK and 9600 baud FSK.
> > 
> > TEST CONDITIONS:  Two THD7 HTs were separated by 100 feet of RG58 with
> > 70 dB or more of SMA RF attenuators on both ends and with RF clamp on
> > ferrite chokes at 4 places along the coax.  Radios were operated at
> > Extremely Low power setting.
> > 
> > PROCEDURE:  Attenuators were adjusted to find the 50% and 100% performance
> > rate at 1200 baud, then we changed to 9600 baud and re-adjusted
> > attenuators to get the same performance.  S signals are the number of LCD
> > segments lit.
> > 
> > BAND  FREQ    DATA RATE   50%          100%      Comments
> > ----  ------- ---------  -------     -------     ----------------
> >  A    145 MHz 1200 baud  S1 -120dBm  S1 -117dBm
> >  A    145 MHz 9600 baud  S5 -113dBm  S5 -110dBm  About 7 dB worse
> > 
> >  B    435 MHz 1200 baud  S0 -119dBm  S0 -116dBm
> >  B    435 MHz 9600 baud  S3 -111dBm  S3 -109dBm  About 7 dB worse
> > 
> > S-METER:  Although the above test shows that there is a difference in
> > the S meter on the two bands, the RF signals are about the same.  Within
> > each S meter range, there appears to be about 3 dB and that is the
> > difference between solid and marginal decoding.
> > 
> > PACSATS:  The above tests are between two identical radios.  My PACSAT
> > experience has shown that the 9600 baud PACSAT downlinks often decode with
> > even less signal, often at S0!  So they do seem to have optimized signals.
> > 
> > CAVEATS:  Even though the radios are identical, it is still an apple and
> > oranges comparison, because, of course, the audio path and filtereing is
> > different  between the two modems.  But if we "asume" that kenwood did the
> > best they could with each, within their other constraints, then this does
> > give us a data point that tends to confirm the performance difference
> > between 1200 AFSK and 9600 FSK baud.
> > 
> > Given that there is a further 9 dB or so PATH loss difference between VHF
> > and UHF, this gives us a baseline for trying to design the optimum simple
> > 2-way digital messaging satellite links to handhelds.  On the downlink, we
> > want to deliver a message to an HT in your pocket.  On the uplink, we want
> > to be able to transnmit to the bird with the whip held in the clear.
> > 
> > Just thought you might be interested in the results of this test.
> > 
> > de WB4APR, Bob
> > 
> > 
> > ---
> > You are currently subscribed to aprssig as: KOSTER@IPSG.MOT.COM
> > To unsubscribe send a blank email to leave-aprssig-2122X@lists.tapr.org
> > Questions regarding the SIG go to the SIG administrator: wa1lou@tapr.org
> -- 
> --
> Motorola			Email:  koster@seattle.css.mot.com
> Seattle Cellular Design Center 	Office: (425) 487-8854
> 24000 35th Ave SE, Suite 100   	Cell  : (206) 499-4872
> Bothell, WA  98021             	Fax   : (425) 483-3400
> 
> 

APRSdos REPLY/COMMENT:

Reply mail addr:   wb4apr@amsat.org   
US mail address:   115 old Farm Ct, Glen Burnie, MD 21060
See DAYTON97 HISTORY:    http://web.usna.navy.mil/~bruninga/dayton.html
See Maryland APRS LIVE:  http://web.usna.navy.mil/~bruninga/aprs.html
See GPS on ANY radio:    http://www.tapr.org/tapr/html/mic-e.html

----
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org