[aprssig] Fw: [nwaprssig] UHF 9K6 Baud vs. VHF 1200 Baud

[R.M. King]

----- Original Message ----- 
From: "R.M. King" <krm1012 at qwest.net>
To: <nwaprssig at nwaprs.info>; <aprssig-request at lists.tapr.org>
Cc: <bruninga at usna.edu>
Sent: Thursday, October 12, 2006 10:45
Subject: [nwaprssig] UHF 9K6 Baud vs. VHF 1200 Baud


> For sometime now I have been puzzled by the response of several persons on 
> both the NWAPRS and National APRS sigs to UHF 9k6 Baud operations.
>
> I have heard:
>
> a)    There is a 9db penalty for using 440Mhz  vs. 144MHz. Implying this 
> is an unacceptable or significant loss in performance for APRS operations.
>
> b)    There is a 2db or so penalty for using G3RUH FSK vs.AFSK.  Implying 
> the wider bandwidth required by 9k6 Baud vs. 1200 Baud causes an increase 
> in noise thus causing a higher BER rate adding to performance degradation 
> at 9k6 Baud.
>
> c)    QSB and phase distortion are worse at the higher frequency and other 
> "complaints".
>
> d)    It is more difficult to get 9k6 Baud operations to work properly.
>
>
> Of course, all of this is keeping some people from trying UHF 9k6 APRS 
> operations.
>
> None of this seemed to agree with my three years of practical on-the-air 
> operation of a small 440.875MHz 9k6 Baud APRS network in the Puget Sound 
> area of Washington State.  My observations on both 144.390 and 440.875MHz 
> seem to point to 440.875MHz 9k6 Baud as a better choice for APRS 
> operations.
>
> Let me address each of the complaints I have heard.
>
> The 9db path loss penalty -   It is true there is a 9.7db path loss when 
> using 440.875MHz compared to 144.390MHz.  However, in every case for 
> practical operations this is NOT a significant factor.  It does NOT reduce 
> performance for distances used on Earth for APRS mobile or fixed 
> operations. The real limiter is the same for 144.390MHz as it is for 
> 440.875MHz. And that limiter is LOS distance and curvature of the Earth. 
> (slant range) The signal strength at the ANTENNA TERMINALS for a typical 
> 440 mobile reciever is well within the lower limit for an acceptable BER 
> for APRS.  A practical working range may be 20 miles.  This is true for 
> 144 and 440MHz.
>
> Using two identically equipped base and mobile stations on 144 and 440MHz 
> (25Watts at the base antenna, a 6db gain gain antenna at the base station, 
> a 3db gain antenna at the mobile station, and a 20 mile path), the power 
> at the mobile station antenna terminals is -55.8dbm for the 144MHz station 
> and 9.7db less or -65.5dbm at the 440MHz station.  BOTH of these signal 
> strengths are sufficient to drive the typical FM receiver into limiting 
> enough to produce an acceptable BER for APRS.  It looks like this is true 
> even at 100 miles.  This is far beyond the slant range for most stations 
> on Earth.
>
> The reverse path loss (mobile to base and  mobile station using 25W) in 
> this case is 3db more.  This is still not enough to cause problems at 
> 440MHz.
>
> The 2db penalty for using G3RUH FSK vs. AFSK -  This breaks down to NOT a 
> conversion penalty for G3RUH FSK but a concern about BANDWIDTH in the I.F. 
> needed for the 9k6 Baud modulation vs. 1200 Baud modulation.  It turns out 
> if you use the same receiver, the bandwidth for both modulations is the 
> same.  About 15KHz.  The noise in the I.F. section is the same whether 
> 1200 Baud or 9k6 Baud is used.
>
> QSB, phase distortion, etc. -  These are very hard to quantify and would 
> take a lot of work to compare between 2 meter and 3/4 meter operation. 
> But let me say this;  The packet length at 9600 Baud is less than 1/3 the 
> length in time at 1200 Baud when a 200ms. TXD overhead is added to both 
> packets. This is significant for several reasons.  It means there is 60% 
> less time for all the man made things to kill an APRS packet.  Although I 
> can't put this into quantifiable units, practical experience indicates 
> this is very significant in improving APRS performance.  The shorter 
> packets do improve throughput.  There is a 3X improvement to be sure.
>
> It is more difficult to get 9k6 Baud operations working.  If this means 
> you must be sure your transmitter deviation is set properly with a 
> deviation meter, then it's true.  It is near impossible to set 9k6 Baud to 
> 3-kHz without a deviation meter.  It's only blind luck if one gets it 
> right without a deviation meter.  So PLEASE, don't even try without a 
> meter.
> I have measured the deviation of several D7's and D700's and have not 
> found a bad one.
>
> At 1200 Baud you can get away with murder.  Even an obscenely overdeviated 
> transmitter may get decoded.  But as Bob Bruninga indicated in an earlier 
> post, just taking a 1200 Baud TNC out of the box and hooking it to a 
> transmitter is VERY bad practice.  And it quite likely will be illegal. 
> You should use a good deviation meter even at 1200 Baud.
>
> So then.  Why try UHF and 9k6 Baud?  The big reason is throughput.  Much 
> more activity can be tolerated on a simplex channel at 9k6 Baud than 1200 
> Baud.  More net reliability is built in, NOT less as many people seem to 
> think. It improves the state of the art.  Better equipment is used, 
> generally, at 9600 Baud.
>
> When 144.390 finally chokes itself to death, UHF and 9k6 will still be 
> working.  Who knows, maybe the equipment manufacturers will pay attention 
> and develop gear for higher speed data operations.  One thing is for sure. 
> We can't grow if we stand still.  We can't learn new things if we don't 
> try.
>
> Anyone want to try 102.4K Baud APRS?
>
> Bob King
> K7OFT
>
> krm1012 at qwest.net
>
>
> _______________________________________________
> nwaprssig mailing list
> nwaprssig at nwaprs.info
> http://www.nwaprs.info/mailman/listinfo/nwaprssig
>