Discussion:
[Freetel-codec2] Maximizing range per Watt?
Tomas Härdin
2016-02-11 11:14:58 UTC
Permalink
Hi list

I've been looking at designing modems for what you might call QRPP for
the last few days, and I've been following Mike's designing his open VHF
FSK modem. What I'm thinking about may follow along lines similar to the
goals of the FreeDV project, and feels more relevant than whatever drama
I saw someone try to stir up on here earlier today

My question is simple: how low can you go in terms of Joules per user
bit, if you're allowed arbitrarily high capacity-to-bandwidth ratios
(say 1:10 or 1:100)?

Terms more relevant to this list might be: what if we ran the 700 bit/s
system in something like a 12.5 kHz channel? Could that improve the
range we get out of a 5W PA? Our old friend the channel capacity theorem
supports this. Letting S/N = (d0/d)^2*SNR_d0 and solving for d/d0 for a
700 bit/s channel gives:

d/d0 = [(2^(700/B)-1)/SNR_d0]^(-1/2)

In the 12.5 kHz example above, assuming we have some range d0 for an SNR
of 0 dB in a 700 Hz band, this suggests we should be able to get five
times the range out of the same amount of energy(!)

Anywho, I'm also in the process of designing an
as-simple-and-compact-as-possible BPSK (really DSB-SC)
modulator+transmitter since BPSK seems to have the best chance in terms
of Eb/No. Part of that will include writing simulations in Octave to try
and work some of this out, so quite a bit of perspiration involved. The
plan is to add on FEC and crank up the bitrate. I'm trying to ensure at
this early stage that I'm not making any fundamental mistakes however,
and good ideas are welcome

73 de sa2tms
/Tomas
David Rowe
2016-02-11 12:17:46 UTC
Permalink
Hello Tomas,

Some calculations on the lower SNR limits for digital voice here:

http://www.rowetel.com/blog/?p=4053

- David
Post by Tomas Härdin
Hi list
I've been looking at designing modems for what you might call QRPP for
the last few days, and I've been following Mike's designing his open VHF
FSK modem. What I'm thinking about may follow along lines similar to the
goals of the FreeDV project, and feels more relevant than whatever drama
I saw someone try to stir up on here earlier today
My question is simple: how low can you go in terms of Joules per user
bit, if you're allowed arbitrarily high capacity-to-bandwidth ratios
(say 1:10 or 1:100)?
Terms more relevant to this list might be: what if we ran the 700 bit/s
system in something like a 12.5 kHz channel? Could that improve the
range we get out of a 5W PA? Our old friend the channel capacity theorem
supports this. Letting S/N = (d0/d)^2*SNR_d0 and solving for d/d0 for a
d/d0 = [(2^(700/B)-1)/SNR_d0]^(-1/2)
In the 12.5 kHz example above, assuming we have some range d0 for an SNR
of 0 dB in a 700 Hz band, this suggests we should be able to get five
times the range out of the same amount of energy(!)
Anywho, I'm also in the process of designing an
as-simple-and-compact-as-possible BPSK (really DSB-SC)
modulator+transmitter since BPSK seems to have the best chance in terms
of Eb/No. Part of that will include writing simulations in Octave to try
and work some of this out, so quite a bit of perspiration involved. The
plan is to add on FEC and crank up the bitrate. I'm trying to ensure at
this early stage that I'm not making any fundamental mistakes however,
and good ideas are welcome
73 de sa2tms
/Tomas
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
_______________________________________________
Freetel-codec2 mailing list
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
Tomas Härdin
2016-02-11 16:10:25 UTC
Permalink
Hi David (got the name wrong the first time, sorry)

Interesting, I somehow overlooked that post. It seems there's really no
limit to how far you can take ideas like these. This is why UWB is a
thing I guess

My goal for the week is to try and analyze where the limit is with
practical (RS-like) codes. Some preliminary computations showed that
Joules/bit can go arbitrarily low, which sounds too good for me to trust
myself not having made a mistake. CCSDS also have good documents in this
area that I should read more closely

/Tomas
Post by David Rowe
Hello Tomas,
http://www.rowetel.com/blog/?p=4053
- David
Post by Tomas Härdin
Hi list
I've been looking at designing modems for what you might call QRPP for
the last few days, and I've been following Mike's designing his open VHF
FSK modem. What I'm thinking about may follow along lines similar to the
goals of the FreeDV project, and feels more relevant than whatever drama
I saw someone try to stir up on here earlier today
My question is simple: how low can you go in terms of Joules per user
bit, if you're allowed arbitrarily high capacity-to-bandwidth ratios
(say 1:10 or 1:100)?
Terms more relevant to this list might be: what if we ran the 700 bit/s
system in something like a 12.5 kHz channel? Could that improve the
range we get out of a 5W PA? Our old friend the channel capacity theorem
supports this. Letting S/N = (d0/d)^2*SNR_d0 and solving for d/d0 for a
d/d0 = [(2^(700/B)-1)/SNR_d0]^(-1/2)
In the 12.5 kHz example above, assuming we have some range d0 for an SNR
of 0 dB in a 700 Hz band, this suggests we should be able to get five
times the range out of the same amount of energy(!)
Anywho, I'm also in the process of designing an
as-simple-and-compact-as-possible BPSK (really DSB-SC)
modulator+transmitter since BPSK seems to have the best chance in terms
of Eb/No. Part of that will include writing simulations in Octave to try
and work some of this out, so quite a bit of perspiration involved. The
plan is to add on FEC and crank up the bitrate. I'm trying to ensure at
this early stage that I'm not making any fundamental mistakes however,
and good ideas are welcome
73 de sa2tms
/Tomas
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
_______________________________________________
Freetel-codec2 mailing list
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
_______________________________________________
Freetel-codec2 mailing list
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
Loading...