Return-Path: X-Spam-DCC: paranoid 1170; Body=2 Fuz1=2 Fuz2=2 X-Spam-Checker-Version: SpamAssassin 3.1.3 (2006-06-01) on lipkowski.org X-Spam-Level: X-Spam-Status: No, score=-2.4 required=5.0 tests=BAYES_00,DNS_FROM_AHBL_RHSBL, HTML_MESSAGE autolearn=no version=3.1.3 Received: from post.thorcom.com (post.thorcom.com [195.171.43.25]) by paranoid.lipkowski.org (8.13.7/8.13.7) with ESMTP id t08MeFOZ019842 for ; Thu, 8 Jan 2015 23:40:16 +0100 Received: from majordom by post.thorcom.com with local (Exim 4.14) id 1Y9LdU-00007I-8K for rs_out_1@blacksheep.org; Thu, 08 Jan 2015 22:33:00 +0000 Received: from [195.171.43.32] (helo=relay1.thorcom.net) by post.thorcom.com with esmtp (Exim 4.14) id 1Y9LdT-000079-QQ for rsgb_lf_group@blacksheep.org; Thu, 08 Jan 2015 22:32:59 +0000 Received: from omr-m4.mx.aol.com ([64.12.226.25]) by relay1.thorcom.net with esmtps (TLSv1:DHE-RSA-AES256-SHA:256) (Exim 4.84) (envelope-from ) id 1Y9LdQ-0000r2-M3 for rsgb_lf_group@blacksheep.org; Thu, 08 Jan 2015 22:32:58 +0000 Received: from mtaout-mad02.mx.aol.com (mtaout-mad02.mx.aol.com [172.26.221.206]) by omr-m4.mx.aol.com (Outbound Mail Relay) with ESMTP id 3649638000154 for ; Thu, 8 Jan 2015 17:32:54 -0500 (EST) Received: from White (ipb21bee4a.dynamic.kabel-deutschland.de [178.27.238.74]) by mtaout-mad02.mx.aol.com (MUA/Third Party Client Interface) with ESMTPA id 700D838000089 for ; Thu, 8 Jan 2015 17:32:53 -0500 (EST) Message-ID: <01D6434A3D2244758CA9F260E950E41E@White> From: "Markus Vester" To: References: <20150108220755.GA20377@cs.utwente.nl> Date: Thu, 8 Jan 2015 23:32:52 +0100 MIME-Version: 1.0 X-Priority: 3 X-MSMail-Priority: Normal Importance: Normal X-Mailer: Microsoft Windows Live Mail 12.0.1606 X-MimeOLE: Produced By Microsoft MimeOLE V12.0.1606 x-aol-global-disposition: G DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mx.aol.com; s=20140625; t=1420756374; bh=t6FH+C/yTU0iYQsxup1aGMOZaPwap4aOOyPuYFl7FL8=; h=From:To:Subject:Message-ID:Date:MIME-Version:Content-Type; b=uOxOoC+rgCr1DrmQzMxjPjOTiPE6BH+AXOYkaCTU5Ep+DbOY+YQJoKQIU8xCFZlqX Nvlpf1GesG1md8De80v79l5dCLiuFQ0eKceiwLpmBbt92fPip6TgkUvmuyq2tYYMba alwqMF1aMHd0cETE+NhKWFTP2dTPJwKJc6o6zJ/8= x-aol-sid: 3039ac1addce54af05954ee6 X-AOL-IP: 178.27.238.74 X-Scan-Signature: b6696563999afa34e36d15007cae73e3 Subject: Re: LF: Eb/N0 values for amateur modes Content-Type: multipart/alternative; boundary="----=_NextPart_000_0020_01D02B9B.6BE0BC40" X-SA-Exim-Scanned: Yes Sender: owner-rsgb_lf_group@blacksheep.org Precedence: bulk Reply-To: rsgb_lf_group@blacksheep.org X-Listname: rsgb_lf_group X-SA-Exim-Rcpt-To: rs_out_1@blacksheep.org X-SA-Exim-Scanned: No; SAEximRunCond expanded to false X-Scanned-By: MIMEDefang 2.56 on 10.1.3.10 Status: O X-Status: X-Keywords: X-UID: 1890 Dies ist eine mehrteilige Nachricht im MIME-Format. ------=_NextPart_000_0020_01D02B9B.6BE0BC40 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Thanks for this Pieter!=20 There are a couple more modes that have been used on LF, and of which = I'd be interested to see quantitative SNR threshold data: - Wolf PSK-10 with variable long integration from Stewart, KK7KA = http://www.scgroup.com/ham/wolf.html, - Jason MFSK-17 from Alberto, I2PHD http://www.weaksignals.com/, - WSQ realtime MFSK from Con ZL2AFP and Murray ZL1BPU = http://www.qsl.net/zl1bpu/SOFT/WSQ.htm. Maybe someone in the group can help? Markus (DF6NM) From: Pieter-Tjerk de Boer=20 Sent: Thursday, January 08, 2015 11:07 PM To: rsgb_lf_group@blacksheep.org=20 Subject: LF: Eb/N0 values for amateur modes A couple of days ago, Markus posted Eb/N0 numbers for WSPR and OPERA, for comparison to the BPSK VLF tests (see below). Some may be interested in a table with Eb/N0 values for some more = amateur modes, which I composed some weeks ago for an article in the Dutch = amateur radio magazine 'Electron': Needed SNR Net datarate Needed Mode in 2500 Hz in bits/s Eb/N0 = Comments =20 SSB voice +10 dB 20 +31 dB = very rough estimate CW (ZRO-test, by ear) -18 dB 0.54 +16 dB = based on avg.pwr; peak 3 dB higher CW (QRSS-3, waterfall) -26 dB 0.13 +14 dB = same CW (RSCW software, 12 wpm) -12 dB 4 +13 dB = same OPERA-2 -23 dB 0.23 +14 dB = peak pwr. 3 dB higher; 2 dB lower if counting CRC-bits as information RTTY -5 dB 32 +14 dB PSK31 -10 dB 31 +9 dB WSPR -29 dB 0.45 +5 dB = not counting energy in sync bits; otherwise 3 dB higher WSPR-15 -38 dB 0.056 +5 dB = same JT65 (for EME) -24 dB 1.54 +5 dB = same Coherent BPSK on VLF -57 dB 0.0058 -1 dB Theoretical limit -1.59 dB (The table is formatted for display using a fixed-width font.) I have to emphasize that the SNR and Eb/N0 values should be taken with a grain of salt, especially for the modes that do not use heavy FEC, since those don't have a no sharp threshold. The SNR values quoted come from a variety of sources on the web, and the BPSK/VLF line is based on Markus and Paul's experiments from last May (obviously, since I created the table before the recent experiments by = Dex and Paul). 73, Pieter-Tjerk, PA3FWM On Sun, Jan 04, 2015 at 04:11:22PM +0100, Markus Vester wrote: > [....] > > The ultimate goal of this work has been to take decoding sensitivity = close to > the theroretical limit. An universal metric for this is Eb/N0, the = ratio of the > received signal energy per payload bit (Eb in Joules) and the noise = spectral > power density (N0 in Watt/Hertz, equivalent to "noise energy" in = Joules). The > Shannon limit for long messages spread to infinite bandwidth is > Eb/N0 =3D ln(2) =3D -1.59 dB, > which (similar to the speed of light) cannot be surpassed by any = possible > encoding scheme. Paul's and Dex' experiments showed that his codes can = come > within about a dB of this limit in a real long-distance propagation = experiment. > =20 > To put that into perspective, let's derive Eb/N0 figures for two = popular > digital modes: > =20 > WSPR-15 transmits 50 information bits in 15 minutes, ie one bit in 18 = seconds. > The decoding threshold is -38 dB in 2.5 kHz, or -4 dBHz. This gives > Eb/N0 =3D 10 log(18) - 4 dB =3D +8.5 dB, > ie. about 10 dB above the Shannon limit. Note that although different = speed > variants (eg WSPR-2) need different power, the minimum energy per bit = has to > remain the same. > =20 > Opera-32 carries 28 information bits in 32.6 minutes, ie. one bit in = 70 > seconds. The threshold is about -39.5 dB in 2.5 kHz, (-5.5 dBHz), = referenced to > the average power of the 50% dutycycle on-off keying. This gives > Eb/N0 =3D 10 log(70) - 5.5 dB =3D +13 dB > or about 14.5 dB above Shannon. Note however that for LF / VLF = transmissions, > the limit will often be antenna voltage and peak power rather than = average > power, which can result in a further 3 dB disadvantage for Opera = against > frequency- or phase-modulated techniques. =20 > =20 > The opds correlation decoder can go about 9 dB lower than Opera. But = of course > it can only find the best match from an a-priori defined list of = callsigns, and > doesn't attempt to decode any message. > =20 > However we must recognize that the amateur modes spend a significant = part of > their energy to provide a reference for synchronisation, so not all of = the Eb/ > N0 difference is due to less efficient encoding. The "nude" FEC-PSK = mode > doesn't contain any such overhead. So it can only work when the link = has a > stable phase (like on VLF), and the decoder has been given accurate = information > on carrier frequency and symbol timing. > =20 > All the best, > Markus (DF6NM)=20 >=20 ------=_NextPart_000_0020_01D02B9B.6BE0BC40 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Thanks for this Pieter!
 
There are a couple more = modes that have=20 been used on LF, and of which I'd be interested to see quantitative = SNR=20 threshold data:
 
- Wolf PSK-10 with variable long = integration from=20 Stewart, KK7KA http://www.scgroup.com/ham/wolf.html,
- Jason MFSK-17 from Alberto, = I2PHD=20 http://www.weaksignals.com/,
- WSQ realtime MFSK from Con ZL2AFP and = Murray=20 ZL1BPU http://www.qsl.net/zl1bpu/SOFT/WSQ.htm.
 
Maybe someone in the group can = help?
 
Markus (DF6NM)

Sent: Thursday, January = 08, 2015 11:07=20 PM
To: rsgb_lf_group@blacksheep.org=20
Subject: LF: Eb/N0 values = for amateur=20 modes


A couple of days ago, Markus posted Eb/N0 numbers = for WSPR and=20 OPERA,
for comparison to the BPSK VLF tests (see below).
Some may = be=20 interested in a table with Eb/N0 values for some more amateur
modes, = which I=20 composed some weeks ago for an article in the Dutch amateur
radio = magazine=20 'Electron':

          &nbs= p;            = ;   =20 Needed SNR    Net datarate  =20 Needed
         =20 Mode           &nb= sp; in=20 2500 Hz     in bits/s     =20 Eb/N0        =20 Comments
          &= nbsp;           &n= bsp;           &nb= sp;           &nbs= p;        =20
        SSB=20 voice           &n= bsp;=20 +10 dB         =20 20        +31=20 dB         very rough = estimate
 =20 CW (ZRO-test, by ear)       -18=20 dB          =20 0.54     +16=20 dB         based on avg.pwr; = peak 3 dB=20 higher
  CW (QRSS-3, waterfall)      = -26=20 dB          =20 0.13     +14=20 dB         same
CW (RSCW = software, 12=20 wpm)    -12=20 dB          =20 4        +13=20 dB        =20 same
       =20 OPERA-2           =    =20 -23 dB          =20 0.23     +14=20 dB         peak pwr. 3 dB = higher; 2 dB=20 lower if counting CRC-bits as=20 information
         =20 RTTY           &nb= sp;    =20 -5 dB         =20 32        +14=20 dB
         =20 PSK31           &n= bsp;  =20 -10 dB         =20 31         +9=20 dB
         =20 WSPR           &nb= sp;   =20 -29 dB          =20 0.45      +5=20 dB         not counting energy = in sync=20 bits; otherwise 3 dB = higher
       =20 WSPR-15           =    =20 -38 dB          =20 0.056     +5=20 dB        =20 same
     JT65 (for=20 EME)           -24=20 dB          =20 1.54      +5=20 dB         same
  = Coherent BPSK=20 on VLF        -57=20 dB          =20 0.0058    -1 dB
     Theoretical=20 limit           &n= bsp;           &nb= sp;          =20 -1.59 dB

(The table is formatted for display using a = fixed-width=20 font.)

I have to emphasize that the SNR and Eb/N0 values should = be taken=20 with a
grain of salt, especially for the modes that do not use heavy = FEC,=20 since
those don't have a no sharp threshold.
The SNR values quoted = come=20 from a variety of sources on the web, and the
BPSK/VLF line is based = on=20 Markus and Paul's experiments from last May
(obviously, since I = created the=20 table before the recent experiments by Dex
and Paul).

73,=20 Pieter-Tjerk, PA3FWM


On Sun, Jan 04, 2015 at 04:11:22PM = +0100, Markus=20 Vester wrote:

> [....]
>
> The ultimate goal of = this work=20 has been to take decoding sensitivity close to
> the theroretical = limit.=20 An universal metric for this is Eb/N0, the ratio of the
> received = signal=20 energy per payload bit (Eb in Joules) and the noise spectral
> = power=20 density (N0 in Watt/Hertz, equivalent to "noise energy" in Joules). = The
>=20 Shannon limit for long messages spread to infinite bandwidth = is
> =20 Eb/N0 =3D ln(2) =3D -1.59 dB,
> which (similar to the speed of = light) cannot=20 be surpassed by any possible
> encoding scheme. Paul's and Dex'=20 experiments showed that his codes can come
> within about a dB of = this=20 limit in a real long-distance propagation experiment.
>  =
> To=20 put that into perspective, let's derive Eb/N0 figures for two = popular
>=20 digital modes:
> 
> WSPR-15 transmits 50 information = bits in=20 15 minutes, ie one bit in 18 seconds.
> The decoding threshold is = -38 dB=20 in 2.5 kHz, or -4 dBHz. This gives
>  Eb/N0 =3D 10 log(18) - = 4 dB =3D=20 +8.5 dB,
> ie. about 10 dB above the Shannon limit. Note that = although=20 different speed
> variants (eg WSPR-2) need different power, the = minimum=20 energy per bit has to
> remain the same.
> 
> = Opera-32=20 carries 28 information bits in 32.6 minutes, ie. one bit in 70
> = seconds.=20 The threshold is about -39.5 dB in 2.5 kHz, (-5.5 dBHz), referenced = to
>=20 the average power of the 50% dutycycle on-off keying. This = gives
> =20 Eb/N0 =3D 10 log(70) - 5.5 dB =3D +13 dB
> or about 14.5 dB above = Shannon.=20 Note however that for LF / VLF transmissions,
> the limit will = often be=20 antenna voltage and peak power rather than average
> power, which = can=20 result in a further 3 dB disadvantage for Opera against
> = frequency- or=20 phase-modulated techniques. 
> 
> The opds = correlation=20 decoder can go about 9 dB lower than Opera. But of course
> it can = only=20 find the best match from an a-priori defined list of callsigns, = and
>=20 doesn't attempt to decode any message.
> 
> However we = must=20 recognize that the amateur modes spend a significant part of
> = their=20 energy to provide a reference for synchronisation, so not all of the = Eb/
>=20 N0 difference is due to less efficient encoding. The "nude" FEC-PSK = mode
>=20 doesn't contain any such overhead. So it can only work when the link has = a
> stable phase (like on VLF), and the decoder has been given = accurate=20 information
> on carrier frequency and symbol = timing.
> =20
> All the best,
> Markus (DF6NM)
>=20


------=_NextPart_000_0020_01D02B9B.6BE0BC40--