Re: Topband: FT8 - How it really works
Ok, ok. I said it wrong. When I think of 160, I say “night,” It was about 4:00 pm local time. Chuck W5PR On Mon, Dec 24, 2018 at 5:46 PM JC wrote: > Jerry > > The new mode FT8 is not all that new, actually, there are several aspects > to > consider, like detect the signal, decode the signal detected, make a > decision to accept the decoded signal. The improvement on signal to noise > ratio concept is very old, just the internet made it possible with time > synchronization. The decode uses new algorithms and some very intelligent > way to guest the decoded signal. > > Check this out. 1975 Sept QST; Coherent cw test! Experiments show 20 db > Signal Boost over QRM, > > http://www.arrl.org/files/file/Technology/tis/info/pdf/7509026.pdf > > The improvement on signal to noise ratio is just because a narrow > bandwidth. > The gates opens at the right millisecond window. On FSK the secret salvage > is time synchronization. You can record the audio and play it back, the > decode will happen only if you synchronize the time of the recording with > the time in ms of the PC clock. > > I did that, and it worked, I have a SDR QS1R and using HDSDR software to > record the I/Q file, RF file. I used to record rare DX expedition signal > and > the bandwidth is 50 KHz, I can see the FT8 guys on 1840, My question was , > can I decode them from the digital file recorded several month ago? > > I started plaining the file at the top of the second count, and voalah!!!, > The WSJT-X decoded several station, weak as -21 db. The weak signals are > there, buried in the noise on my old digital recorded file. > > Then I decided to test my HWF, the practical result measuring cw signal is > that the signal to noise increase around 20 db, 10 db due the directivity > RDF 11.5 and another 10 db from the polarization filter. The Horizontal WF > attenuation on vertical signals is over -90 db. The manmade noise vertical > polarized is reduced below the MDS of the receiver and cannot be amplified > by the receiver. > > The IC-7800 has two identical receivers. I connected my HWF on receiver > MAIN > and the TX antenna on the receiver SUB, I installed two instances of the > WSJTX program, one for each receiver. After 15 minutes the number of > decodes > on the HWF was 20 or times more than the vertical full size vertical, my TX > antenna 120 Ft high. > > Signals decoded around -21 db on the vertical was decoded on the HWF 0 to > +1 > db. Signals less the -5db decoded on the HWF was not decoded using the > vertical, The HWF was decoding hundreds of signals that would be -40 db on > the decode using the vertical. > > I think the s/n reported by the program as ball part is actually very good > and close to the real s/n improvement of 2 Hz BW, depending on the mode. > > The only real way to increase signal to noise ratio is increasing the > directivity of the RX antenna, more real RDF means real signal to noise > ratio improvement. I used real because it is very easy to destroy the > directivity with integration, leaking, intermodulation, low noise figure > etc. > > One bad concept, bidirectional unterminated beverage with two lobes one in > the back and one front, it just does not work because the RDF is 6 db down > a > terminated beverage. Same for BOG's the RDF is bad, a K9AY works better > because has more RDF. A simple Flag can deliver 9 db RDF is tis easy to > hide > too. Two Flags in phase 11.5 db and four Flags 14 db RDF, and a very clean > pattern besides real broadband from 1 MHz to 10 MHz > > As you can see on the ARRL 1975 article, there is nothing new about > improvement of signal to noise ratio reducing the bandwidth. On the > article, > the test was CW at 12 wpm and 9 Hz filter BW , no ring using WWV as time > source for the synchronization. > > That was state of the art back in the early 70's, almost 50 years ago. > > 73's > JC > N4IS > > > -Original Message- > From: Topband [mailto:topband-boun...@contesting.com] On Behalf Of K4SAV > Sent: Monday, December 24, 2018 3:10 PM > To: topband@contesting.com > Subject: Re: Topband: FT8 - How it really works > > Although I have finished my FT8 testing, there is one final thought I would > like to leave with you, and also to correct one statement I made earlier. > Someone thought FT8 measured the noise in the interval when the FT8 signals > were off, and I replied that would result in a real S/N number. That is > not > true as you will see in the info below. You would get a real S/N number if > the RF was sampled, but not if the audio is sampled. > > I spent many years designing electronic circuits professionally, so I still > think that way. So for a few minutes
Re: Topband: FT8 - How it really works
Jerry The new mode FT8 is not all that new, actually, there are several aspects to consider, like detect the signal, decode the signal detected, make a decision to accept the decoded signal. The improvement on signal to noise ratio concept is very old, just the internet made it possible with time synchronization. The decode uses new algorithms and some very intelligent way to guest the decoded signal. Check this out. 1975 Sept QST; Coherent cw test! Experiments show 20 db Signal Boost over QRM, http://www.arrl.org/files/file/Technology/tis/info/pdf/7509026.pdf The improvement on signal to noise ratio is just because a narrow bandwidth. The gates opens at the right millisecond window. On FSK the secret salvage is time synchronization. You can record the audio and play it back, the decode will happen only if you synchronize the time of the recording with the time in ms of the PC clock. I did that, and it worked, I have a SDR QS1R and using HDSDR software to record the I/Q file, RF file. I used to record rare DX expedition signal and the bandwidth is 50 KHz, I can see the FT8 guys on 1840, My question was , can I decode them from the digital file recorded several month ago? I started plaining the file at the top of the second count, and voalah!!!, The WSJT-X decoded several station, weak as -21 db. The weak signals are there, buried in the noise on my old digital recorded file. Then I decided to test my HWF, the practical result measuring cw signal is that the signal to noise increase around 20 db, 10 db due the directivity RDF 11.5 and another 10 db from the polarization filter. The Horizontal WF attenuation on vertical signals is over -90 db. The manmade noise vertical polarized is reduced below the MDS of the receiver and cannot be amplified by the receiver. The IC-7800 has two identical receivers. I connected my HWF on receiver MAIN and the TX antenna on the receiver SUB, I installed two instances of the WSJTX program, one for each receiver. After 15 minutes the number of decodes on the HWF was 20 or times more than the vertical full size vertical, my TX antenna 120 Ft high. Signals decoded around -21 db on the vertical was decoded on the HWF 0 to +1 db. Signals less the -5db decoded on the HWF was not decoded using the vertical, The HWF was decoding hundreds of signals that would be -40 db on the decode using the vertical. I think the s/n reported by the program as ball part is actually very good and close to the real s/n improvement of 2 Hz BW, depending on the mode. The only real way to increase signal to noise ratio is increasing the directivity of the RX antenna, more real RDF means real signal to noise ratio improvement. I used real because it is very easy to destroy the directivity with integration, leaking, intermodulation, low noise figure etc. One bad concept, bidirectional unterminated beverage with two lobes one in the back and one front, it just does not work because the RDF is 6 db down a terminated beverage. Same for BOG's the RDF is bad, a K9AY works better because has more RDF. A simple Flag can deliver 9 db RDF is tis easy to hide too. Two Flags in phase 11.5 db and four Flags 14 db RDF, and a very clean pattern besides real broadband from 1 MHz to 10 MHz As you can see on the ARRL 1975 article, there is nothing new about improvement of signal to noise ratio reducing the bandwidth. On the article, the test was CW at 12 wpm and 9 Hz filter BW , no ring using WWV as time source for the synchronization. That was state of the art back in the early 70's, almost 50 years ago. 73's JC N4IS -Original Message- From: Topband [mailto:topband-boun...@contesting.com] On Behalf Of K4SAV Sent: Monday, December 24, 2018 3:10 PM To: topband@contesting.com Subject: Re: Topband: FT8 - How it really works Although I have finished my FT8 testing, there is one final thought I would like to leave with you, and also to correct one statement I made earlier. Someone thought FT8 measured the noise in the interval when the FT8 signals were off, and I replied that would result in a real S/N number. That is not true as you will see in the info below. You would get a real S/N number if the RF was sampled, but not if the audio is sampled. I spent many years designing electronic circuits professionally, so I still think that way. So for a few minutes lets think about a circuit that can decode something below the noise floor .If you think about FT8 or anything similar, from a designers point of view, you suddenly realize that making a statement of "the circuit can decode down to X dBs below the noise floor" is almost an impossible task, that is, if you are talking RF noise floor as most people will be assuming. Since you will be dealing with audio, not RF, the receiver will convert the RF into audio and compress it into something that has a lot less dynamic range. How much less? Say the volume is set to a level such that the strongest signals do not clip, then ho
Re: Topband: FT8 - How it really works
Chuck, I suspect something wasn’t quite right with your setup? At my QTH in W3 I can hear multiple FT8 signals on 1840kc USB (2.4khz bandwidth) from before my sunset until after sunrise. They are whining/droning carriers for 13 seconds every 15 seconds. Only for that less than 2 seconds every 15 seconds do I hear just band noise, at all other times I hear multiple FT8 signals just fine. And the computer can decode more signals than I can hear. Tim N3QE > On Dec 24, 2018, at 4:17 PM, Chuck Dietz wrote: > > I think I understand much of what you are saying, but I know that I was on > 160 meter FT=8 two nights ago with the speaker up fairly loud. I only heard > noise. I set the AGC off and adjusted the RF gain so that it did not > overload. Still no hint of any signals, but I decoded two stations! > > Just sayin’. > > Chuck W5PR > > Sent from Mail for Windows 10 > > From: K4SAV > Sent: Monday, December 24, 2018 2:10 PM > To: topband@contesting.com > Subject: Re: Topband: FT8 - How it really works > > Although I have finished my FT8 testing, there is one final thought I > would like to leave with you, and also to correct one statement I made > earlier. Someone thought FT8 measured the noise in the interval when > the FT8 signals were off, and I replied that would result in a real S/N > number. That is not true as you will see in the info below. You would > get a real S/N number if the RF was sampled, but not if the audio is > sampled. > > I spent many years designing electronic circuits professionally, so I > still think that way. So for a few minutes lets think about a circuit > that can decode something below the noise floor .If you think about FT8 > or anything similar, from a designers point of view, you suddenly > realize that making a statement of "the circuit can decode down to X dBs > below the noise floor" is almost an impossible task, that is, if you are > talking RF noise floor as most people will be assuming. > > Since you will be dealing with audio, not RF, the receiver will convert > the RF into audio and compress it into something that has a lot less > dynamic range. How much less? Say the volume is set to a level such > that the strongest signals do not clip, then how far down is the noise? > You can expect that to vary on each band too. > > Now comes a real complication. If you were taking samples in the RF > world, you could see the noise level on your S meter and estimate it > relative to the strongest signals. However your circuit will be dealing > with audio. Surprisingly, when the signals disappear, the receiver AGC > voltage drops and the receiver gain increases. That produces a lot more > audio signal. The audio noise in the case of no signals becomes higher > than the audio level for strong signals if you are using USB bandwidth > and receiving something similar to FT8. That condition is not nearly as > pronounced when using a narrow CW bandwidth. Even if you put the > receiver into AGC slow mode it won't hold for the 3 seconds when FT8 is > off, so you still get the increased audio in the off period. Then there > will be a sudden increase in audio when the first signal reappears, > until the ACG kicks in and lowers it. This happens even with fast AGC > selected. It's fast enough that you don't notice it when listening, but > if you put a scope on it you can see it. Yeah, all that surprised me > too when first thinking about it. Take a close listen and see if you > agree. If you can't hear it, put it on a scope or anything that displays > an audio waveform and it will become very obvious. > > If you made a statement that this circuit can decode X dBs below the > noise floor, most people will be thinking RF noise floor. So what is it > in the audio world that represents the noise floor in the RF world, and > what would your statement mean? > > Of course you could turn off the AGC and decrease the receiver RF gain > and that would make the audio very low when the signals disappear. That > would also severely limit the dynamic range for your circuit since you > would no longer have the compression supplied by the receiver.. Your > circuit would have to cover a much wider dynamic range, similar to what > a receiver does. So your circuit would need what? maybe 100 dB dynamic > range to cover the strongest signals to the weakest noise floor, > forgetting about decoding below the noise floor. Actually that wouldn't > really happen because receivers can't produce a dynamic range of 100 dB > in the audio. They may do it in the RF world, but not in audio. > Receivers have no need to do that. > > Jerry > _ > Searchable Archives: http://www.contesting.com/_topband - Topband Reflector > > _ > Searchable Archives: http://www.contesting.com/_topband - Topband Reflector _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Hi, we use 4ports-RX-splitter https://www.ebay.com/itm/4-WAY-HF-ANTENNA-SPLITTER-COMBINER-RX-0-1-50-MHz-SO-239-connectors/322564884873?hash=item4b1a5d8989:g:JJ4AAOxyM89Sbujo:rk:1:pf:0 73! de Eugene RA0FF http://www.qsl.net/ra0ff/ >Вторник, 25 декабря 2018, 8:17 +11:00 от Chuck Dietz : > >I think I understand much of what you are saying, but I know that I was on 160 >meter FT=8 two nights ago with the speaker up fairly loud. I only heard noise. >I set the AGC off and adjusted the RF gain so that it did not overload. Still >no hint of any signals, but I decoded two stations! > >Just sayin’. > >Chuck W5PR > >Sent from Mail for Windows 10 > >From: K4SAV >Sent: Monday, December 24, 2018 2:10 PM >To: topband@contesting.com >Subject: Re: Topband: FT8 - How it really works > >Although I have finished my FT8 testing, there is one final thought I >would like to leave with you, and also to correct one statement I made >earlier. Someone thought FT8 measured the noise in the interval when >the FT8 signals were off, and I replied that would result in a real S/N >number. That is not true as you will see in the info below. You would >get a real S/N number if the RF was sampled, but not if the audio is >sampled. > >I spent many years designing electronic circuits professionally, so I >still think that way. So for a few minutes lets think about a circuit >that can decode something below the noise floor .If you think about FT8 >or anything similar, from a designers point of view, you suddenly >realize that making a statement of "the circuit can decode down to X dBs >below the noise floor" is almost an impossible task, that is, if you are >talking RF noise floor as most people will be assuming. > >Since you will be dealing with audio, not RF, the receiver will convert >the RF into audio and compress it into something that has a lot less >dynamic range. How much less? Say the volume is set to a level such >that the strongest signals do not clip, then how far down is the noise? >You can expect that to vary on each band too. > >Now comes a real complication. If you were taking samples in the RF >world, you could see the noise level on your S meter and estimate it >relative to the strongest signals. However your circuit will be dealing >with audio. Surprisingly, when the signals disappear, the receiver AGC >voltage drops and the receiver gain increases. That produces a lot more >audio signal. The audio noise in the case of no signals becomes higher >than the audio level for strong signals if you are using USB bandwidth >and receiving something similar to FT8. That condition is not nearly as >pronounced when using a narrow CW bandwidth. Even if you put the >receiver into AGC slow mode it won't hold for the 3 seconds when FT8 is >off, so you still get the increased audio in the off period. Then there >will be a sudden increase in audio when the first signal reappears, >until the ACG kicks in and lowers it. This happens even with fast AGC >selected. It's fast enough that you don't notice it when listening, but >if you put a scope on it you can see it. Yeah, all that surprised me >too when first thinking about it. Take a close listen and see if you >agree. If you can't hear it, put it on a scope or anything that displays >an audio waveform and it will become very obvious. > >If you made a statement that this circuit can decode X dBs below the >noise floor, most people will be thinking RF noise floor. So what is it >in the audio world that represents the noise floor in the RF world, and >what would your statement mean? > >Of course you could turn off the AGC and decrease the receiver RF gain >and that would make the audio very low when the signals disappear. That >would also severely limit the dynamic range for your circuit since you >would no longer have the compression supplied by the receiver.. Your >circuit would have to cover a much wider dynamic range, similar to what >a receiver does. So your circuit would need what? maybe 100 dB dynamic >range to cover the strongest signals to the weakest noise floor, >forgetting about decoding below the noise floor. Actually that wouldn't >really happen because receivers can't produce a dynamic range of 100 dB >in the audio. They may do it in the RF world, but not in audio. >Receivers have no need to do that. > >Jerry >_ >Searchable Archives: http://www.contesting.com/_topband - Topband Reflector > >_ >Searchable Archives: http://www.contesting.com/_topband - Topband Reflector _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
I think I understand much of what you are saying, but I know that I was on 160 meter FT=8 two nights ago with the speaker up fairly loud. I only heard noise. I set the AGC off and adjusted the RF gain so that it did not overload. Still no hint of any signals, but I decoded two stations! Just sayin’. Chuck W5PR Sent from Mail for Windows 10 From: K4SAV Sent: Monday, December 24, 2018 2:10 PM To: topband@contesting.com Subject: Re: Topband: FT8 - How it really works Although I have finished my FT8 testing, there is one final thought I would like to leave with you, and also to correct one statement I made earlier. Someone thought FT8 measured the noise in the interval when the FT8 signals were off, and I replied that would result in a real S/N number. That is not true as you will see in the info below. You would get a real S/N number if the RF was sampled, but not if the audio is sampled. I spent many years designing electronic circuits professionally, so I still think that way. So for a few minutes lets think about a circuit that can decode something below the noise floor .If you think about FT8 or anything similar, from a designers point of view, you suddenly realize that making a statement of "the circuit can decode down to X dBs below the noise floor" is almost an impossible task, that is, if you are talking RF noise floor as most people will be assuming. Since you will be dealing with audio, not RF, the receiver will convert the RF into audio and compress it into something that has a lot less dynamic range. How much less? Say the volume is set to a level such that the strongest signals do not clip, then how far down is the noise? You can expect that to vary on each band too. Now comes a real complication. If you were taking samples in the RF world, you could see the noise level on your S meter and estimate it relative to the strongest signals. However your circuit will be dealing with audio. Surprisingly, when the signals disappear, the receiver AGC voltage drops and the receiver gain increases. That produces a lot more audio signal. The audio noise in the case of no signals becomes higher than the audio level for strong signals if you are using USB bandwidth and receiving something similar to FT8. That condition is not nearly as pronounced when using a narrow CW bandwidth. Even if you put the receiver into AGC slow mode it won't hold for the 3 seconds when FT8 is off, so you still get the increased audio in the off period. Then there will be a sudden increase in audio when the first signal reappears, until the ACG kicks in and lowers it. This happens even with fast AGC selected. It's fast enough that you don't notice it when listening, but if you put a scope on it you can see it. Yeah, all that surprised me too when first thinking about it. Take a close listen and see if you agree. If you can't hear it, put it on a scope or anything that displays an audio waveform and it will become very obvious. If you made a statement that this circuit can decode X dBs below the noise floor, most people will be thinking RF noise floor. So what is it in the audio world that represents the noise floor in the RF world, and what would your statement mean? Of course you could turn off the AGC and decrease the receiver RF gain and that would make the audio very low when the signals disappear. That would also severely limit the dynamic range for your circuit since you would no longer have the compression supplied by the receiver.. Your circuit would have to cover a much wider dynamic range, similar to what a receiver does. So your circuit would need what? maybe 100 dB dynamic range to cover the strongest signals to the weakest noise floor, forgetting about decoding below the noise floor. Actually that wouldn't really happen because receivers can't produce a dynamic range of 100 dB in the audio. They may do it in the RF world, but not in audio. Receivers have no need to do that. Jerry _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Although I have finished my FT8 testing, there is one final thought I would like to leave with you, and also to correct one statement I made earlier. Someone thought FT8 measured the noise in the interval when the FT8 signals were off, and I replied that would result in a real S/N number. That is not true as you will see in the info below. You would get a real S/N number if the RF was sampled, but not if the audio is sampled. I spent many years designing electronic circuits professionally, so I still think that way. So for a few minutes lets think about a circuit that can decode something below the noise floor .If you think about FT8 or anything similar, from a designers point of view, you suddenly realize that making a statement of "the circuit can decode down to X dBs below the noise floor" is almost an impossible task, that is, if you are talking RF noise floor as most people will be assuming. Since you will be dealing with audio, not RF, the receiver will convert the RF into audio and compress it into something that has a lot less dynamic range. How much less? Say the volume is set to a level such that the strongest signals do not clip, then how far down is the noise? You can expect that to vary on each band too. Now comes a real complication. If you were taking samples in the RF world, you could see the noise level on your S meter and estimate it relative to the strongest signals. However your circuit will be dealing with audio. Surprisingly, when the signals disappear, the receiver AGC voltage drops and the receiver gain increases. That produces a lot more audio signal. The audio noise in the case of no signals becomes higher than the audio level for strong signals if you are using USB bandwidth and receiving something similar to FT8. That condition is not nearly as pronounced when using a narrow CW bandwidth. Even if you put the receiver into AGC slow mode it won't hold for the 3 seconds when FT8 is off, so you still get the increased audio in the off period. Then there will be a sudden increase in audio when the first signal reappears, until the ACG kicks in and lowers it. This happens even with fast AGC selected. It's fast enough that you don't notice it when listening, but if you put a scope on it you can see it. Yeah, all that surprised me too when first thinking about it. Take a close listen and see if you agree. If you can't hear it, put it on a scope or anything that displays an audio waveform and it will become very obvious. If you made a statement that this circuit can decode X dBs below the noise floor, most people will be thinking RF noise floor. So what is it in the audio world that represents the noise floor in the RF world, and what would your statement mean? Of course you could turn off the AGC and decrease the receiver RF gain and that would make the audio very low when the signals disappear. That would also severely limit the dynamic range for your circuit since you would no longer have the compression supplied by the receiver.. Your circuit would have to cover a much wider dynamic range, similar to what a receiver does. So your circuit would need what? maybe 100 dB dynamic range to cover the strongest signals to the weakest noise floor, forgetting about decoding below the noise floor. Actually that wouldn't really happen because receivers can't produce a dynamic range of 100 dB in the audio. They may do it in the RF world, but not in audio. Receivers have no need to do that. Jerry _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Jerry, If you would like to do some mid-day comparison testing between FT8 and CW, let me know. Looks like the path length is about 400 miles. Jim, K9SE _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Still doing FT8 testing this morning 3 hours after sunrise I looked for something resembling dead band conditions with only a few weak stations. There was nothing on 160 but the west coast guys were still on 80. They were weak but there. They would have been good copy on CW but FT8 was having much difficulty. It was getting only about one decode every 5 minutes or so. Never got a decode of someone sending CQ so that I could see what FT8 was calculating for a report. This was on my receiving antenna. I couldn't hear them on the transmit antenna and FT8 never decoded anything on that antenna. Based on my tests of mixing noise into the audio and seeing FT8 decode below the noise floor, I really expected to see better performance from FT8, but it didn't happen. The dead band test yesterday didn't bode well for FT8 either. I think for the test yesterday when mixing in audio noise to cover up the signals, I had FT8 in deep decode mode. I wonder if those decodes were just guesses because it knew which stations were on which frequencies before I added that noise. This deserves a retest. So I did a retest. This time I set up the noise such that I could just barely hear the FT8 signals. I did that before I let WSJTX hear the band, and I set it to decode normally. It didn't do nearly as well. There weren't many decodes and the signals were being reported in the range -12 to -17 dB. There were no -24 dB reports like the last test. where they were all -24 dB I changed the mode to deep decode and let WSJTX look at the band for a while. Then I added the noise back. Now there are many more decodes and the reported numbers were much lower. Hmm. Guesses? I think I have concluded my FT8 testing. My curiosity of how this works is sufficiently satisfied. Others can form their opinions from this data, their own tests, or from whatever source they desire. Thanks to others that showed some interest in this experiment. Now that I am feeling better, hope to be up early enough to work some of that early morning DX that I have been missing and all you other guys have been getting. (CW of course) Jerry, K4SAV On 12/21/2018 5:11 PM, K4SAV wrote: Yesterday I said " I don't have a measurement with the results of that showdown of CW versus FT8 in dead band conditions but the answer would be interesting to know." This afternoon I tried to get an answer to that. I wanted to see if FT8 would decode a station I couldn't hear. I wanted to find that happen and then switch to narrow band CW and see if I could hear him. I first went to 6 meters but there was nothing there. Then I went to 160 and nothing there. I waited until it was sunset in the northeast. As soon as that happens the band fills up with stations from that area. It's not dark here but the S meter immediately goes to S9 plus. NEWS FLASH for those guys. You can work stations at any time on 160. You don't have to wait until dark. High power, CW, and a good receiving antenna is good for about 600-800 miles on 160 at noon. Low power should also be good but at a little less distance. Don't know what FT8 will do to the distance. All those stations were defeating my test to find a condition close to that of a dead band and a weak signal or two. So I switched to my receiving antenna and pointed it west to reduce signal amplitudes and still have some noise. I can still hear a bunch of signals in there but FT8 isn't decoding anything with the antenna pointed west. There seems to be some high level static when pointed west. It's not noticeable on CW but it sounds bad on USB. Maybe that is causing FT8 to not decode anything. So I consider this test pretty much unsuccessful. If anyone else runs a similar test please post the results. I will try again. I know this isn't a top priority question for most people, but curiosity gets to me sometimes. Jerry, K4SAV _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Yesterday I said " I don't have a measurement with the results of that showdown of CW versus FT8 in dead band conditions but the answer would be interesting to know." This afternoon I tried to get an answer to that. I wanted to see if FT8 would decode a station I couldn't hear. I wanted to find that happen and then switch to narrow band CW and see if I could hear him. I first went to 6 meters but there was nothing there. Then I went to 160 and nothing there. I waited until it was sunset in the northeast. As soon as that happens the band fills up with stations from that area. It's not dark here but the S meter immediately goes to S9 plus. NEWS FLASH for those guys. You can work stations at any time on 160. You don't have to wait until dark. High power, CW, and a good receiving antenna is good for about 600-800 miles on 160 at noon. Low power should also be good but at a little less distance. Don't know what FT8 will do to the distance. All those stations were defeating my test to find a condition close to that of a dead band and a weak signal or two. So I switched to my receiving antenna and pointed it west to reduce signal amplitudes and still have some noise. I can still hear a bunch of signals in there but FT8 isn't decoding anything with the antenna pointed west. There seems to be some high level static when pointed west. It's not noticeable on CW but it sounds bad on USB. Maybe that is causing FT8 to not decode anything. So I consider this test pretty much unsuccessful. If anyone else runs a similar test please post the results. I will try again. I know this isn't a top priority question for most people, but curiosity gets to me sometimes. Jerry, K4SAV _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Jerry: Clever way to evaluate FT8 SNR reports with your VFO 1 and 2 comparisons. I often see positive numbers on FT8 160 meter signal reports for strong stations on FT8. For example -- if a station is S9 + 10 dB audible, then it reads a positive number for the FT8 signal report. At the risk of giving away a tip to improve FT8 SNR performance -- I use my 500 Hz CW filter for FT8 at times. It can definitely improve decoding on DX signals, particularly if many strong signals are in the 2,500 Hz passband on 160 meters. I encourage everyone to upgrade your software to WSJT-X v 2.0 Best wishes to all for the Holidays -- Jon N0JK While sitting around being bored and recovering from a gall bladder operation, I decided to do some experiments with FT8. First thing I did was upgrade the software to WSJT-X v2.0. I hope this post doesn't turn into another FT8 bashing session. My only goal was to understand how this mode works and what it can do and what it cannot do. -- Jerry, K4SAV _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Correct me if im wrong: A K3 owner could easily check this. You need 2 computers, running WSJT on both. Don't know if you can run 2 instances of WSJT on a single computer. Feed the left channel from LINE OUT (=mainRX) to one instance of FT-8 and the right channel (=sub-RX) to the other instance. Crank down the bandwith to 1khz or less on main RX and leave the bandwith on sub RX 2.7khz or wider. While your computers do the decoding, you can hear the 2 RXes in your headphone (or not). > In a condition of only one weak signal on the band, I haven't run a > test > that says whether FT8 decodes better than CW or not. NN4T said that > using FT8 on 6 meter sporadic E that he observes signals being decoded > with no audio in the receiver. That is probably with a wide > bandwidth, > and it would be interesting to know if the signals would be audible > with > a narrow bandwidth. -- Ohne CW ist es nur CB.. 73, Martin DM4iM _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Hi,Narroving filter in WSJT-X digi modes will degrade decodes! Best performance you'll get using wide open filters on you radio.I sugest you to look at the WSJT-X archives and find the answer to your question by the author it self - K1JT.As we are radio Amateurs i beleive its worth to experiment with the filters and it would be nice to see those results posted here. So far i didnt try yet FT8 on 160m, i did it on on 6m during the summer season and 40 to 10m daily from home with simple wires. I prefer to work CW on 160m ;)On 6m i have almost no noise, worked about 106 DXCC's in 3 months or less, 200 NA stations and about 60 JAs in the log. Many of those are Topbanders.With modest setup, old IC706MK2 + 3cx800, 6el.YU7EF and of course JTDX software (worth to try it). 160m and 6m are very different, starting from noise level, QSB and propagation modes. Multiple receivers are needed with multiple JT program's running, each his own dedicated QTF direction. Otherwise you will loose a needed DX for sure. One thing we agreed b4 the season, that EU will transmit always on the 1st period. That was good decision to minimise local QRM. Best regards, MX and 73 9A5CW Poslano s pametnog telefona Samsung Galaxy. _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
On 2018-12-20 11:45 AM, K4SAV wrote: > In the case of a crowded band it becomes obvious that CW is much > superior to decoding a weak signal because all those strong signals > limit the ability of FT8 to decode a weak signal. That is only true if you leave AGC enabled and the strong signals result in the AGC decreasing the overall sensitivity. The WSJT-X Users Guide (instructions) recommend turning off AGC. It is possible (depending on the transceiver design) to reduce the IF bandwidth to remove some or all of the strong signals to reduce overload and minimize the gain reduction by using the RF Gain control to introduce the least gain reduction necessary to prevent distortion. On 160 meters, typical nighttime noise levels are as high as -90 to -100 dBm. Using the RF attenuator and/or RF gain to set the recovered audio from that noise approximately 15 dB above the sound card noise floor should provide sufficient dynamic range to decode the weakest signal (~ 10 dB below the 200 Hz noise floor) while not distorting on the strong (S9+40 dB) signals if one has a *quality* sound card. 73, ... Joe, W4TV On 2018-12-20 11:45 AM, K4SAV wrote: Thanks to the folks commenting on how FT8 works. VE3KI said: "The noise floor the wsjt-x signal is referenced to is the noise within the bandpass during the two-second period when no-one is transmitting, not the signal level when people are transmitting." That was what I originally thought might be a possibility because that would result in a real S/N number. However that doesn't seem to be the case. That S9+40 dB signal I referenced would result in a huge S/N number, probably greater than 50 dB. FT8 gives a report of -1 dB. Doing it that way would also have some problems produced by people transmitting at the wrong time and other out of band stations, however it seems that FT8 doesn't make that measurement. Thanks to Arunas, LY2IJ . Your comments agree 100% with what I measured. As to your question of can FT8 decode signals below the noise floor and below the level that can be decoded by CW. My experiment of adding audio noise which covered up the signals and the software still being able to decode signals says that under some conditions FT8 can decode signals below the noise floor. Of course that experiment was done at audio levels, not at RF. If you use CW you get the benefit of a much narrowed passband, I can't run that test using audio mixing. In a condition of only one weak signal on the band, I haven't run a test that says whether FT8 decodes better than CW or not. NN4T said that using FT8 on 6 meter sporadic E that he observes signals being decoded with no audio in the receiver. That is probably with a wide bandwidth, and it would be interesting to know if the signals would be audible with a narrow bandwidth. In the case of a crowded band it becomes obvious that CW is much superior to decoding a weak signal because all those strong signals limit the ability of FT8 to decode a weak signal. That was the basis for my conclusion that FT8 didn't seem to be useful for working weak signal DX because most of the bands are very crowded. However there may be a case where FT8 can beat CW, that is if you are tying to decode a signal on an essentially dead band. Since FT8 seems to be able to decode below the noise floor, the noise floor in that case would just be real noise, not signals. You could improve FT8's ability to decode by narrowing the bandwidth, although that's not normally done. That would cut into CW's advantage obtained by using a narrow passband. I don't have a measurement with the results of that showdown of CW versus FT8 in dead band conditions but the answer would be interesting to know. Jerry, K4SAV _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
Thanks to the folks commenting on how FT8 works. VE3KI said: "The noise floor the wsjt-x signal is referenced to is the noise within the bandpass during the two-second period when no-one is transmitting, not the signal level when people are transmitting." That was what I originally thought might be a possibility because that would result in a real S/N number. However that doesn't seem to be the case. That S9+40 dB signal I referenced would result in a huge S/N number, probably greater than 50 dB. FT8 gives a report of -1 dB. Doing it that way would also have some problems produced by people transmitting at the wrong time and other out of band stations, however it seems that FT8 doesn't make that measurement. Thanks to Arunas, LY2IJ . Your comments agree 100% with what I measured. As to your question of can FT8 decode signals below the noise floor and below the level that can be decoded by CW. My experiment of adding audio noise which covered up the signals and the software still being able to decode signals says that under some conditions FT8 can decode signals below the noise floor. Of course that experiment was done at audio levels, not at RF. If you use CW you get the benefit of a much narrowed passband, I can't run that test using audio mixing. In a condition of only one weak signal on the band, I haven't run a test that says whether FT8 decodes better than CW or not. NN4T said that using FT8 on 6 meter sporadic E that he observes signals being decoded with no audio in the receiver. That is probably with a wide bandwidth, and it would be interesting to know if the signals would be audible with a narrow bandwidth. In the case of a crowded band it becomes obvious that CW is much superior to decoding a weak signal because all those strong signals limit the ability of FT8 to decode a weak signal. That was the basis for my conclusion that FT8 didn't seem to be useful for working weak signal DX because most of the bands are very crowded. However there may be a case where FT8 can beat CW, that is if you are tying to decode a signal on an essentially dead band. Since FT8 seems to be able to decode below the noise floor, the noise floor in that case would just be real noise, not signals. You could improve FT8's ability to decode by narrowing the bandwidth, although that's not normally done. That would cut into CW's advantage obtained by using a narrow passband. I don't have a measurement with the results of that showdown of CW versus FT8 in dead band conditions but the answer would be interesting to know. Jerry, K4SAV _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
HI, Thanks for sharing experience! Question to FT8 and "noise" relations - In usual case everything in passband is noise except signal of interest. So with only one FT8 signal and white noise in passband S/N can be increased narrowing passband till it matches signal width. EME guys know that JT65 -16 dB signals are audible. But if there are many other FT8 signals (which are "noise" since are out of interest for me) then energy from all these 59++ signals are integrated as noise and my 57 signal became "very deep in the noise". If to narrow bandwidth in this case we get rid of QRMing 59++ signals (and proportional part of band white noise) - they are not integrated and our signal gets "above noise". Is this right or there is some magic excluding other FT8 signals from S/N calculations? Still very doubt in FT8 ability to be decoded 10dB below CW. -3 dB at best. 73 Arunas LY2IJ 2018-12-20 05:57, Joe Subich, W4TV rašė: >> Is the definition of "noise floor" being changed for FT8? > WSJT-X (and WSJT before that) defines noise as the integrated value > of noise (noise power) across the 2500 Hz (approximately based on > the receiver filter) receive bandwidth. > > 73, > > ... Joe, W4TV _ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
Re: Topband: FT8 - How it really works
That would be my definition of noise power also. That would not help explain the numbers produced by FT8. It's curious that my VFO1 - VFO2 measurement produces numbers very close to what FT8 reports. I have no information as to why that should be, only measurements that produce those results. Jerry, K4SAV On 12/19/2018 9:57 PM, Joe Subich, W4TV wrote: Is the definition of "noise floor" being changed for FT8? WSJT-X (and WSJT before that) defines noise as the integrated value of noise (noise power) across the 2500 Hz (approximately based on the receiver filter) receive bandwidth. 73, ... Joe, W4TV On 2018-12-19 9:57 PM, K4SAV wrote: Joe, thanks for the information. I am not exactly sure what all that means. My conclusions were based on observed data. It seems pretty obvious to me that a signal that is more than 50 dB above the noise floor should not receive a S/N number of -1 dB, which is what FT8 gives. I don't know how the information you provided can make a calculation like that. I judge that a signal reading S9+40 dB on the S meter should be more than 50 dB above the noise floor when I can tune of to a spot where there are no signals and the S meter reads about S2 or S3 in SSB mode or less than S1 in a narrow bandwidth. Is the definition of "noise floor" being changed for FT8? Jerry, K4SAV On 12/19/2018 7:27 PM, Joe Subich, W4TV wrote: On 2018-12-19 4:28 PM, K4SAV wrote: > The official documentation for FT8 says it will decode signals 24 dB > below the noise floor. That is not a correct statement most of the > time. No, that is a correct statement. Signal reports in WSJT-X for FT8, JT65 and JT9 are *all* measured *with regard to the noise in 2500 Hz*. Note that the tone filters in WSJT-X are on the order of less than 12 Hz or so wide so the SNR *for an individual tone in the DSP filter bandwidth* at 0 dB is -23 dB relative to the *total noise in 2500 Hz bandwidth*. The actual filter bandwidth will change from mode to mode due to the differences in keying rated and tone spacing ... the actual SNR limit is shown in section 17.2.7 of the WSJT_X 2.0 User Guide. CW operators understand this from experience ... a quality 200 Hz filter will have ~12 dB less noise than a 2800 Hz filter. Thus a CW signal with a 200 Hz filter will have 12 dB better SNR than the same CW signal with a 2800 Hz filter (excluding any "processing gain" from the ear- brain filter). With FT8, JT65, JT9, etc. coding (forward error correction) provides some additional SNR (called "coding gain") but the *measurement* is based on strength of the individual tone to total noise. Thus, the lowest accurate report is -24 dB although some signals will be decoded at levels below that. 73, ... Joe, W4TV On 2018-12-19 4:28 PM, K4SAV wrote: While sitting around being bored and recovering from a gall bladder operation, I decided to do some experiments with FT8. First thing I did was upgrade the software to WSJT-X v2.0. I hope this post doesn't turn into another FT8 bashing session. My only goal was to understand how this mode works and what it can do and what it cannot do. The official description of FT8's signal reporting cannot be correct. It is obviously not a signal to noise number and it is not an S meter reading. What is it? That was the first question to answer. It's obviously not an S/N number because how do you give a report of -1 dB for a signal that is S9+40 dB on a quiet band. I was unable to find any info on how the signal report was calculated so I tried to correlate those reports to observations. I think I have figured out a method that results in very close to the same number that FT8 reports. Here is the experiment. I set up my main VFO to USB 2500 Hz bandwidth and set the second VFO to CW at about 150 Hz bandwidth. I look for a station calling CQ and tune the second VFO to him and measure his signal strength. I also look at the S meter for the signal level on the main VFO. I also look at the signal report calculated by the software. For stations calling CQ that report is calculated by the software in my computer. The FT8 report is usually very close to the difference in signal levels (VFO1 - VFO2). For example if the main VFO reads S9+10 and the second VFO reads S9, the FT8 number will be -10 dB. Note that the FT8 says that -24 dB is the lowest it can decode. With VFO1 = S9+10, that's about S7 for the smallest signal it can decode. Observations agree. Those numbers will vary a little depending on how your S meter is calibrated. In order to decode a weak signal, all those close USA stations will have to go silent. The official documentation for FT8 says it will decode signals 24 dB below the noise floor. That is not a correct statement most of the time. That statement should be that FT8 will decode signals 24 dB below the sum total of everything in a 2500 Hz bandwidth. If the total of all signals on the band are below the
Re: Topband: FT8 - How it really works
Is the definition of "noise floor" being changed for FT8? WSJT-X (and WSJT before that) defines noise as the integrated value of noise (noise power) across the 2500 Hz (approximately based on the receiver filter) receive bandwidth. 73, ... Joe, W4TV On 2018-12-19 9:57 PM, K4SAV wrote: Joe, thanks for the information. I am not exactly sure what all that means. My conclusions were based on observed data. It seems pretty obvious to me that a signal that is more than 50 dB above the noise floor should not receive a S/N number of -1 dB, which is what FT8 gives. I don't know how the information you provided can make a calculation like that. I judge that a signal reading S9+40 dB on the S meter should be more than 50 dB above the noise floor when I can tune of to a spot where there are no signals and the S meter reads about S2 or S3 in SSB mode or less than S1 in a narrow bandwidth. Is the definition of "noise floor" being changed for FT8? Jerry, K4SAV On 12/19/2018 7:27 PM, Joe Subich, W4TV wrote: On 2018-12-19 4:28 PM, K4SAV wrote: > The official documentation for FT8 says it will decode signals 24 dB > below the noise floor. That is not a correct statement most of the > time. No, that is a correct statement. Signal reports in WSJT-X for FT8, JT65 and JT9 are *all* measured *with regard to the noise in 2500 Hz*. Note that the tone filters in WSJT-X are on the order of less than 12 Hz or so wide so the SNR *for an individual tone in the DSP filter bandwidth* at 0 dB is -23 dB relative to the *total noise in 2500 Hz bandwidth*. The actual filter bandwidth will change from mode to mode due to the differences in keying rated and tone spacing ... the actual SNR limit is shown in section 17.2.7 of the WSJT_X 2.0 User Guide. CW operators understand this from experience ... a quality 200 Hz filter will have ~12 dB less noise than a 2800 Hz filter. Thus a CW signal with a 200 Hz filter will have 12 dB better SNR than the same CW signal with a 2800 Hz filter (excluding any "processing gain" from the ear- brain filter). With FT8, JT65, JT9, etc. coding (forward error correction) provides some additional SNR (called "coding gain") but the *measurement* is based on strength of the individual tone to total noise. Thus, the lowest accurate report is -24 dB although some signals will be decoded at levels below that. 73, ... Joe, W4TV On 2018-12-19 4:28 PM, K4SAV wrote: While sitting around being bored and recovering from a gall bladder operation, I decided to do some experiments with FT8. First thing I did was upgrade the software to WSJT-X v2.0. I hope this post doesn't turn into another FT8 bashing session. My only goal was to understand how this mode works and what it can do and what it cannot do. The official description of FT8's signal reporting cannot be correct. It is obviously not a signal to noise number and it is not an S meter reading. What is it? That was the first question to answer. It's obviously not an S/N number because how do you give a report of -1 dB for a signal that is S9+40 dB on a quiet band. I was unable to find any info on how the signal report was calculated so I tried to correlate those reports to observations. I think I have figured out a method that results in very close to the same number that FT8 reports. Here is the experiment. I set up my main VFO to USB 2500 Hz bandwidth and set the second VFO to CW at about 150 Hz bandwidth. I look for a station calling CQ and tune the second VFO to him and measure his signal strength. I also look at the S meter for the signal level on the main VFO. I also look at the signal report calculated by the software. For stations calling CQ that report is calculated by the software in my computer. The FT8 report is usually very close to the difference in signal levels (VFO1 - VFO2). For example if the main VFO reads S9+10 and the second VFO reads S9, the FT8 number will be -10 dB. Note that the FT8 says that -24 dB is the lowest it can decode. With VFO1 = S9+10, that's about S7 for the smallest signal it can decode. Observations agree. Those numbers will vary a little depending on how your S meter is calibrated. In order to decode a weak signal, all those close USA stations will have to go silent. The official documentation for FT8 says it will decode signals 24 dB below the noise floor. That is not a correct statement most of the time. That statement should be that FT8 will decode signals 24 dB below the sum total of everything in a 2500 Hz bandwidth. If the total of all signals on the band are below the noise floor, it would be interesting to know if FT8 will decode any of them. I haven't observed that yet in a real situation. I did however try to simulate that condition by adding enough noise to the signals such that all the signals were below the noise. The software did continue to decode signals. All the reports were -24 dB. This was a very crude test
Re: Topband: FT8 - How it really works
Joe, thanks for the information. I am not exactly sure what all that means. My conclusions were based on observed data. It seems pretty obvious to me that a signal that is more than 50 dB above the noise floor should not receive a S/N number of -1 dB, which is what FT8 gives. I don't know how the information you provided can make a calculation like that. I judge that a signal reading S9+40 dB on the S meter should be more than 50 dB above the noise floor when I can tune of to a spot where there are no signals and the S meter reads about S2 or S3 in SSB mode or less than S1 in a narrow bandwidth. Is the definition of "noise floor" being changed for FT8? Jerry, K4SAV On 12/19/2018 7:27 PM, Joe Subich, W4TV wrote: On 2018-12-19 4:28 PM, K4SAV wrote: > The official documentation for FT8 says it will decode signals 24 dB > below the noise floor. That is not a correct statement most of the > time. No, that is a correct statement. Signal reports in WSJT-X for FT8, JT65 and JT9 are *all* measured *with regard to the noise in 2500 Hz*. Note that the tone filters in WSJT-X are on the order of less than 12 Hz or so wide so the SNR *for an individual tone in the DSP filter bandwidth* at 0 dB is -23 dB relative to the *total noise in 2500 Hz bandwidth*. The actual filter bandwidth will change from mode to mode due to the differences in keying rated and tone spacing ... the actual SNR limit is shown in section 17.2.7 of the WSJT_X 2.0 User Guide. CW operators understand this from experience ... a quality 200 Hz filter will have ~12 dB less noise than a 2800 Hz filter. Thus a CW signal with a 200 Hz filter will have 12 dB better SNR than the same CW signal with a 2800 Hz filter (excluding any "processing gain" from the ear- brain filter). With FT8, JT65, JT9, etc. coding (forward error correction) provides some additional SNR (called "coding gain") but the *measurement* is based on strength of the individual tone to total noise. Thus, the lowest accurate report is -24 dB although some signals will be decoded at levels below that. 73, ... Joe, W4TV On 2018-12-19 4:28 PM, K4SAV wrote: While sitting around being bored and recovering from a gall bladder operation, I decided to do some experiments with FT8. First thing I did was upgrade the software to WSJT-X v2.0. I hope this post doesn't turn into another FT8 bashing session. My only goal was to understand how this mode works and what it can do and what it cannot do. The official description of FT8's signal reporting cannot be correct. It is obviously not a signal to noise number and it is not an S meter reading. What is it? That was the first question to answer. It's obviously not an S/N number because how do you give a report of -1 dB for a signal that is S9+40 dB on a quiet band. I was unable to find any info on how the signal report was calculated so I tried to correlate those reports to observations. I think I have figured out a method that results in very close to the same number that FT8 reports. Here is the experiment. I set up my main VFO to USB 2500 Hz bandwidth and set the second VFO to CW at about 150 Hz bandwidth. I look for a station calling CQ and tune the second VFO to him and measure his signal strength. I also look at the S meter for the signal level on the main VFO. I also look at the signal report calculated by the software. For stations calling CQ that report is calculated by the software in my computer. The FT8 report is usually very close to the difference in signal levels (VFO1 - VFO2). For example if the main VFO reads S9+10 and the second VFO reads S9, the FT8 number will be -10 dB. Note that the FT8 says that -24 dB is the lowest it can decode. With VFO1 = S9+10, that's about S7 for the smallest signal it can decode. Observations agree. Those numbers will vary a little depending on how your S meter is calibrated. In order to decode a weak signal, all those close USA stations will have to go silent. The official documentation for FT8 says it will decode signals 24 dB below the noise floor. That is not a correct statement most of the time. That statement should be that FT8 will decode signals 24 dB below the sum total of everything in a 2500 Hz bandwidth. If the total of all signals on the band are below the noise floor, it would be interesting to know if FT8 will decode any of them. I haven't observed that yet in a real situation. I did however try to simulate that condition by adding enough noise to the signals such that all the signals were below the noise. The software did continue to decode signals. All the reports were -24 dB. This was a very crude test because I don't know how exactly much the signals were below the noise. This should be of benefit to those people that have S9+ noise on the bands they operate. They should be able to decode the strongest signals on the band. The (VFO1 - VFO2) test just described should always result in a
Re: Topband: FT8 - How it really works
On 2018-12-19 4:28 PM, K4SAV wrote: > The official documentation for FT8 says it will decode signals 24 dB > below the noise floor. That is not a correct statement most of the > time. No, that is a correct statement. Signal reports in WSJT-X for FT8, JT65 and JT9 are *all* measured *with regard to the noise in 2500 Hz*. Note that the tone filters in WSJT-X are on the order of less than 12 Hz or so wide so the SNR *for an individual tone in the DSP filter bandwidth* at 0 dB is -23 dB relative to the *total noise in 2500 Hz bandwidth*. The actual filter bandwidth will change from mode to mode due to the differences in keying rated and tone spacing ... the actual SNR limit is shown in section 17.2.7 of the WSJT_X 2.0 User Guide. CW operators understand this from experience ... a quality 200 Hz filter will have ~12 dB less noise than a 2800 Hz filter. Thus a CW signal with a 200 Hz filter will have 12 dB better SNR than the same CW signal with a 2800 Hz filter (excluding any "processing gain" from the ear- brain filter). With FT8, JT65, JT9, etc. coding (forward error correction) provides some additional SNR (called "coding gain") but the *measurement* is based on strength of the individual tone to total noise. Thus, the lowest accurate report is -24 dB although some signals will be decoded at levels below that. 73, ... Joe, W4TV On 2018-12-19 4:28 PM, K4SAV wrote: While sitting around being bored and recovering from a gall bladder operation, I decided to do some experiments with FT8. First thing I did was upgrade the software to WSJT-X v2.0. I hope this post doesn't turn into another FT8 bashing session. My only goal was to understand how this mode works and what it can do and what it cannot do. The official description of FT8's signal reporting cannot be correct. It is obviously not a signal to noise number and it is not an S meter reading. What is it? That was the first question to answer. It's obviously not an S/N number because how do you give a report of -1 dB for a signal that is S9+40 dB on a quiet band. I was unable to find any info on how the signal report was calculated so I tried to correlate those reports to observations. I think I have figured out a method that results in very close to the same number that FT8 reports. Here is the experiment. I set up my main VFO to USB 2500 Hz bandwidth and set the second VFO to CW at about 150 Hz bandwidth. I look for a station calling CQ and tune the second VFO to him and measure his signal strength. I also look at the S meter for the signal level on the main VFO. I also look at the signal report calculated by the software. For stations calling CQ that report is calculated by the software in my computer. The FT8 report is usually very close to the difference in signal levels (VFO1 - VFO2). For example if the main VFO reads S9+10 and the second VFO reads S9, the FT8 number will be -10 dB. Note that the FT8 says that -24 dB is the lowest it can decode. With VFO1 = S9+10, that's about S7 for the smallest signal it can decode. Observations agree. Those numbers will vary a little depending on how your S meter is calibrated. In order to decode a weak signal, all those close USA stations will have to go silent. The official documentation for FT8 says it will decode signals 24 dB below the noise floor. That is not a correct statement most of the time. That statement should be that FT8 will decode signals 24 dB below the sum total of everything in a 2500 Hz bandwidth. If the total of all signals on the band are below the noise floor, it would be interesting to know if FT8 will decode any of them. I haven't observed that yet in a real situation. I did however try to simulate that condition by adding enough noise to the signals such that all the signals were below the noise. The software did continue to decode signals. All the reports were -24 dB. This was a very crude test because I don't know how exactly much the signals were below the noise. This should be of benefit to those people that have S9+ noise on the bands they operate. They should be able to decode the strongest signals on the band. The (VFO1 - VFO2) test just described should always result in a number equal to or less than zero. I notice sometimes the software will report a small positive number. That seems to happen more often when the bandwidth is set to something less than 2500 Hz and there are very few signals on the band. I think this may be related to the fact that FT8 does all its calculations using audio signals and the receiver S meter is operating on RF. Audio shaping in the receiver will affect the FT8 calculations. Audio processing in your computer sound card may be a factor too. This becomes really apparent when the radio is set to CW and the audio peaking filter is turned on. With SSB bandwidth and flat audio response, S meter readings are a good indication of what will be decoded. It
Topband: FT8 - How it really works
While sitting around being bored and recovering from a gall bladder operation, I decided to do some experiments with FT8. First thing I did was upgrade the software to WSJT-X v2.0. I hope this post doesn't turn into another FT8 bashing session. My only goal was to understand how this mode works and what it can do and what it cannot do. The official description of FT8's signal reporting cannot be correct. It is obviously not a signal to noise number and it is not an S meter reading. What is it? That was the first question to answer. It's obviously not an S/N number because how do you give a report of -1 dB for a signal that is S9+40 dB on a quiet band. I was unable to find any info on how the signal report was calculated so I tried to correlate those reports to observations. I think I have figured out a method that results in very close to the same number that FT8 reports. Here is the experiment. I set up my main VFO to USB 2500 Hz bandwidth and set the second VFO to CW at about 150 Hz bandwidth. I look for a station calling CQ and tune the second VFO to him and measure his signal strength. I also look at the S meter for the signal level on the main VFO. I also look at the signal report calculated by the software. For stations calling CQ that report is calculated by the software in my computer. The FT8 report is usually very close to the difference in signal levels (VFO1 - VFO2). For example if the main VFO reads S9+10 and the second VFO reads S9, the FT8 number will be -10 dB. Note that the FT8 says that -24 dB is the lowest it can decode. With VFO1 = S9+10, that's about S7 for the smallest signal it can decode. Observations agree. Those numbers will vary a little depending on how your S meter is calibrated. In order to decode a weak signal, all those close USA stations will have to go silent. The official documentation for FT8 says it will decode signals 24 dB below the noise floor. That is not a correct statement most of the time. That statement should be that FT8 will decode signals 24 dB below the sum total of everything in a 2500 Hz bandwidth. If the total of all signals on the band are below the noise floor, it would be interesting to know if FT8 will decode any of them. I haven't observed that yet in a real situation. I did however try to simulate that condition by adding enough noise to the signals such that all the signals were below the noise. The software did continue to decode signals. All the reports were -24 dB. This was a very crude test because I don't know how exactly much the signals were below the noise. This should be of benefit to those people that have S9+ noise on the bands they operate. They should be able to decode the strongest signals on the band. The (VFO1 - VFO2) test just described should always result in a number equal to or less than zero. I notice sometimes the software will report a small positive number. That seems to happen more often when the bandwidth is set to something less than 2500 Hz and there are very few signals on the band. I think this may be related to the fact that FT8 does all its calculations using audio signals and the receiver S meter is operating on RF. Audio shaping in the receiver will affect the FT8 calculations. Audio processing in your computer sound card may be a factor too. This becomes really apparent when the radio is set to CW and the audio peaking filter is turned on. With SSB bandwidth and flat audio response, S meter readings are a good indication of what will be decoded. It should decode signals down to 24 dB below whatever your S meter reads. I also narrowed the bandwidth of VFO1 and chopped out a bunch of signals. I got S7 on VFO1. Then a station calling CQ also measured S7 on VFO2. The FT8 report was 0 dB. Agrees.` That test brings up a possibility. If you can narrow VFO1 to a very narrow bandwidth hopefully containing only a very weak signal, then you may be able to decode it. A strong signal in the passband of VFO1 will kill the decode. It works. I decreased the bandwidth of VFO1 to 200 Hz and it decoded an S2 signal. I had VFO1 in USB mode with that bandwidth. My receiver will go to zero bandwidth in USB mode. I put VFO1 into CW mode at 100 Hz bandwidth and it decoded a signal that was moving the meter between S0 and S1. That signal would have also been easy copy if it was CW instead of FT8. I was using a good receiving antenna on 160 meters immediately after sunset. While this seems to work for weak signals it is a non-starter for normal operation. How do you tune around with a very narrow bandwidth looking for a station calling CQ or any other station that might be DX? It's not like CW, unless you learn to copy FT8 by ear. You can't find him with a wide bandwidth because the software won't decode him. He is only there when the bandwidth is very narrow. Given the number of USA stations on FT8 that
