Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63)
Skip, MT63-1000 has a -5 dB minimum S/N, but MFSK16 has a -13.5 dB minimum S/N, so the static tests you made must be at signal levels high enough that MT63-1000 decodes, which may not be a realistic level. That is true. Fortunately, there are times when signals are above the decode threshold for the majority of modes. That gives us the chance to test the higher throughput modes to see what works in heavy static. MFSK16 turned out (after three months of testing) to be the most static-resistant mode of all That is interesting Skip. It did seem to do slightly better than THOR22 during n simulated tests. Did you see any advantage in throughput with MT63 during the static crash tests when signals were adequate? Tony -K2MO
Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63)
We did not test MT63, because only MT63-2000 could work with flarq and ARQ, and we think it would be irresponsible to use that on the shared ham bands for the little benefit it would bring compared to much more narrow modes. It is OK to use on MARS, because each MARS frequency channel is dedicated, not shared (well, time-shared by different nets, and the channels are voice-bandwidth as they are also used interchangebly with voice. My experience with MT63-1000 on MARS is that it works very well under QRM and static, as expected, but that is with S5-S9 signals in the South Carolina - Florida corridor, and weaker stations often report negative copy, probably because the S/N is not good enough at their locations. Will find out more about the MT63-1000 real-world static resistance as summertime approaches. 73, Skip KH6TY http://kh6ty.home.comcast.net - Original Message - From: Tony To: digitalradio@yahoogroups.com Sent: Sunday, March 22, 2009 3:03 AM Subject: Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63) Skip, MT63-1000 has a -5 dB minimum S/N, but MFSK16 has a -13.5 dB minimum S/N, so the static tests you made must be at signal levels high enough that MT63-1000 decodes, which may not be a realistic level. That is true. Fortunately, there are times when signals are above the decode threshold for the majority of modes. That gives us the chance to test the higher throughput modes to see what works in heavy static. MFSK16 turned out (after three months of testing) to be the most static-resistant mode of all That is interesting Skip. It did seem to do slightly better than THOR22 during n simulated tests. Did you see any advantage in throughput with MT63 during the static crash tests when signals were adequate? Tony -K2MO
RE: [digitalradio] THOR is static-proof (Re: KV9U - MT63)
I would like to remind all, if you are not already aware, to turn AGC off when static crashes are an issue. If you are fortunate enough to operate in a mixed mode net, turn it to fast, or for inland stations, medium. Slow recovery time of the rig in response to a strong signal cannot be corrected by a sound card protocol; no matter how robust. While we are at it, when using MT-63 at 1K long, keep in mind that most software hard codes a starting frequency of 500 Hz, and that is a 1.5Khz total width. It doesn't work well if you have your filters set for PSK, or a narrow-band mode. In running digital training nets for newcomers to MT-63, it is absolutely amazing how many ways can be found to lessen it's effectiveness; primarily due to not understanding where the signal is, where it is going, and how it is getting there. It took me a long time to factor out many of the common reasons it didn't work. That is one of the main reasons that PSK-31 is so popular; even a caveman can do it. (Sorry Geico; couldn't resist) David KD4NUE -Original Message- From: digitalradio@yahoogroups.com [mailto:digitalra...@yahoogroups.com] On Behalf Of Tony Sent: Sunday, March 22, 2009 3:04 AM To: digitalradio@yahoogroups.com Subject: Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63) Skip, MT63-1000 has a -5 dB minimum S/N, but MFSK16 has a -13.5 dB minimum S/N, so the static tests you made must be at signal levels high enough that MT63-1000 decodes, which may not be a realistic level. That is true. Fortunately, there are times when signals are above the decode threshold for the majority of modes. That gives us the chance to test the higher throughput modes to see what works in heavy static. MFSK16 turned out (after three months of testing) to be the most static-resistant mode of all That is interesting Skip. It did seem to do slightly better than THOR22 during n simulated tests. Did you see any advantage in throughput with MT63 during the static crash tests when signals were adequate? Tony -K2MO
Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63)
David, I would like to remind all, if you are not already aware, to turn AGC off when static crashes are an issue. Good advise. A fast AGC setting may help as well if there's no way to turn it off. Tony -K2MO - Original Message - From: David Little dalit...@bellsouth.net To: digitalradio@yahoogroups.com Sent: Sunday, March 22, 2009 9:58 AM Subject: RE: [digitalradio] THOR is static-proof (Re: KV9U - MT63) I would like to remind all, if you are not already aware, to turn AGC off when static crashes are an issue. If you are fortunate enough to operate in a mixed mode net, turn it to fast, or for inland stations, medium. Slow recovery time of the rig in response to a strong signal cannot be corrected by a sound card protocol; no matter how robust. While we are at it, when using MT-63 at 1K long, keep in mind that most software hard codes a starting frequency of 500 Hz, and that is a 1.5Khz total width. It doesn't work well if you have your filters set for PSK, or a narrow-band mode. In running digital training nets for newcomers to MT-63, it is absolutely amazing how many ways can be found to lessen it's effectiveness; primarily due to not understanding where the signal is, where it is going, and how it is getting there. It took me a long time to factor out many of the common reasons it didn't work. That is one of the main reasons that PSK-31 is so popular; even a caveman can do it. (Sorry Geico; couldn't resist) David KD4NUE -Original Message- From: digitalradio@yahoogroups.com [mailto:digitalra...@yahoogroups.com] On Behalf Of Tony Sent: Sunday, March 22, 2009 3:04 AM To: digitalradio@yahoogroups.com Subject: Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63) Skip, MT63-1000 has a -5 dB minimum S/N, but MFSK16 has a -13.5 dB minimum S/N, so the static tests you made must be at signal levels high enough that MT63-1000 decodes, which may not be a realistic level. That is true. Fortunately, there are times when signals are above the decode threshold for the majority of modes. That gives us the chance to test the higher throughput modes to see what works in heavy static. MFSK16 turned out (after three months of testing) to be the most static-resistant mode of all That is interesting Skip. It did seem to do slightly better than THOR22 during n simulated tests. Did you see any advantage in throughput with MT63 during the static crash tests when signals were adequate? Tony -K2MO
Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63)
Tony wrote: The most impressive thing about MT63 is how it seems to resist heavy static crashes. I made a few recordings with short segments of the signal removed to simulate this type of QRN and there was little effect on copy. What about THOR? Thor stated to be more static-proof. Jaak es1hj/qrp -- Kirjutas ja tervitab Jaak Hohensee
Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63)
Jaak, What about THOR? Thor stated to be more static-proof. It depends which THOR mode is used. It seems THOR-22 is the best of the bunch for static crash resistance. I've done a few static crash tests by generating noise at regular intervals; the noise obliterates the signal in short bursts. I would imagine this method would give some indication of on-air performance. I'm sure there are simulators out there that can produce more accurate results. The list of variables that would add to the mix are endless; ionospheric distortion, weak / strong signal performance, QRM etc. As the disclaimers say, your mileage may vary! See below... Tony -K2MO ___ Text Message: Quick Brown Fox Pangram Static Crash: Duration: 1 second Interval: Every 5 seconds THOR-11 µ9i$:neíICK olrsplnOX JUAnopco vsR THE l¶unknOG TËq ©E QUICK BRetqksˆX JUMPS«aa±n THE )txeTaTic DOG X erEÒtCK BROsbßnn”X JU 5¶R THE ¡t,a0ssY DOG TŒi R ta BROWN THOR-22 THE QUICK BRwnoacebnOX JUMPS OVER THE Lti ) tla ey tktzlQ HE QUICK BROWtzoh JUMPS OVER THE Lpc·¢fG THE QUICK BROWN L xth Ítl t1 JUMPS OVER THE LAZYk rNyp+THE QUICK $ MT63 1K Long Interleave THE QUICK BREWQUICK BROWN FOX JUMPS OVER THE LAZY DOG THERQUICK BROWN FOX JUMPS OVER THE LAZY DOG THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG MFSK16 THE QUICKl||½ OWN FOX JUMPS hqPeavHE LAZY DOG THEvaŽÊICK BROWNza«cpFOX JUMPS OVER Taetf ‡E LAZY DOG THE Qh tCK BROWN FOX JU3 ]S OVER THE LA¬cc tsa ÕOG ___ Text - Quick Brown Fox Pangram Static Crash: Duration: 2 seconds Interval: Every 5 seconds THOR-11 Tseor'Ka °ANROWN F7ueNpg r epitUX s 3àn MDBxhvuntF^yš THE õ ¾bSyK BROWN tq?yõP×7 eZ ²opHE L 8p!t es OGCK Ä A/pttªOX JUMPS OfdròSe THE LAZY Do trtn THOR-22 THE QUICK BuA qklt ¬ JUMPS OVER ta97tncx2td/RZY DOG THE QUIceË Ái daÖWN FOae t pQ R m ©t OVER THE elNtîi oMcsiG THE QUICK rLbu otiSoWN FOX MT63 1K Long Interleave THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG THE QUICK BROWN FOX JUMPS OVER THE AOY JOMPS OVEU THE LAZY DOG THE QUICKEBRAWN FOX THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG MFSK16 CK BROWN FOX JUMPS THE LAZY DOGqnæwbih THE QUbs up,‡CK BROWN FOEl„UMPS OVER THtY DOG G¨¨aId-E QUICK BROW)o tÌieEX JUMPS OVER gt - Original Message - From: Jaak Hohensee jaak.hohen...@eesti.ee To: digitalradio@yahoogroups.com Sent: Saturday, March 21, 2009 3:55 AM Subject: Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63) Tony wrote: The most impressive thing about MT63 is how it seems to resist heavy static crashes. I made a few recordings with short segments of the signal removed to simulate this type of QRN and there was little effect on copy. What about THOR? Thor stated to be more static-proof. Jaak es1hj/qrp -- Kirjutas ja tervitab Jaak Hohensee
Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63)
Tony, Static crash resistance is not the only parameter to consider. The problem is that you can have static and weak signals at the same time. MT63-1000 has a -5 dB minimum S/N, but MFSK16 has a -13.5 dB minimum S/N, so the static tests you made must be at signal levels high enough that MT63-1000 decodes, which may not be a realistic level. Last summer, during the lightning season in Florida, MFSK16 turned out (after three months of testing) to be the most static-resistant mode of all, even surpassing Thor, which we had worked on so hard to harden against static crashes. However, THOR is tolerant of mistuning, whereas MFSK16 is not, and MFSK16 needs AFC, which Thor does not, but overall, we concluded that MFSK16 was the best for NBEMS messaging on HF unless conditions (QSB and QRN) were such that a faster mode would work. Of course our tests were to find the best mode for messaging, which has to be a combination of reasonable speed and minimum S/N, and MT63-2000 is the only MT63 variant that is fast enough to overcome the extreme latency of MT63 and allow successful ARQ transfers without unreasonable wait times. MT63-1000 is not fast enough. The problem is that MT63-2000 is 3 dB worse on minimum S/N than MT63-1000, so the spread in minimum S/N between MT63-2000 and MFSK16 grows to about 11 dB, which is a LOT! As you point out, the list of variables is very long, and a mode for one situation may not work for another. As you observed during the MT63-1000 tests we made together, MFSK16 would print 80% when MT63-1000 would not print at all, and Olivia was printing 100% under roughly the same conditions. There is a resonably acceptable speed for message transfers, with and without ARQ (ARQ cuts the speed in about half), and a different reasonably acceptable speed for QSO's, just as JT65A is acceptable for short exchanges, but not so much for QSO's. So, for NBEMS, since the primary objective is messaging, on HF we found MFSK16 to be most suitable overall, but on VHF, where there is no static, for instance on 30m there is little static (where PSKmail operates), PSK250 can be used instead, when it is impossible to control the static crashes, or even noise, on the lower HF bands from capturing the AFC and shifting the tuning off frequency on HF, simply because you need to have AFC for PSK250, and between ARQ exchanges, there is no signal to lock on, so the AFC locks on a noise burst. Olivia would be great to use, but takes forever to get a message through, so the better minimum S/N of Olivia has to be sacrificed for greater speed in messaging and use MFSK16 instead, and let the ARQ just resend blocks when necessary. Of course, at some point, enough blocks may be damaged that the link simply fails or times out. Once you add ARQ to MFSK16, you have a speed of only about 20 wpm, which is very slow for anything than a very short message, but the ARQ guarantees error-free reception in return for the slow speed. Minimum S/N, QSB, QRN, doppler distortion, inter-symbol interference, tolerance to operator tuning, transceiver frequency stability, minimum necessary bandwidth, etc. etc., all figure into the decision as to which mode is best. No one shoe fits all, and we can only choose the best mode for our particular mission out of all the many available choices. 73, Skip KH6TY NBEMS Development Team - Original Message - From: Tony To: digitalradio@yahoogroups.com Sent: Saturday, March 21, 2009 5:05 PM Subject: Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63) Jaak, What about THOR? Thor stated to be more static-proof. It depends which THOR mode is used. It seems THOR-22 is the best of the bunch for static crash resistance. I've done a few static crash tests by generating noise at regular intervals; the noise obliterates the signal in short bursts. I would imagine this method would give some indication of on-air performance. I'm sure there are simulators out there that can produce more accurate results. The list of variables that would add to the mix are endless; ionospheric distortion, weak / strong signal performance, QRM etc. As the disclaimers say, your mileage may vary! See below... Tony -K2MO ___ Text Message: Quick Brown Fox Pangram Static Crash: Duration: 1 second Interval: Every 5 seconds THOR-11 µ9i$:neíICK olrsplnOX JUAnopco vsR THE l¶unknOG TËq ©E QUICK BRetqksˆX JUMPS«aa±n THE )txeTaTic DOG X erEÒtCK BROsbßnn”X JU 5¶R THE ¡t,a0ssY DOG TŒi R ta BROWN THOR-22 THE QUICK BRwnoacebnOX JUMPS OVER THE Lti ) tla ey tktzlQ HE QUICK BROWtzoh JUMPS OVER THE Lpc·¢fG THE QUICK BROWN L xth Ítl t1 JUMPS OVER THE LAZYk rNyp+THE QUICK $ MT63 1K Long Interleave THE QUICK BREWQUICK BROWN FOX JUMPS OVER THE LAZY DOG THERQUICK BROWN FOX JUMPS OVER THE LAZY DOG THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG MFSK16
Re: [digitalradio] THOR is static-proof (Re: KV9U - MT63)
Tony, Further complicating the static crash test conclusions is the effect of the static on the receiver AGC. If a long AGC constant is being used, the static crash is going to desensitize the receiver for as long as the AGC holds the receiver sensitivity above the decoding threshold. In such a case, the mode with the lower minimum S/N may recover sooner to the decoding threshold than the mode with the higher S/N. This may be why MFSK16 appears to beat out Thor (on the average). MFSK16 has both a low minimum S/N AND FEC, which appears to be a winning combination, especially as the band is starting to go out, as we experienced during our MT63-1000 trials (but without a lot of QRN, since we were on 20m). Depending upon the proximity of lightning strikes, and when signals are fairly strong, MT63-1000 may easily be the best mode - even better than Olivia - but there is ALWAYS some point that the last mode standing (probably the one with the lowest minimum S/N) is going to win when the band is going out. The idea behind using NVIS antennas for NBEMS on HF is that propagation is more constant, since there is less dependence on the skywave, and also that noise arrives at a lower angle than the NVIS cloud burner signal. This reduces the effect of the static crashes, but limits the distance on 80m and 40m to about 300 miles. 73, Skip KH6TY NBEMS Development Team
