5717RE: [wsjtgroup] FSK441 vs JT6M -- when to use
- Feb 2, 2009
Thanks Bill, this is another great post and very informative for newcomers and old-timers alike.
Jim W6OUU DN22sn
From: email@example.com [mailto:firstname.lastname@example.org] On Behalf Of Bill W5WVO
Sent: Sunday, February 01, 2009 9:41 PM
Subject: [wsjtgroup] FSK441 vs JT6M -- when to use
I hope nobody is upset with my posting these little tutorials for HSMS/WSJT newbies. If anyone thinks this is an inappropriate use of this list, please email the list owner and let him know. On the other hand, if you just don't find it useful or interesting, please simply delete. Thanks.
------------ --------- --
Many, many WSJT newbies have asked me what the difference is between FSK441 and JT6M, and when each mode should be used in preference to the other.
First, some history and a little technical explanation.
FSK441 was developed as a mode expressly optimized for meteor-scatter propagation. It is capable of decoding an entire message in considerably less than 1 second. FSK441, like all WSJT modes, is AFSK, or audio frequency-shift keying. Your sound card creates four audio tones in various sequences to encode the alphabet, numerals, and a few special characters.
FSK441 is not a "smart" protocol, meaning that it does not analyze data patterns to determine message contents (as does JT65), nor does it create "most likely" message contents by averaging multiple decodes of the message string (as do both JT65 and JT6M). What FSK441 decodes, character by character, is what you see in the decode window. It is a very MANUAL mode! And to get the best decoding performance out of it, more operator smarts are arguably required than for any of the other WSJT modes. (Advanced use of the FSK441 tools in WSJT is beyond the scope of this short article, but such a paper should be written.)
JT6M came along later. It is not a "dumb" protocol like FSK441 because it is capable of making some intelligent guesses about the content of the message being decoded. For this reason, and because it is a slower protocol and therefore requires less bandwidth than FSK441, JT6M is a more sensitive mode. On average, you can decode a more-or-less steady strength JT6M signal at better than 10 dB below the noise floor. In comparison, FSK441 requires signals at least 1 or 2 dB above the noise floor.
For this reason, a lot of hams jump to the conclusion that JT6M must be better for long-distance meteor-scatter contacts because it is "more sensitive". This conclusion is incorrect, and we'll discuss the reasons here.
As anyone knows who has had any experience with meteor-scatter work, meteor-scatter radio propagation events (variously called pings, burns, or bursts) come in all strengths and sizes, from the weak 1-dB, 100-millisecond (ms) ping up to the 20-second long, speaker-rattling "blue whizzer" caused by a bigger than usual meteor hitting the atmosphere in exactly the right place. Now, here's the technical part:
The amount of ionization created by a meteor burning up (called the plasma density) is determined by a number of variables, including the meteor's composition, its mass, its speed, and its angle of entry. When a very high plasma density is achieved momentarily, the effective Maximum Usable Frequency (MUF) for that meteor trail is quite high, sometimes as high as 432 MHz or more. However, since this extremely high plasma density occurs only in a small portion of the meteor's ion trail and begins to weaken quickly, pings in the high VHF and UHF ranges are infrequent and quite narrow, usually less than 1 second duration. The higher in frequency you go, the more narrow and the more infrequent meteor pings are.
FSK441 was designed to decode meteor pings as brief as 150 ms duration. Therefore, under normal circumstances on 2 meters and above, FSK441 is really the only mode that will reliably provide meteor-scatter communication.
On 6 meters, however, propagation can be sustained with a lower plasma density, and meteor pings tend, on average, to be longer in duration, often lasting more than a second or two. JT6M requires a burn of at least a second, and more reliably two seconds, in order to encode/decode a worst-case length message (both calls plus reports). These longer pings allow the more sensitive JT6M mode to be usable for some meteor-scatter work on 6 meters.
Now, there is a catch. Here is the nexus of the problem:
Very weak pings that might seem to make the higher JT6M sensitivity advantageous also tend to be very short. Therefore, even if a 200-ms ping several dB below the noise floor could theoretically be decoded by JT6M (and could not be decoded by FSK441), the ping is oftentimes too short to contain a reliably decodable JT6M message. So your extra sensitivity is of little use! Pings that are long enough to contain full JT6M messages are also typically more than strong enough to be decodable by FSK441. And FSK441 will also decode comparatively weaker messages only a few hundreds of milliseconds long.
Therefore, in most cases, FSK441 is still a better performer in 6-meter meteor-scatter work than JT6M, though plenty of meteor-scatter QSOs can be made and are made every day using JT6M. It's just a matter of the percentages.
If conditions are strong and the station you are trying to work is in the "sweet spot" for meteor-scatter propagation (500-800 miles), JT6M will virtually always work just fine. If, however, you are: (1) working under poor MS conditions, or (2) you are trying to work a station at the edge of MS range, or (3) you are trying to work a station very close to you via high-angle, high-MUF meteors, it is better to use FSK441. Why? Because all three of these conditions tend to produce pings that are comparatively shorter in duration, and FSK441 simply works better for short pings.
JT6M, however, has a wonderful redeeming value. It is an outstanding mode for very weak signal work when signal strengths are either fairly stable or variable over a period of at least several seconds. Such signals can come from tropospheric, D-layer ionoscatter, and weak sporadic-E propagation. When F2 propagation begins to return to 6 meters in a few more years, we may find that JT6M is an outstanding mode for completing 6-meter QSOs when propagation is still too weak to support SSB or even CW communication.
In general, then:
(1) On bands above 6 meters, use FSK441 for meteor-scatter QSO attempts. JT6M will generally not be useful on 2 meters and above.
(2) On 6 meters, use FSK441 as your default mode for meteor-scatter QSO attempts. Use JT6M only when meteor burns are fairly long -- at least a full second. More often than not, these conditions will also produce burns that are fairly strong as well, again on average.
(3) Use JT6M for weak-signal QSO attempts using atmospheric propagation modes like tropo, ionoscatter, and sporadic-E.
These are not laws, merely suggestions based on observation and a little science. Exceptions to these generalizations do occur regularly on 6 meters.
- << Previous post in topic