FT8 Using Real FSK

Discussion in 'General Technical Questions and Answers' started by VK4TS, Jul 12, 2019.

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: Left-2
ad: abrind-2
ad: Subscribe
ad: L-MFJ
  1. VK4TS

    VK4TS Ham Member QRZ Page

    de VK4TS

    Correct me if I am wrong - a question for the gurus..

    RTTY is generated by two main methods Audio Shift Keying AFSK and Frequency Shift Keying FSK - the difference is remarkable.

    Using AFSK using the SSB audio path setting of audio levels are critical and over driving is a problem. Using FSK is basically the same as CW there is a carrier and the radio moves the frequency between the two points.

    Because it is CW there is no audio setting issues, rigs run cooler and power can be increased.

    FT8 is a series of audio tones that are driven down the audio path and susceptible to over driving. Could FT8 be adopted to a true frequency shift keying format that keys the rig without using the AFSK audio path, varying the frequency via the serial port thus allowing a better interface, greater power and better efficiency?
  2. KA0HCP

    KA0HCP XML Subscriber QRZ Page

    Sure, if you build a tone generator for each tone and devise the interfaces between the device and radio and device and computer. Simple.
  3. K8JD

    K8JD Ham Member QRZ Page

    Isn't FT8 just a variety of high speed phase shift keying ?
  4. KA0HCP

    KA0HCP XML Subscriber QRZ Page

    Summary of FT8
    • FT8 is named after its developers, Steven Franke, K9AN, and Joe Taylor, K1JT.
    • The “8” denotes mode's 8-frequency shift keying format.
    • Tones are spaced at 6.25 Hz, and an FT8 signal occupies just 50 Hz

    Introduction to FT8 by W0WTN
    http://www.w0wtn.org/downloads/n0dl/Introduction to Ham Radio Digital Mode FT8.pdf
    W9KEY likes this.
  5. W9KEY

    W9KEY XML Subscriber QRZ Page

    What I find interesting - in spite of having over 24,000 lookups, Dr. Franke, K9AN does not have an active page on QRZ.com. But you can learn a bit more about him here: https://ece.illinois.edu/directory/profile/s-franke
  6. VA3VF

    VA3VF Ham Member QRZ Page

    KA0HCP likes this.
  7. KM1H

    KM1H Ham Member QRZ Page

    With his creds he doesnt need a page!! And his antenna

    And for K1JT who packs an awful lot of antennas in a small lot on a crankup!

    It is those that bluster on here and elsewhere over and over about things they know little about in a peer reviewed forum and provide no backround bio or even a thread to establish some level of credibility. Those deserve no respect.

  8. G0GSR

    G0GSR Ham Member QRZ Page

    In reality, there would be no difference between the two systems. Same with AFSK and FSK. But this assumes that the internal processing was as good as the "JT" software.
    "Digitally generated tones" and "digitised audio tones" have exactly the same characteristics when converted back to an analogue RF signal, then amplified.
    The "JT" software generates very low distortion tones with phase coherent shifts so there is only a very slight degradation within the first D to A converter.
    All the user needs to do is to avoid driving this into digital clipping which is very easy.

    Last edited: Jul 13, 2019
  9. G0GSR

    G0GSR Ham Member QRZ Page

    Sorry too late to edit. "First A to D converter".

  10. SM0AOM

    SM0AOM Ham Member QRZ Page

    FT8 and its "siblings" use a constant envelope emission, which makes them quite immune to distorsion in the signal path.

    MFSK could be generated with whatever signal generation scheme which is convenient.

    If it is generated directly without mixing from an audio passband it becomes emission F1D, when a baseband is used it is emission J2D, which both have identical characteristics at the transmitter output.

    Predecessors like Piccolo and Coquelet that were used for military and diplomatic communications as early as the 60s used somewhat shaped signal envelopes to minimise signal bandwidth, and they could be transmitted even through the mediocre transmitters of yesteryear.

    Before that, a quite common signalling protocol on commercial radiotelegraph circuits was "four-frequency diplex" where two synchronous data streams were multiplexed in one transmitter using four shift frequencies.

    The most common transmitter implementation in the pre-SSB era was to use two crystal oscillators that were "pulled" by reactance tubes which then were multiplied to the appropriate output frequencies.

    For reasonable amounts of distorsion in the SSB transmitter signal paths, their influence on FT8 signal quality may be safely neglected.


Share This Page