00210d1141
Minor code cleanup in subtractft9.f90.
2020-02-26 13:14:49 -05:00
65fda32a05
Previous commit was in error. This is the best-performing subtractft8.f90.
2020-02-25 09:04:18 -05:00
ff46c5a0c4
In subtractft8.f90: refine DT for early decodes before subtracting them from dd().
2020-02-24 16:18:42 -05:00
adb7dfcac3
Satisfy fussy compilers like gcc 10: Type mismatch between actual argument at (1) and actual argument at (2)
2020-02-21 09:52:20 -05:00
ac1d0f837d
Comment out diagnostic write statement.
2020-02-17 10:56:47 -06:00
20b942b802
FT8: Add an experimental 4'th bit metric. Make dupe removal unconditional.
2020-02-17 10:53:52 -06:00
207fd05f9f
Minor simplification - remove unnecessary nint() functions in get_spectrum_baseline.f90
2020-02-14 16:54:56 -06:00
09b47b0e8e
Change lower limit of analysis window from 200 Hz to 100 Hz.
2020-02-14 15:28:16 -06:00
4a4864d2ae
FT8 early decoding is now working.
2020-02-13 15:05:37 -05:00
ff109a60ca
Implement AP decoding for Hounds when Fox has a compound call with standard base call.
2020-02-03 10:48:51 -06:00
70b84478f0
Add WW_DIGI to the special activities section. Make necessary changes to
...
the decoders. Also, an experimental change to the FT4 decoder to base
AP decoding passes on 4-symbol block detection instead of single symbol
detection. This provides about 1 dB improvement on the AWGN channel.
Sensitivity changes on other channels are TBD.
2020-01-24 11:26:36 -06:00
f069477dbb
Bring test28.f90, ft4_testmsg.f90, ft8_testmsg.f90, ft8code.f90 up to date.
2020-01-03 13:34:56 -05:00
3397bf49b4
Add utility program ft4code.
2019-12-18 10:02:00 -05:00
a0d471cb0b
Improve FT8 SNR estimates in two ways:
...
(i) SNR no longer saturates at +20 dB
(ii) a large signal in the passband no longer causes the
SNR of weaker signals to be biased low.
2019-11-27 15:58:52 -06:00
091b7f03ed
Delete un-needed file.
2019-11-27 15:57:40 -06:00
d1530b0c19
Rename ft8sim files as follows:
...
ft8sim_gfsk.f90 => ft8sim.f90
ft8sim.f90 => ft8sim_fsk.f90
This change reflects the fact that the GFSK waveform is now used in WSJT-X
for ft8 and ft4.
2019-11-27 09:08:17 -06:00
f224790569
Tweak for consistency.
2019-07-12 10:41:09 -05:00
396bf1a013
Correct a flaw in ft8sim when generating data with large abs(DT).
2019-06-17 11:26:46 -04:00
3611be8ba0
Try to avoid an array bounds error that has been reported by a user. Cause unknown.
2019-05-03 10:51:45 -05:00
cf2219173a
Squash a SEGV
2019-04-20 01:14:23 +01:00
e8d17a9898
FT4: Subtraction is basically working.
2019-04-18 14:16:39 -05:00
be72461142
FT4: Some groundwork for subtraction.
2019-04-18 14:15:24 -05:00
f296e37630
Fix a bug that didn't show up when I compiled on the Mac.
2019-04-12 18:54:33 -05:00
cd8ea16f40
FT8: Use GFSK waveform with BT=4.0 as reference for subtraction, for now. May want to change to BT=2.0 before v2.1 release?
2019-04-12 17:11:26 -05:00
278c544a6a
Merge branch 'develop' into feat-ft2
2019-02-26 11:52:14 +00:00
3759198982
Remove ramp-up and ramp-down symbols. Instead, shape the first 1/8th of first and last real symbols.
2019-02-25 15:03:43 -06:00
ec2d5b6233
Fix the sign of one of the Nuttal window coefficients. Change BT to 2.0.
2019-02-25 12:57:34 -06:00
17ac99344c
For now, change to BT=1.5 for FT8. This commit enables the waterfall analysis window. This decreases sidelobes significantly in exchange for increased noise bandwidth. Sensitivity to weak tones will be degraded somewhat.
2019-02-23 14:18:22 -06:00
5d914a1eb1
Make ft8sim_gfsk work with recent changes to gen_ft8wave.f90.
2019-02-22 15:16:29 -06:00
25bfa3798a
Transmit FT8 using a gaussian-filtered waveform.
2019-02-22 13:39:39 -05:00
d9c99601ae
Fix bugs in gen_ft8wave.f90.
2019-02-21 15:56:21 -06:00
1acb5d5d99
Add routines needed to test FT8 with GFSK pulses.
2019-02-21 15:48:02 -06:00
7d66f442d6
Fix up some compiler errors when using gfortran v9
2019-02-10 20:43:51 +00:00
c52f755afc
Make Watterson simulator causal.
2019-01-25 16:26:13 -06:00
aee9c09851
Merge branch 'develop' into feat-ft2
2019-01-16 20:44:44 +00:00
8f1baae048
If delay=0.0 make Watterson use only a single component to simulate flat fading.
2019-01-13 09:25:30 -06:00
c4d078f55f
Remove unnecessary character string space-padding.
2019-01-03 14:02:17 -06:00
1a8ddad307
In ft8apset, mycall and dxcall must be at least 3 characters long, otherwise they are treated as non-existent.
2018-12-24 11:19:07 -06:00
b44d7b472d
Fix a typo.
2018-12-24 10:49:40 -06:00
318ce0020e
Try harder to ensure that we don't waste time trying to do ap decoding that would involve nonstandard calls. Current code does not support ap decoding with non-standard calls.
2018-12-24 10:45:02 -06:00
5034e21291
Remove some diagnostic writes.
2018-12-21 09:51:39 -06:00
7a40bf2e4b
Use separate hash tables for 10, 12, and 22 bit hashes. Make unpacking i3=4 messages depend on whether unpack is being done on a received message, or a to-be-transmitted message. Give mycall13 priority over hash table entries in certain contexts.
2018-12-20 15:08:29 -06:00
307948a25d
Remove an unused array.
2018-12-16 11:55:36 -06:00
dfceef3662
Fix up ft8sim so that it prints a warning when SNR is set too high so that data will be clipped.
2018-12-07 11:10:25 -06:00
337bae6d0d
Housekeeping. Remove unused code and unused variables. Silence some compiler warnings.
2018-12-05 16:06:33 -06:00
a5e82ac94c
Delete an unused routine. Fix a typo in a comment.
2018-12-04 20:01:55 -06:00
6fc27d550d
Fix an egregious bug and another not-so-egregious bug. The egregious bug prevented decodes of most signals with negative DT.
2018-12-03 07:18:52 -06:00
df37d9b1f7
Remove cqhund stuff and some other unused code.
2018-11-30 13:46:54 -06:00
461b87f047
Minor improvement in code that checks for bad FT8 messages.
2018-11-27 15:34:01 -05:00
9e74125799
Don't allow an illegal message to crash routine genft8().
2018-11-27 15:26:24 -05:00
Joe Taylor
Steven Franke
Bill Somerville