Nick Foster
28366627c5
fix crap, use confidence
2011-06-11 21:40:51 -07:00
Nick Foster
8b1c1791c7
Everything changed.
...
* Rewrote preamble detector. Uses integrate-and-dump along with correlator to perform optimal receive.
* No longer using framer. Determining packet length via header. Missing some anti-FRUIT stuff.
* Pulled out the slicer logic into its own function.
Lots of todos and fixmes.
2011-06-11 21:32:28 -07:00
Nick Foster
c8dce82b75
Major cosmetic changes to preamble detector
2011-06-08 22:02:13 -07:00
Nick Foster
140f91537c
add subdev spec, also clean up cpr.py
2011-06-08 21:44:53 -07:00
Nick Foster
bca80f26ad
CPR cleanup
2011-06-01 20:37:51 -07:00
Nick Foster
9b3b296a22
Merge branch 'master' into nolocal
2011-06-01 20:14:59 -07:00
Nick Foster
ebd74a2587
fix bug in framer that causes the occasional segfault
2011-06-01 20:14:45 -07:00
Nick Foster
d814dc085f
disable local decoding. should probably be an option.
2011-06-01 16:36:42 -07:00
Nick Foster
e9cf22eb33
use eng float for rate so you can say rate=10e6
2011-06-01 16:36:00 -07:00
Nick Foster
ef5c0fbd53
cosmetics
2011-05-30 21:00:57 -07:00
Nick Foster
a1249b8c4d
pull numpy out of the int-and-dump generator
2011-05-30 20:54:24 -07:00
Nick Foster
52ebfcf4d4
rip out the energy stuff, int-and-dump filter takes care of it
2011-05-30 20:45:59 -07:00
Nick Foster
f86635430e
first stab at converting front end to use integrate & dump instead of cheesy "bit_energy" thing
2011-05-30 20:26:58 -07:00
Nick Foster
46573db399
Tags updated for sparse tags (which aren't actually in gr-uhd yet).
2011-05-30 19:57:47 -07:00
Nick Foster
773ddf8dec
Modified to work with proposed gr-uhd tags timestamp interface. Probably going to change.
2011-05-16 18:06:54 -07:00
Nick Foster
462b7fbed3
Merge branch 'speed'
...
Conflicts:
src/python/uhd_modes.py
2011-05-16 16:08:41 -07:00
Nick Foster
f0b913fe35
make parser invocations match new prototype
2011-05-06 10:39:43 +02:00
Nick Foster
756ccc8548
Merge in UHD change to master
2011-03-21 15:18:08 -07:00
Nick Foster
3341a62511
Updated to reflect newer UHD interface (multi-USRP)
2011-03-21 15:17:21 -07:00
Nick Foster
35ca3c8869
Modified mlat solver to solve to a threshold and quit. Also sorts incoming timestamps.
2010-12-06 10:28:44 -08:00
Nick Foster
1acea7c9fd
Added multilateration library mlat.py and test program mlat-test.py.
...
No integration into the application yet.
2010-12-05 17:43:13 -08:00
Nick Foster
6f62364051
fixed tune() call
2010-11-30 12:25:31 -08:00
Nick Foster
f69d8a2ecf
Derp
2010-11-28 17:12:02 -08:00
Nick Foster
b1b990ad49
Use single source rather than simple source.
2010-11-28 17:08:58 -08:00
Nick Foster
00bcc041b2
Cosmetics
2010-11-28 17:07:09 -08:00
Nick Foster
f8966f0d10
All tags, all the time. Still some weirdness with set_history() that requires using the check widths in the framer & slicer.
2010-11-20 17:06:48 -08:00
Nick Foster
cd394e1863
Preamble now passes tags to framer.
...
Note: Using set_history cocks up the nitems_written() call.
2010-11-20 16:34:44 -08:00
Nick Foster
6af11394d1
Removed gain print which bombs for file input
2010-11-20 15:09:23 -08:00
Nick Foster
d173f9e256
Preamble detector cleanup. Better for >2 samps/symbol.
2010-11-19 16:38:32 -08:00
Nick Foster
72248eaecf
Timestamps work.
2010-11-19 14:39:56 -08:00
Nick Foster
eb18bdc308
First stab at tagged timestamps using TR's tagging branch. Currently segfaults due to some oddness in the tag subsystem.
2010-11-18 14:25:07 -08:00
Nick Foster
d23530615f
Whoops. Had an SVN dir lurking in the config/ dir.
2010-11-01 12:24:16 -07:00
Nick Foster
4b3a225f7b
Various fixes for stability.
2010-10-24 19:00:10 -07:00
Nick Foster
a2aad97f5b
Reverting a fix to the loop -- misunderstood my own code! Un-broke the receive.
2010-10-24 16:13:44 -07:00
Nick Foster
dedf60fd4e
Exception bug fix in modes_parse
2010-10-24 08:55:37 -07:00
Nick Foster
4d93b8f821
Collected all the local position stuff so it's specified once in uhd_modes.py.
...
Should probably add in a command-line parameter to specify this.
2010-10-23 18:00:58 -07:00
Nick Foster
0d4194ef84
Fixed some indenting so outputs are only called when there's something to output. Hopefully also fixed that pesky NoneType error.
2010-10-23 16:51:09 -07:00
Nick Foster
839596a4b2
Derp. Forgot to git-add the KML parser. Also added an LPF to the main app. Some other cleanups in prep for release.
2010-10-23 16:05:08 -07:00
Nick Foster
01a0b635c6
Forgot the airplane icon.
2010-10-21 23:20:16 -07:00
Nick Foster
058b0b9af5
I should probably set up a .gitignore
2010-10-21 23:19:28 -07:00
Nick Foster
0bf5e47749
Various fixes for various exceptions.
...
You are seeing a couple of impossible altitudes and a couple of impossible longitudes. Looks like they're coming from the same aircraft. Possibly a nonstandard squitter? They pass checksum, but they don't contain the data you think they do.
2010-10-21 23:02:48 -07:00
Nick Foster
0edcc29380
KML at least works some. Need to pass DB filename into KML constructor.
2010-10-19 10:05:33 -07:00
Nick Foster
76f884e81f
Copyright assignment.
2010-10-18 09:59:08 -07:00
Nick Foster
ac0b796d1d
KML generation in there but not tested
2010-10-18 00:14:28 -07:00
Nick Foster
627b820b6e
Database inputs working.
2010-10-17 22:17:48 -07:00
Nick Foster
468cb726cb
Removed old USRP driver version. We are UHD-only now.
2010-10-17 19:47:09 -07:00
Nick Foster
33b04c292c
Reorganized things quite a bit. Much cleaner output interface.
...
Made a single registry of output plugins which gets populated on init with classes to output on a socket, print to screen, database, etc. Next step is to implement a SQLite database class.
2010-10-17 19:45:19 -07:00
Nick Foster
7727b7c902
Fixed some bugs where I was treating sample data as an integer. Allows use with UHD without multiplying everything up by 64K. =)
...
Also changed the reference level printout to show dB instead of sample value. This seems more useful. 0dB is full scale.
2010-09-20 11:09:59 -07:00
Nick Foster
13c5acca52
Added uhd_modes.py, a UHD port of the usrp_modes.py program. Basically the same but takes a rate instead of a decimation.
...
Right now there's a 1e6 gain block in there because old USRP spit out roughly 1e6x data as USRP2. Haven't investigated this for USRP vs USRP2 yet. Don't know why blocks care, actually.
2010-09-15 15:25:09 -07:00
Nick Foster
4a92b250ea
Compiles. Ready to go.
2010-09-14 22:10:57 -07:00
