David Protzman
94297ff21c
Added decode block
2022-09-22 21:15:13 -04:00
David Protzman
08cd68ab52
Added fft and filter libraries
2022-09-22 21:06:44 -04:00
David Protzman
ce69f6fc3c
Working on porting to 3.10
2022-09-22 21:03:39 -04:00
David Protzman
3bc45fe703
Added comments
2022-06-26 17:10:11 -04:00
David Protzman
39d419e318
No longer outputting NaN values for divide by zeros
2022-06-26 15:17:16 -04:00
David Protzman
a4c21fdf91
Now using the volk inverse sqrt function
2022-06-26 15:16:49 -04:00
David Protzman
df985ba877
Moved dot_prod to normalized_xcorr_estimate
...
The dot_prod module ended up being a full on normalized cross correlation module, so it made sense to replace the old normalized xcorr module
2022-06-26 10:24:05 -04:00
David Protzman
85f39daf0b
Added complex noise function
2022-06-24 01:07:12 -04:00
David Protzman
4cb98e4429
Removed the epsilon add
...
This was an attempt at getting the divide by zeros to stop, but if this didn't do it then it might be that the issue isn't a divide by zero. Will need to create a MATLAB test to find out what's actually happening
2022-06-24 01:06:49 -04:00
David Protzman
3335f6e3ea
Working much faster now
...
There are still some blanks in the o'scope plots which makes me think that there's either a divide by zero, or some gaps in the output values
2022-06-24 01:04:40 -04:00
David Protzman
574f1aaccd
Added dot product block
...
It's working, but very slowly at the moment
2022-06-24 00:21:36 -04:00
David Protzman
885262b97e
Moved the plotting logic out of the main test
2022-06-23 22:18:00 -04:00
David Protzman
957ad6934d
Added CMake config option for MATLAB install directory
2022-06-23 22:17:31 -04:00
David Protzman
ba5b1a2593
Added variance block
2022-06-23 21:42:57 -04:00
David Protzman
675fc65dfc
Added missing CMake file for the norm xcorr est block
2022-06-20 16:43:49 -04:00
David Protzman
1ba794cbee
Converted normalized_xcorr to use VOLK malloc'd arrays
...
This is to get better performance from the VOLK calls
2022-06-20 16:43:32 -04:00
David Protzman
c6598f9253
Added normalized xcorr estimate block
...
This is just a wrapper around the normalized_xcorr class
2022-06-20 16:42:47 -04:00
David Protzman
4b46d30248
Stripped down normalized cross correlator
...
This new version is much more of an estimate of the cross correlation, but it's a *lot* faster. Still need to do more real world testing of it
2022-06-20 13:07:31 -04:00
David Protzman
2e33165cbe
Turning on all available optimizations
...
This is to help the correlation module run at line rate
2022-06-20 13:06:08 -04:00
David Protzman
7606a61e8a
Added runtime configurable threshold to extractor
2022-06-20 13:02:27 -04:00
David Protzman
66b749c3ad
Added abs squared function
2022-06-20 12:49:02 -04:00
David Protzman
0fdac0e259
Fixed bug in running average tracking logic
2022-06-19 14:47:17 -04:00
David Protzman
10052a6b8d
Converted to running average to get a small performance gain
2022-06-19 13:58:16 -04:00
David Protzman
c89d6e6004
Normalization is working properly, but it's crazy slow
2022-06-19 13:42:58 -04:00
David Protzman
17f3465f3a
Initial commit for normalized cross correlation function
...
This is the crazy slow version!
2022-06-11 12:56:32 -04:00
David Protzman
bb791dbac5
Added functions for mean and var
2022-06-11 12:55:31 -04:00
David Protzman
5d59a14618
Updated drone_id_test GNU Radio example graph
...
Increased number of extracted samples to help with processing in MATLAB
Changed default RX gain from .35 to .55
Disabled the offset block (not used for SDR based processing)
Removed the time sync block as it's no longer needed
2022-06-09 00:32:34 -04:00
David Protzman
112454c3a8
Now saving off more vectors for debugging
2022-06-09 00:28:18 -04:00
David Protzman
ad1ad97181
Fixed typo in received ZC sequence saving logic
2022-06-09 00:27:48 -04:00
David Protzman
b915386ead
Dropped threshold down to .2
2022-06-09 00:27:23 -04:00
David Protzman
769daa80cc
Printing message if the total sample count isn't valid
...
This should never happen, but for when it does...
2022-06-09 00:27:13 -04:00
David Protzman
beaeec30fc
Fixed copy paste error in sample saving
2022-06-09 00:26:44 -04:00
David Protzman
3aa1b77e0b
Now outputting each burst that comes into demodulation
...
This used to happen in the time_sync block, but that block is no longer used
2022-06-09 00:12:48 -04:00
David Protzman
c863755ff8
Now writing out each received ZC sequence with a counter
2022-06-09 00:01:58 -04:00
David Protzman
7a3de6b0ce
Increased number of samples extracted from the beginning of the burst
...
This is to help MATLAB do demod since it needs a fair amount of padding before and after the burst
2022-06-08 23:57:29 -04:00
David Protzman
e36349b8a1
Now incrementing the file number for bursts coming into time sync
2022-06-08 23:56:48 -04:00
David Protzman
0729e4620e
Moved the XOR_BIT_VEC to a constant static member variable
...
This was being created for every burst which was just wasteful. It was done that way because I was in a hurry =\
2022-06-08 23:55:14 -04:00
David Protzman
4aa81eb2b9
Added CMake build dirs to ignore file
2022-06-08 23:53:35 -04:00
David Protzman
2af27b093b
Many, many updates... like.. a lot
2022-05-10 00:32:30 -04:00
David Protzman
506a43c736
Made qpsk_to_bits static
2022-05-09 18:12:04 -04:00
David Protzman
736ccbe8b1
Added bit_vec_to_string and fixed write() bug
2022-05-09 18:08:29 -04:00
David Protzman
c7e7587c24
Added debug path argument, and now running LTE decode
2022-05-09 18:07:55 -04:00
David Protzman
9518926892
Updated graph to include decode block
2022-05-09 00:58:00 -04:00
David Protzman
8cdc2c262b
Disabled printing bits
...
This was locking GNU Radio up due to the large number of console outputs
2022-05-09 00:57:19 -04:00
David Protzman
a5ce98aa54
Working on turbo decoding and rate matching
2022-05-09 00:55:46 -04:00
David Protzman
2b8b54d746
Now outputting just the data carriers
2022-05-09 00:55:18 -04:00
David Protzman
9d3b379aa7
Added util functions
...
get_data_carrier_indices
extract_data_carriers
write (generic and uint32_t)
print_bits
2022-05-09 00:54:33 -04:00
David Protzman
b11ec00b7d
Added actual impls for the decode block
2022-05-09 00:54:14 -04:00
David Protzman
62a384a942
Added decode block
2022-05-09 00:53:29 -04:00
David Protzman
0608c97ec1
Not outputting if debug path is empty
2022-05-08 23:41:30 -04:00