You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+27Lines changed: 27 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -19,6 +19,31 @@ This implementation should be more suited to trellis having more transitions bet
19
19
* Viterbi Volk (state parallelization): implements the classical Viterbi algorithm, but uses Volk to enable parallell processing of states (Add-Compare-Select is done on multiple states at the same time).
20
20
This implementation should be more suited to trellis having states than transitions between states (it is the case of most error correcting codes).
21
21
22
+
# Installation
23
+
24
+
## Requirements
25
+
26
+
`gr-lazyviterbi` depends on:
27
+
* GNURadio >= 3.8.0
28
+
* Volk >= 2.0.0
29
+
* cmake (same version as for GNURadio 3.8)
30
+
31
+
## Under linux
32
+
33
+
Within `gr-lazyviterbi` directory, and as a regular user:
34
+
```sh
35
+
$ mkdir build
36
+
$ cd build
37
+
$ cmake ..
38
+
$ make -j4
39
+
```
40
+
A root:
41
+
```sh
42
+
# make install
43
+
```
44
+
45
+
# Examples
46
+
22
47
One GRC example is provided in the examples/ directory:
23
48
*`simple.grc` simple example showing how each decoder should be use in a typical flowgraph. You will have to populate the variable `base_dir`, and put the full path to the source folder of `gr-lazyviterbi` (for example: `/usr/local/src/gr-lazyviterbi`).
24
49
@@ -35,6 +60,8 @@ The convolutional code used in this script can easily be changed by supplying it
35
60
treillis in a FSM file (see gr-trellis documentation), to see what kind of speedup
36
61
can be expected for a particular use case.
37
62
63
+
# Performance
64
+
38
65
There is no best implementation. Performance depends on the trellis, the SNR of the transmission, the processor in your computer.
39
66
Bellow is bitrate comparisons for some trellis, using a computer equiped with an Intel core i5 775 (see `examples/bitrate_vs_ebn0.py` for details on simulation parameters).
0 commit comments