lorenlugosch / neural-min-sum-decoding Goto Github PK

Hi,
Thanks for the code!
I got a seemly right result when the decoder type is OMS. However, when the type is NNMS, I found that the decoding performance is terrible.

Here is the command I ran:
python main.py 0 1 8 1 100 10000000000000000 5 BCH_63_36.alist BCH_63_36.gmat laskdjhf 0.5 100 FNNMS

And here are the terrible results:
BERs:
[0.1279322573687082, 0.10423613490160252, 0.08364134089553754, 0.06601353837564869, 0.05195795744356895, 0.04082400746069811, 0.03281945871873929, 0.027175561138390876]
FERs:
[0.9998800959232614, 0.9991207034372502, 0.9960531574740208, 0.9864908073541168, 0.9621203037569944, 0.9145283772981615, 0.8490007993605115, 0.7742106314948042]

cant run the simulation

Hi,

I tried to run the script on BCH (63, 45) using your default setting. However, I didn't find loss decreasing for even 100K iterations. The final BER is very close to original belief propagation. Could you help me fix it?

The output is following:

0 minibatches completed 0.26780248
20 minibatches completed 0.2812021
40 minibatches completed 0.27524856
60 minibatches completed 0.2865098
80 minibatches completed 0.2665105
.......
99800 minibatches completed 0.26420146
99820 minibatches completed 0.33095878
99840 minibatches completed 0.30447066
99860 minibatches completed 0.2416447
99880 minibatches completed 0.26412797
99900 minibatches completed 0.2476056
99920 minibatches completed 0.27065468
99940 minibatches completed 0.23565353
99960 minibatches completed 0.2468922
99980 minibatches completed 0.27805367
Trained decoder on 100000 minibatches.
....
SNR: 1.0
frame count: 106080
bit errors: 640124
BER: 0.09578335607747372
FER: 0.9434106334841629
SNR: 2.0
frame count: 124800
bit errors: 545150
BER: 0.06933633496133496
FER: 0.8014663461538462
SNR: 3.0
frame count: 186960
bit errors: 495371
BER: 0.04205729431981037
FER: 0.5349326059050065
SNR: 4.0
frame count: 399000
bit errors: 483991
BER: 0.019254127381946932
FER: 0.25065413533834585
SNR: 5.0
frame count: 1255680
bit errors: 508151
BER: 0.006423522624306263
FER: 0.07963971712538226
SNR: 6.0
frame count: 5370600
bit errors: 560241
BER: 0.0016558139287443277
FER: 0.018620079693144154

Why can't it be used for LDPC codes? Is this a requirement? Can anyone provide a working LDPC code?

Hi,

Nice work!

I see that you have H matrix in alist format. I am trying to work on a standard data set, i.e., https://www.uni-kl.de/channel-codes and only find they have H matrix.

May I ask where do you get your gmat file?

Thanks!