Research Article
Performance Evaluation of Conventional and Neural Network-Based Decoder for an Audio of Low-Girth LDPC Code
Table 4
Obtained BER, PSNR, and MSE of low-girth audio sample 3 of train noise of conventional and neural network decoders [
21].
| SNR (dB) | BER conventional decoder | BER NN decoder | PSNR conventional decoder | PSNR NN decoder | MSE conventional decoder | MSE NN decoder |
| 1 | 0.872454 | 0.056285 | 39.70957 | 40.25216 | 0.956956 | 0.466867 | 7 | 0.900528 | 0.000732 | 39.71099 | 41.74078 | 0.955217 | 0.050925 | 8 | 0.904614 | 0.000214 | 39.71115 | 41.96716 | 0.955029 | 0.035124 | 9 | 0.904491 | 2.92E − 05 | 39.71028 | 44.3992 | 0.956092 | 0.000331 | 10 | 0.909436 | 0 | 39.70917 | 96.32947 | 0.957452 | 0 | 11 | 0.911195 | 0 | 39.71127 | 96.32947 | 0.954872 | 0 |
|
|