Convolutional Neural Network-based Successive Cancellation List Decoder for the Polar Code.
Gespeichert in:
| Titel: | Convolutional Neural Network-based Successive Cancellation List Decoder for the Polar Code. |
|---|---|
| Autoren: | Kshirsagar, Sunil Yadav1 sunilksagar143@gmail.com, Marka, Venkatrajam2 mvraaz.nitw@gmail.com |
| Quelle: | IAENG International Journal of Computer Science. Sep2025, Vol. 52 Issue 9, p3191-3206. 16p. |
| Schlagwörter: | Convolutional neural networks, Error-correcting codes, Computer performance, Channel coding, Computational complexity, Decoding algorithms, 5G networks |
| Abstract: | In modern communication systems, polar codes play an essential role as error-correcting codes due to their inclusion in 5G technology. Successive cancellation list (SCL) decoders with cyclic redundancy checks (CRCs) enhance polar code performance, but their computational complexity and latency are high. To overcome this limitation, this study proposes a convolutional neural network (CNN)-based SCL (CNN-based SCL) decoder. The proposed decoder significantly improves the error correction capability and decoding speed in terms of bit error rate (BER), block error rate (BLER), and frame error rate (FER) versus signal-to-noise ratio (SNR). For L = 32, the gained BER, BLER, and FER, respectively, are 2:34×10-6 at 3:5 dB SNR, 0:005, and 0:002 at 3 dB SNR. We conducted a simulation of the proposed decoder and compared the results with conventional decoders. The results indicate that the proposed CNN-based SCL decoder outperforms other decoders across various list sizes. [ABSTRACT FROM AUTHOR] |
| Datenbank: | Supplemental Index |
Full Text Finder 
Nájsť tento článok vo Web of Science Schreiben Sie den ersten Kommentar!