Λεπτομέρειες βιβλιογραφικής εγγραφής
| Τίτλος: |
Low-Latency and High-Efficient Multi-Kernel Polar Encoding with Deep Deterministic Policy Gradient for Successive Cancellation Decoder. |
| Συγγραφείς: |
Pushpa, B. Yamini1 (AUTHOR) yamini.pushpa@gmail.com, Panda, Sunita1 (AUTHOR) sunita.nano@gmail.com |
| Πηγή: |
International Journal of Computational Intelligence & Applications. Jul2026, p1. 31p. 26 Illustrations. |
| Θεματικοί όροι: |
*Telecommunication systems, Decoding algorithms, Error-correcting codes, Reinforcement learning, Error rates, Phase shift keying |
| Περίληψη: |
Multi-Kernel Polar Codes (MKPCs) play a crucial role in extending the flexibility of polar code construction. Combining different kernel matrices enables a wider range of block lengths and greater adaptability to various communication scenarios while preserving the polarization effect. However, polar codes often suffer from limited scalability and poor error-rate performance at short block lengths, which affects their decoding performance. To overcome these drawbacks, a Successive Cancellation (SC) decoder based on the Deep Deterministic Policy Gradient (DDPG) algorithm was developed. Initially, communication within the network begins with generating a sequence of information bits that represent a specific message acquired from a device. The message is converted to a binary representation using American Standard Code for Information Interchange (ASCII) values, resulting in a binary matrix. Multi-kernel polar encoding is used to construct a generator matrix that supports flexible codeword lengths. Frozen bits are inserted at predefined positions to improve reliability. After encoding, the message results are mapped to Binary Phase Shift Keying (BPSK) modulation, which encodes binary 0 and 1 s using specific phase shifts of a carrier wave. The modulated symbols are then transmitted over Additive White Gaussian Noise (AWGN) channels, which simulate real-world communication conditions by adding random Gaussian noise to the signals. After the signal is transmitted over the channel, reinforcement learning-based SC decoding is applied using the DDPG algorithm. It dynamically adjusts decision thresholds using Log-Likelihood Ratios (LLRs) to minimize latency and reconstruct the original bits. The results show 98.50% accuracy, 92.60% NPV, 92.60% F1-score, and 6.70% FDR. The proposed approach improves transmission by converting information bits into a robust format and reconstructing them accurately despite noise interference. This ensures higher decoding precision and minimizes errors for reliable communication. [ABSTRACT FROM AUTHOR] |
|
Copyright of International Journal of Computational Intelligence & Applications is the property of World Scientific Publishing Company and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) |
| Βάση Δεδομένων: |
Business Source Index |