Multi Kernel Polar Code using Nut cracker optimization based GAN for Successive Cancellation Decoder to attain low latency and high efficiency

Multi-Kernel Polar Code is a type of polar code that utilizes multiple kernels for information encoding and decoding. The selection and optimization of multiple kernels can be challenging. Previous research has demonstrated that existing deep learning models may achieve high decoding accuracy and sp...

Full description

Saved in:
Bibliographic Details
Published in:Evolving systems Vol. 16; no. 3; p. 93
Main Authors: Pushpa, B. Yamini, Panda, Sunita
Format: Journal Article
Language:English
Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.09.2025
Springer Nature B.V
Subjects:
Accuracy
Algorithms
Artificial Intelligence
Binary phase shift keying
Codes
Coding
Coding theory
Complex Systems
Complexity
Decoding
Efficiency
Energy consumption
Engineering
Error correction & detection
Generative adversarial networks
Generative artificial intelligence
Literature reviews
Machine learning
Methods
Optimization
Original Paper
Random noise
Wireless communications
GEN AI
Multi Kernel Polar Code
Decoder
Transmission matrix
Encoder
Successive cancellation
ISSN:1868-6478, 1868-6486
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Multi-Kernel Polar Code is a type of polar code that utilizes multiple kernels for information encoding and decoding. The selection and optimization of multiple kernels can be challenging. Previous research has demonstrated that existing deep learning models may achieve high decoding accuracy and speed for polar code when the block length can be very tiny. Its speed, however, dramatically drops with longer codes because of the huge network structure. A successful Generative Artificial Intelligence (GEN AI) is developed in this work for decoding polar codes. The input sequence has been encoded using multiple kernel polar codes, giving a polar encoded output. After encoding the message using the multi-kernel polar encoder, the resulting bits are mapped to binary phase-shift keying (BPSK) symbols prior to transmission. The Gaussian noise term with zero mean and variance in additive white Gaussian noise (AWGN) is used to receive the signal. The improved Generative Adversarial Network (GAN) improves the decoder performance under different channel conditions after the signals have been transmitted via the channel. Computation is employed to determine the approximate reliability of the bit channel. The proposed approach achieves 95.30% of accuracy, 4.70% error, 91.70% precision and 96.80% specificity. Thus, the designed optimized GAN model is the best option for successive cancellation in the decoder. Graphical abstract
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1868-6478
1868-6486
DOI:10.1007/s12530-025-09722-9