Memory Management in Successive-Cancellation based Decoders for Multi-Kernel Polar Codes

TL;DR

This paper introduces an efficient memory management framework for successive-cancellation decoders of multi-kernel polar codes, enabling flexible hardware implementation and improved handling of various kernel sizes and code lengths.

Contribution

It presents a generalized memory management approach tailored for multi-kernel polar code decoders, adaptable across hardware architectures and decoding algorithms.

Findings

Framework reduces memory usage in decoding

Applicable to various kernel sizes and code lengths

Complexity estimates provided for different configurations

Abstract

Multi-kernel polar codes have recently been proposed to construct polar codes of lengths different from powers of two. Decoder implementations for multi-kernel polar codes need to account for this feature, that becomes critical in memory management. We propose an efficient, generalized memory management framework for implementation of successivecancellation decoding of multi-kernel polar codes. It can be used on many types of hardware architectures and different flavors of SC decoding algorithms. We illustrate the proposed solution for small kernel sizes, and give complexity estimates for various kernel combinations and code lengths.