Low-Complexity Block-Based Decoding Algorithms for Short Block Channels

2023-09-20

2023 IEEE AFRICON

AI Summary

In 3GPP short-block transmission scenarios, conventional DMRS-based least-squares (LS) channel estimation and maximum-likelihood (ML) decoding suffer from prohibitively high computational complexity for long codewords. To address this, this paper proposes a low-complexity decoding framework jointly designed for first-order Reed–Muller (RM) codes and orthogonal DMRS. The key contributions are: (1) a novel block-wise encoding structure that partitions the RM code into independent subblocks; (2) a block-based fast Hadamard transform (FHT) decoding algorithm reducing complexity from $O(N^2)$ to $O(N log N)$; and (3) DMRS-assisted LS channel estimation with adaptive power allocation, achieving minimal performance loss while preserving high receiver sensitivity. Experimental results demonstrate that the proposed scheme achieves an optimal trade-off between error-rate performance and decoding latency in short-packet communications, significantly reducing decoding time without compromising reliability.

Technology Category

• Machine Learning: Hardware-aware ML
• Search and Optimization: Mixed Discrete/Continuous Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Security and Privacy: Large-scale security measurements

Abstract

This paper presents low-complexity block-based encoding and decoding algorithms for short block length channels. In terms of the precise use-case, we are primarily concerned with the baseline 3GPP Short block transmissions in which payloads are encoded by Reed-Muller codes and paired with orthogonal DMRS. In contemporary communication systems, the short block decoding often employs the utilization of DMRS- based least squares channel estimation, followed by maximum likelihood decoding. However, this methodology can incur substantial computational complexity when processing long bit length codes. We propose an innovative approach to tackle this challenge by introducing the principle of block/segment encoding using First-Order RM Codes which is amenable to low-cost decoding through block-based fast Hadamard transforms. The Block-based FHT has demonstrated to be cost-efficient with regards to decoding time, as it evolves from quadric to quasilinear complexity with a manageable decline in performance. Additionally, by incorporating an adaptive DMRS/data power adjustment technique, we can bridge/reduce the performance gap and attain high sensitivity, leading to a good trade-off between performance and complexity to efficiently handle small payloads.

Problem

• Develops low-complexity decoding for short block channels
• Addresses computational complexity in 3GPP Short block transmissions
• Enhances efficiency using block-based fast Hadamard transforms

Innovation

• Block-based fast Hadamard transforms
• First-Order Reed-Muller Codes
• Adaptive DMRS/data power adjustment