Iterative Semantic Decoding for Short Block Codes
Abstract
This paper proposes an iteratively enhanced semantic receiver for natural-language text transmission over noisy wireless channels using multiple short block codes. At the transmitter, each sentence is permuted by a character-level interleaver, partitioned into segments, and independently encoded by short block codes. At the receiver, we develop an iterative decoder consisting of a channel decoder and a language model, where a de-interleaver between them disperses the burst decoding errors within each segment across the sentence. In each iteration, the language model denoises the channel decoding output, and the denoised characters verified to be consistent with the channel observations are fed back to the channel decoder as semantic information for the next iteration. Simulation results on the Stanford Natural Language Inference (SNLI) corpus over the additive white Gaussian noise (AWGN) channel show that the proposed receiver achieves approximately 1.5 dB block error rate (BLER) gain over conventional short-block coding, while maintaining BLEU and ROUGE scores above 99% at SNRs beyond 1.0 dB.
Related
- On the Multi-Weight Shift-Sum Iterative Decoding of Short BCH Codes
- Iterative Bounded Distance Decoding With 2-Bit Reliability Marker For Product-Like Codes
- Semantic SCL Decoding for Polar Codes
- MacKay-Neal Codes for DNA Channels under Block-MAP Decoding
- Study of Iterative Detection and Decoding for {Mixed Near- and Far-Field} XL-MIMO Systems
BibTeX
@misc{hao2026iterative,
title = {{Iterative Semantic Decoding for Short Block Codes}},
author = {Jiafu Hao and Chentao Yue and Mingcheng Nie and Branka Vucetic and Yonghui Li},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.03256},
eprint = {2609.03256},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.03256},
}
RIS
TY - UNPB
TI - Iterative Semantic Decoding for Short Block Codes
AU - Jiafu Hao
AU - Chentao Yue
AU - Mingcheng Nie
AU - Branka Vucetic
AU - Yonghui Li
PY - 2026
DA - 2026/09/03
PB - arXiv
DO - 10.48550/arxiv.2609.03256
UR - https://arxiv.org/abs/2609.03256
AB - This paper proposes an iteratively enhanced semantic receiver for natural-language text transmission over noisy wireless channels using multiple short block codes. At the transmitter, each sentence is permuted by a character-level interleaver, partitioned into segments, and independently encoded by short block codes. At the receiver, we develop an iterative decoder consisting of a channel decoder and a language model, where a de-interleaver between them disperses the burst decoding errors within each segment across the sentence. In each iteration, the language model denoises the channel decoding output, and the denoised characters verified to be consistent with the channel observations are fed back to the channel decoder as semantic information for the next iteration. Simulation results on the Stanford Natural Language Inference (SNLI) corpus over the additive white Gaussian noise (AWGN) channel show that the proposed receiver achieves approximately 1.5 dB block error rate (BLER) gain over conventional short-block coding, while maintaining BLEU and ROUGE scores above 99% at SNRs beyond 1.0 dB.
ER -