Metin için sayısal semantik iletişim yöntemleri

Semantic communication, a paradigm concentrating on correctly transmitting underlying semantic information instead of bit sequences, has proved effective. Deep learning (DL) enabled methods are mainly used with basic natural language processing (NLP) techniques for the semantic communication of text. However, most previous DL-enabled semantic communication systems use continuous-valued channel vectors incompatible with existing digital modulation-based systems and binary channels. First, to address this issue, we propose Sememe-based Semantic Communications (SememeSC), a semantic communication paradigm that utilizes a fundamental concept from linguistics, sememes. We design a semantic source encoding-decoding strategy based on linguistics and DL, combined with conventional communication algorithms. We provide experimental results verifying that the proposed SememeSC performs superior to baselines in additive white Gaussian noise (AWGN) and Rayleigh fading channels. Second, we extend DL-based joint source-channel coding architecture to binary channels with insertions, deletions, and substitutions (IDS). We propose a novel three-stage training algorithm combining gated recurrent unit (GRU) networks for marker detection, transformer-based semantic communication for continuous latent space, and lookup-free quantization for binarized latent space optimization. The proposed DeepJSOC is the first to integrate DL-based error correction networks into joint-source channel coding schemes for binary channels with synchronization errors. We demonstrate the effectiveness of DeepJSOC by numerical experiments, achieving significant improvements over existing methods in text transmission over IDS channels.