With the rapid integration of communication and intelligence,semantic communication has garnered significant attention for its potential to enhance transmission efficiency and reduce redundancy.Among these developments,the integration of semantic communication with multiple-input multiple-output(MIMO)systems has emerged as a research focus.However,ensuring the integrity and accuracy of semantic information transmitted via MIMO systems in noisy environments remains a pressing challenge.To address this issue,a novel noise-resilient semantic communication MIMO model,SC-MIMO-Anti,is proposed.Leveraging deep learning techniques,the model incorporates an anti-noise neural network module that optimizes semantic information transmission by jointly considering semantic content and channel state information.Simulation results demonstrate that the SC-MIMO-Anti model not only ensures high-quality semantic transmission but also exhibits enhanced robustness and noise resistance.Notably,its performance advantage is particularly evident under adverse channel conditions.Comparative experiments further validate the superiority of the proposed anti-noise approach for semantic communication.Specifically,at an SNR of 0 dB,SC-MIMO-Anti achieves approximately a 6.1%improvement in sentence similarity compared to traditional noise-resilient MIMO systems.
维普
万方数据