کنفرانس بین المللی فناوری اطلاعات و دانش

English

صفحه اصلی / شانزدهمین کنفرانس بین المللی فناوری اطلاعات و دانش

Adaptive Semantic Communication for Non-Terrestrial Networks

نویسندگان :

Soroosh Miri¹ Sepehr Abolhasani² S. Mohammad Razavizadeh³

1- دانشگاه علم و صنعت ایران 2- دانشگاه علم و صنعت ایران 3- دانشگاه علم و صنعت ایران

کلمات کلیدی :

semantic communication،non-terrestrial networks (NTN)،deep joint source-channel coding (DJSCC)،adaptive feature selection،image transmission،wireless robustness

چکیده :

In recent years, non-terrestrial networks (NTNs) have demonstrated substantial potential for delivering ubiquitous, continuous, and scalable wireless services. Nevertheless, these networks encounter significant challenges, such as limited bandwidth, unstable channel conditions, and dynamic link variability— particularly in aerial communication scenarios. Concurrently, semantic communication has emerged as a compelling paradigm that reduces network overhead and improves wireless efficiency by transmitting semantic content rather than raw data, positioning it as an ideal solution for NTN environments. To achieve reliable and efficient image transmission under these constraints, this paper proposes a Non-Terrestrial Adaptive Deep Joint Source-Channel Coding framework (NTA-DJSCC). The framework adaptively selects the most salient image features for transmission, incorporating both the input image content and prevailing channel conditions to ensure robust and high-fidelity image delivery. By integrating semantic awareness with adaptive feature selection, NTA-DJSCC attains superior performance in image fidelity and robustness compared to conventional approaches, thereby offering a robust foundation for advanced NTN-enabled applications.