Deep Joint Semantic Coding and Beamforming for Near-Space Airship-Borne Massive MIMO Network

TL;DR

This work tackles the challenge of reliable, high-capacity Airship-to-X links in an integrated ground-air-space network by marrying semantic communication with massive MIMO. It introduces a deep Joint Semantic Coding and Beamforming (JSCBF) framework that jointly optimizes image semantics and CSI semantics via transformer-based extraction and fusion, and integrates hybrid data-driven and model-driven beamforming in an end-to-end trainable pipeline. The approach yields significant gains in pixel-level and perceptual image quality (PSNR, MS-SSIM, LPIPS), especially under imperfect CSI and limited pilot overhead, outperforming traditional separation-based designs and standard deep JSCC. The proposed system offers improved spectral efficiency, robustness, and practical viability for near-space, airship-borne broadband networks in 6G-era IGASN deployments.

Abstract

Near-space airship-borne communication network is recognized to be an indispensable component of the future integrated ground-air-space network thanks to airships' advantage of long-term residency at stratospheric altitudes, but it urgently needs reliable and efficient Airship-to-X link. To improve the transmission efficiency and capacity, this paper proposes to integrate semantic communication with massive multiple-input multiple-output (MIMO) technology. Specifically, we propose a deep joint semantic coding and beamforming (JSCBF) scheme for airship-based massive MIMO image transmission network in space, in which semantics from both source and channel are fused to jointly design the semantic coding and physical layer beamforming. First, we design two semantic extraction networks to extract semantics from image source and channel state information, respectively. Then, we propose a semantic fusion network that can fuse these semantics into complex-valued semantic features for subsequent physical-layer transmission. To efficiently transmit the fused semantic features at the physical layer, we then propose the hybrid data and model-driven semantic-aware beamforming networks. At the receiver, a semantic decoding network is designed to reconstruct the transmitted images. Finally, we perform end-to-end deep learning to jointly train all the modules, using the image reconstruction quality at the receivers as a metric. The proposed deep JSCBF scheme fully combines the efficient source compressibility and robust error correction capability of semantic communication with the high spectral efficiency of massive MIMO, achieving a significant performance improvement over existing approaches.

Figures (13)

• Figure 1: Schematic diagram of the IGASN architecture, consisting of space satellites, near-space airships, aerial aircraft, and terrestrial BSs.
• Figure 2: Proposed deep JSCBF scheme for near-space airship-borne massive MIMO image transmission network.
• Figure 3: Image semantic extraction network, where MLP stands for multilayer perceptron, and $\times U$ ($\times T$) means the module repeats $U$ times ($T$ times).
• Figure 4: CSI semantic extraction network.
• Figure 5: Semantic fusion network.