ViT-FiLM Conditional SSB Codebook Optimization for Cell-Free mmWave Massive MIMO Systems
IEEE Communications Letters · 2026 · 30 · 2665-2669
DOI: 10.1109/lcomm.2026.3713946 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Communications Letters, 2026. Verfasst von Yulai Han und 3 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben. Gefördert von: ZTE Industry-University-Institute Cooperation Funds, Beijing Natural Science Foundation, National Natural Science Foundation of China.
Abstract (Originalsprache)
Cell-free (CF) network architecture has been considered promising for supporting millimeter-wave (mmWave) multiple-input multiple-output (MIMO) in the future 6G networks, by coordinating multiple distributed access points (APs). In CF mmWave massive MIMO systems, synchronization signal block (SSB) codebook is a key enabler of fast and reliable initial beam sweeping. However, existing SSB codebook optimization methods designed for single-AP scenarios cannot match the multi-AP cooperative channel environment, which leads to beamforming gain loss. To solve this problem, we propose a Vision Transformer-Feature-wise Linear Modulation (ViT-FiLM) conditional SSB codebook optimization method in this letter. In the proposed method, a structured beamspace representation that integrates the original multi-AP codebooks is first established to serve as the primary input, and a low-dimensional conditioning vector is formed by aggregating SSB measurements of multiple APs. Then, the designed FiLM-conditional ViT blocks enable single-pass and unified synthesis of multi-AP codebooks, by using layers modulated by the low-dimensional conditioning vector. Simulations show that the proposed method consistently outperforms the DFT codebook, the NBL learning-based codebook optimization baseline, and the CF-oriented FastBT baseline in aggregated RSRP performance with a gain around 6-9 dB. Moreover, it reduces floating-point operations (FLOPs) by approximately 47%, yielding a computationally efficient beam-management solution for CF mmWave massive MIMO systems.
Förderung
- ZTE Industry-University-Institute Cooperation Funds
- Beijing Natural Science Foundation
- National Natural Science Foundation of China
Zitieren
APA
Han, Y., Tan, J., Zhang, X., & Wang, J. (2026). ViT-FiLM Conditional SSB Codebook Optimization for Cell-Free mmWave Massive MIMO Systems. IEEE Communications Letters, 30, 2665-2669. https://doi.org/10.1109/lcomm.2026.3713946
Vancouver
Han Y, Tan J, Zhang X, Wang J. ViT-FiLM Conditional SSB Codebook Optimization for Cell-Free mmWave Massive MIMO Systems. IEEE Communications Letters. 2026;30:2665-2669. doi:10.1109/lcomm.2026.3713946
BibTeX
@article{Han2026,
author = {Han, Yulai and Tan, Jingbo and Zhang, Xudong and Wang, Jintao},
title = {ViT-FiLM Conditional SSB Codebook Optimization for Cell-Free mmWave Massive MIMO Systems},
journal = {IEEE Communications Letters},
year = {2026},
volume = {30},
pages = {2665--2669},
doi = {10.1109/lcomm.2026.3713946}
}
EUVIMED