abstract

• In order to realize the future Sixth-Generation (6G), the research on Integrated Sensing and Communication (ISAC) channel modeling has attracted considerable attention. In classical cluster-based models, it is difficult to effectively build high density point clouds to outline object features. As an alternative, slightly-distributed-scatterer-based signal models are utilized in this paper, and a first-order generalized array manifold is used to approximate the spread features of channel. By combining this with the space-alternating generalized expectation-maximization (SAGE) algorithm, we achieve a framework for the efficient estimation of the dispersive-path components including the spread on the angle and the Doppler frequency domains. Based on measured data in vehicular environments, the effectiveness of the algorithm is verified, and a time-varying statistical channel model is formulated. The results obtained can help to accurately model vehicular channels and provide guidance for other applications of autonomous driving. © 2024 IEEE.

authors

• Gutiérrez Díaz De León Carlos Adrián
• Li X.

publication date

• 2024-01-01

published in

• IEEE Vehicular Technology Conference  Journal

keywords

• Dispersive-path; generalized array manifold; SAGE; time-varying channel model Channel estimation; Frequency estimation; Array manifolds; Channel modelling; Dispersive-path; Distributed scatterers; Generalized array manifold; Integrated sensing; Space alternating generalized expectation maximization; Time varying channel; Time-varying channel model; Vehicular environments; Expectation maximization algorithm

Digital Object Identifier (DOI)

• 10.1109/VTC2024-Spring62846.2024.10683219

PubMed ID

start page

end page

volume

issue