Analysis of Local Channel Profile Mismatch Reduction Methods in Geographically Segmented Localcasting Scenarios
Hao Ju, Xudong Guan, Yin Xu, Liang Gong, Dazhi He, Haoyang Li, Wenjun Zhang, Yiyan Wu · 2025
In the past year, Geographically Segmented Localcasting (GSL) has garnered significant attention from the broadcast industry. This is due to its implementation of Local Content Insertion (LCI) within a Single Frequency Network (SFN) environment, which aids local multimedia and data broadcasting services. LCI is achieved through Layered Division Multiplexing (LDM). The Core Layer (CL) of LDM provides wide-area broadcasting, forming a large-scale SFN, while the Enhanced Layer (EL) facilitates localcasting services. Since the Advanced Television Systems Committee (ATSC) 3.0 standard was established over a decade ago, LCI/GSL represents a new business model. Originally designed for channel estimation in SFN, the pilot signals are not suitable for estimating the localcasting channels in ELs. This can lead to Local Channel Profile Mismatch (LCPM) in the ELs, significantly impacting the performance of GSL. Therefore, this paper models and analyzes the impact of LCPM on GSL scenario reception under various conditions. A method based on the Denoising Diffusion Probabilistic Model (DDPM) is proposed to reduce LCPM and co-channel interference. This approach can achieve effective performance gains solely through algorithmic changes at the receiver, without altering the ATSC 3.0 protocol framework. It is also applicable to similar scenarios in cellular wireless networks (5G/6G) for enhancement