Radio Map Estimation Based on Generative Artificial Intelligence: Evolution from Point-Level to Cell-Level Prediction

Document Type

Article

Source of Publication

IEEE Communications Magazine

Publication Date

4-23-2025

Abstract

This article explores the application of generative artificial intelligence (AI) in refining radio map estimation for complex urban wireless networks, shifting from conventional point-level to advanced cell-level predictions. Addressing the challenges posed by urban density and diversity in wireless propagation models, the study employs deep learning techniques to surmount these barriers. We propose a three-step deployment framework for generative AI, encompassing scene analysis, adaptation to new environments, and efficient network establishment. This framework empha-sizes the synergy of human expertise and AI in optimizing base station placement and network design, particularly in urban contexts. The article underscores the significance of integrating deep learning with traditional wireless communication knowledge, aiming to enhance the precision and efficiency of network configurations. By offering innovative solutions and methodolo-gies, this research contributes to the evolution of 5G-Advanced and 6G networks, highlighting the transformative role of generative AI in advancing future wireless communication technologies.

ISSN

0163-6804

Volume

63

Issue

5

First Page

150

Last Page

156

Disciplines

Computer Sciences

Keywords

Generative AI, Radio map estimation, Deep learning, Urban wireless networks, 5G-Advanced

Scopus ID

05003863918

Indexed in Scopus

yes

Open Access

no

Link to Full Text

Share

COinS