Erba, UgurBekiryazici, TahirAydemir, GurkanTabakcioglu, Mehmet Baris2026-02-082026-02-082024979-8-3503-8897-8979-8-3503-8896-12165-0608https://doi.org/10.1109/SIU61531.2024.10600900https://hdl.handle.net/20.500.12885/590832nd IEEE Signal Processing and Communications Applications Conference (SIU) -- MAY 15-18, 2024 -- Tarsus Univ Campus, Mersin, TURKEYIn wireless communication systems, one of the most crucial problems is to solve the problem where to deploy the transmitter. It is essential to determine the coverage area before installing the transmitter. In simulation programs, coverage areas are calculated using ray tracing techniques and numerical integral-based wave propagation models, and positioning is done accordingly. However, when dealing with a large area and obstacles between the transmitter and receiver, the computational load significantly increases. Considering the inadequacy of traditional methods in complex urban environments and rapidly changing conditions, the integration of deep learning techniques is aimed at providing a more accurate and flexible solution. In this context, deep learning models Convolutional Neural Networks (CNNs) based, particularly, are highlighted as a potential solution for base station positioning. Among the advantages of CNNs is their ability to adapt more quickly to complex environmental variables. The developed CNN-based model has shown promising results in coverage area estimation and has the potential to enhance the performance of wireless communication networks. This study aims to contribute to the future reliability, speed, and effectiveness of wireless communication networks.trinfo:eu-repo/semantics/closedAccesscoverage areadeep convolutional neural networkgeometric opticsUTDConference Object10.1109/SIU61531.2024.10600900WOS:0012978947001462-s2.0-85200916345N/AN/A