When transmitter characteristics are unknown, path loss inference model depend on electromagnetic signal samples to deduce signal propagation characteristics. Due to spatial discontinuities in sampling data, these mod...
详细信息
When transmitter characteristics are unknown,path loss inference model depend on electromagnetic signal samples to deduce signal propagation *** to spatial discontinuities in sampling data,these models require advance...
详细信息
ISBN:
(数字)9789887581581
ISBN:
(纸本)9798350366907
When transmitter characteristics are unknown,path loss inference model depend on electromagnetic signal samples to deduce signal propagation *** to spatial discontinuities in sampling data,these models require advanced inferential techniques to address missing data and enhance accuracy,highlighting existing methods' need for improved precision in electromagnetic *** response to the aforementioned requirement,this paper proposes a deep learning framework named LDFF-Net to generate continuous path loss values in geographic space,thereby creating path loss *** framework is characterized by a lightweight feature fusion strategy,a multi-level autoencoder structure,and a progressively enhanced dilated convolution *** learns the propagation characteristics of electromagnetic signals through measured electromagnetic data and the distribution of urban *** conducted on open-source datasets have demonstrated that this model achieves an accuracy that is 10% higher than the most advanced methods in various scenarios and experimental setups,while also exhibiting superior robustness.
暂无评论