Publications
Detailed Information
Path Loss Exponent Prediction for Outdoor Millimeter Wave Channels through Deep Learning
Cited 2 time in
Web of Science
Cited 29 time in Scopus
- Authors
- Issue Date
- 2019-04
- Publisher
- IEEE
- Citation
- 2019 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE (WCNC), p. 8885668
- Abstract
- In this paper, we propose a new algorithm for predicting the path loss exponent of outdoor millimeter-wave band channels through deep learning method. The proposed algorithm has the advantage of requiring less inference time compared to existing deterministic channel models while concretely considering the topographical characteristics. We used three-dimensional ray tracing to generate the outdoor millimeter-wave band channel and path loss exponent. We trained a neural network with generated path loss exponent. To evaluate the performance of the proposed method, we analyzed the influence of the hyperparameters and environmental features, for example, building density and average distance from the transmitter.
- ISSN
- 1525-3511
- Files in This Item:
- There are no files associated with this item.
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.