Communication inside aircraft cabin has gained more and more attentions to provide in-flight connectivity for high-quality user experience and enable the wireless communication of electromechanical equipment. In this paper, a power prediction method considering the coordinates of transceiver, propagation distance and carrier frequency based on the parameterized channel model and machine learning (ML) framework is proposed for wireless communication inside the aircraft cabin. In the proposed channel model, a random forest (RF) network is trained for channel characteristics based on ray tracing (RT) simulation data and applied to estimate the power parameters. The proposed method is verified inside a typical aircraft cabin at 1GHz and compared with traditional model and background propagation neural network (BPNN). The simulation results show that RF can accurately predict the power parameter and reproduce the channel environment. The prediction results of power distribution can provide valuable reference for the layout and optimization of communication nodes inside the aircraft cabin.
Machine Learning-Based Channel Power Analysis inside Aircraft Cabin
24.02.2023
4071390 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
MIMO Channel Measurements Inside an Aircraft Cabin with Planar Transmit Antennas
Tema Archiv | 2009
|Europäisches Patentamt | 2021
|Europäisches Patentamt | 2017
|CABIN-INSIDE DETECTION DEVICE AND CABIN-INSIDE DETECTION METHOD
Europäisches Patentamt | 2023
|Cabin-inside detection device and cabin-inside detection method
Europäisches Patentamt | 2024
|