A linear array beamforming algorithm based on RBF neural network

Beamforming has been a concern in array signal processing. Realizing the maximum gain in the ideal direction and the mulling in the interference direction are the main purposes of beamforming. However, traditional beamforming algorithms would suffer from complex computation to achieve the above goal...

Full description

Saved in:
Bibliographic Details
Published in2021 International Conference on Microwave and Millimeter Wave Technology (ICMMT) pp. 1 - 3
Main Authors Zhang, Yinghao, Hu, Haoquan, Lei, Shiwen, Xie, Qi, Shi, Honghai, Xu, Jun
Format Conference Proceeding
LanguageEnglish
Published IEEE 23.05.2021
Subjects
Array signal processing
Beamforming
Microwave integrated circuits
Microwave technology
Millimeter wave technology
MVDR
Neural networks
RBF neural network
real-time
Signal processing algorithms
Training
Online AccessGet full text
DOI10.1109/ICMMT52847.2021.9617891

Cover

More Information
Summary:Beamforming has been a concern in array signal processing. Realizing the maximum gain in the ideal direction and the mulling in the interference direction are the main purposes of beamforming. However, traditional beamforming algorithms would suffer from complex computation to achieve the above goals, resulting in the quick respond requirement unsatisfied. In order to compute the array excitation in a real-time way, this paper designs a beamforming model based on RBF neural network. The neural network can be well-trained with the known array excitation and the array steering vector covariance matrix. With this trained neural network, the array excitation can be quickly achieved, which, as a result, would be able to greatly reduce the computation time.
DOI:10.1109/ICMMT52847.2021.9617891