HyperHyperNetwork for the Design of Antenna Arrays

Shahar Lutati, Lior Wolf
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:7214-7223, 2021.

Abstract

We present deep learning methods for the design of arrays and single instances of small antennas. Each design instance is conditioned on a target radiation pattern and is required to conform to specific spatial dimensions and to include, as part of its metallic structure, a set of predetermined locations. The solution, in the case of a single antenna, is based on a composite neural network that combines a simulation network, a hypernetwork, and a refinement network. In the design of the antenna array, we add an additional design level and employ a hypernetwork within a hypernetwork. The learning objective is based on measuring the similarity of the obtained radiation pattern to the desired one. Our experiments demonstrate that our approach is able to design novel antennas and antenna arrays that are compliant with the design requirements, considerably better than the baseline methods. We compare the solutions obtained by our method to existing designs and demonstrate a high level of overlap. When designing the antenna array of a cellular phone, the obtained solution displays improved properties over the existing one.

Cite this Paper


BibTeX
@InProceedings{pmlr-v139-lutati21a, title = {HyperHyperNetwork for the Design of Antenna Arrays}, author = {Lutati, Shahar and Wolf, Lior}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {7214--7223}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/lutati21a/lutati21a.pdf}, url = {https://proceedings.mlr.press/v139/lutati21a.html}, abstract = {We present deep learning methods for the design of arrays and single instances of small antennas. Each design instance is conditioned on a target radiation pattern and is required to conform to specific spatial dimensions and to include, as part of its metallic structure, a set of predetermined locations. The solution, in the case of a single antenna, is based on a composite neural network that combines a simulation network, a hypernetwork, and a refinement network. In the design of the antenna array, we add an additional design level and employ a hypernetwork within a hypernetwork. The learning objective is based on measuring the similarity of the obtained radiation pattern to the desired one. Our experiments demonstrate that our approach is able to design novel antennas and antenna arrays that are compliant with the design requirements, considerably better than the baseline methods. We compare the solutions obtained by our method to existing designs and demonstrate a high level of overlap. When designing the antenna array of a cellular phone, the obtained solution displays improved properties over the existing one.} }
Endnote
%0 Conference Paper %T HyperHyperNetwork for the Design of Antenna Arrays %A Shahar Lutati %A Lior Wolf %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-lutati21a %I PMLR %P 7214--7223 %U https://proceedings.mlr.press/v139/lutati21a.html %V 139 %X We present deep learning methods for the design of arrays and single instances of small antennas. Each design instance is conditioned on a target radiation pattern and is required to conform to specific spatial dimensions and to include, as part of its metallic structure, a set of predetermined locations. The solution, in the case of a single antenna, is based on a composite neural network that combines a simulation network, a hypernetwork, and a refinement network. In the design of the antenna array, we add an additional design level and employ a hypernetwork within a hypernetwork. The learning objective is based on measuring the similarity of the obtained radiation pattern to the desired one. Our experiments demonstrate that our approach is able to design novel antennas and antenna arrays that are compliant with the design requirements, considerably better than the baseline methods. We compare the solutions obtained by our method to existing designs and demonstrate a high level of overlap. When designing the antenna array of a cellular phone, the obtained solution displays improved properties over the existing one.
APA
Lutati, S. & Wolf, L.. (2021). HyperHyperNetwork for the Design of Antenna Arrays. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:7214-7223 Available from https://proceedings.mlr.press/v139/lutati21a.html.

Related Material