Cavity Duplexer Tuning with 1d Resnet-like Neural Networks
This addresses a domain-specific engineering problem for RF filter tuning, but it is incremental as it adapts existing neural network architectures to a new application.
The paper tackles the problem of tuning cavity duplexers with many adjustment screws by developing a supervised learning method using a 1D ResNet-like neural network that processes S-parameters, achieving near-tuned states within 4-5 rotations per screw.
This paper presents machine learning method for tuning of cavity duplexer with a large amount of adjustment screws. After testing we declined conventional reinforcement learning approach and reformulated our task in the supervised learning setup. The suggested neural network architecture includes 1d ResNet-like backbone and processing of some additional information about S-parameters, like the shape of curve and peaks positions and amplitudes. This neural network with external control algorithm is capable to reach almost the tuned state of the duplexer within 4-5 rotations per screw.