A Multi-feature Fusion Modeling and Debugging Method for Microwave Filters

Abstract

The invention provides a multi-feature fusion modeling and debugging method for a microwave filter. The adjustable component state x is changed many times, the S parameter s is sampled and measured, the original data set including x, s and the sampling frequency f is constructed, and the s and s are measured. f is preprocessed to obtain the original features, the adjustable component state x and the original features constitute the training set; the feature fusion part of the model is constructed using the convolutional layer, the pooling layer and the activation function layer, and the feature mapping part of the model is constructed using the fully connected layer , obtain a high-precision multi-feature debugging decision-making model through training; perform the same preprocessing on the S parameters s* and f that meet the requirements of the indicators, and then input them into the high-precision multi-feature debugging decision-making model after training to obtain the corresponding adjustable components state x*, and adjust the state of the adjustable components of the microwave filter to be debugged to x*. The beneficial effects of the present invention are that the precision of the debugging decision model is improved, thereby improving the debugging efficiency.

Description

technical field [0001] The invention relates to the field of intelligent manufacturing, in particular to the field of microwave filter debugging, in particular to a microwave filter multi-feature fusion modeling and debugging method. Background technique [0002] As the international 5G competition intensifies, my country has successively proposed to accelerate the construction of 5G base stations in important documents. Microwave filters are the core frequency selection devices in 5G base stations, and their filtering performance has a great impact on the quality of frequency selection. However, in its production process, due to inevitable processing tolerances, microwave filters usually cannot meet the requirements of filtering performance indicators, and the debugging process is essential. The traditional debugging method relies on experienced debugging workers to calculate the gap between the current performance index and the target performance index according to the me...

AI Technical Summary

Problems solved by technology

The existing debugging decision-making model only uses S parameters, but does not use the target performance index, which cannot fully reflect the characteristics of the debugging process, and the obtained model has low precision, which affects the debugging efficiency

On the other hand, the dimensions of S parameters and target debugging indicators are quite different, and the relationship between them is difficult to express

Images