Intelligent tuning for microwave filters based on multi-kernel machine learning model

This paper presents an intelligent alignment method for the automatic tuning device of microwave filters. In the method, a model that reveals the relationships between the tunable screws and filter electrical performance is firstly developed by using an improved machine learning algorithm which can incorporate multi-kernel into the traditional linear programming support vector regression to improve the modeling accuracy. Then, a tuning procedure of filters is constructed by using the developed machine-learning model. Finally, some experiments are carried out, and the experimental results confirm the effectiveness of the method. The approach is particularly suited to the computer-aided tuning devices or an automatic tuning robot of volume-producing filters.