A Novel Measurement Based Method Enabling Rapid Extraction of Bayesian Inference-Based Behavioral Model
Fei Wang, Jialin Cai, Jun Liu, Jiangtao Su · 2020
In this paper, a novel model extraction method is proposed, which can extract behavioral model based on Bayesian inference accurately and efficiently. This method uses a simple active load-pull architecture, and only needs to change the amplitude and phase of the incident wave A2at the load port in the process during the test, so that the training data can be acquired for the Bayesian algorithm, without the need for a complete loadpull test. Compared with the traditional scheme, this scheme can save the impedance iteration time and greatly improve the model extracting efficiency. The experiment results prove that the method has greatly increase the extracting speed without compromising the accuracy.