Artificial Neural Network Measurement-Based Transistor Models

Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

Project JJB1

• Title: Artificial Neural Network Measurement-Based Transistor Models

• Description: Behavioural models work by mapping the functional behaviour of a device/system with an equation of some sort. The “Cardiff Model” is a behavioural modelling solution for characterising transistors behaviour under different operational conditions so that measurement data can be quickly used within CAD for the design of Power Amplifiers (PAs) etc. This project aims to replace and improve upon the current “Cardiff Model” Least Mean Squares (LMS) extraction methodology by implementing an Artificial Neural Network (ANN) to extract the model coefficients. For this project, ADS will be used to generate the data for model extraction, however, the platform for ANN implementation is unknown and will be an important part of the project.

• Explore the use of neural networks in relation to RF device characterisation and modelling, and develop an ANN model for various transistor data sets that include harmonics, power variation, and bias variation.