Prediction Model for Microwave Radio Unit Testing
Typ
Examensarbete för masterexamen
Program
Embedded electronic system design (MPEES), MSc
Publicerad
2020
Författare
Balasubramaniam, Sivasenapathi
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
During radio unit testing, the radio unit is tested in different test flows in order to
ensure its conformance. But testing radio unit is time consuming and increases the
unit’s production time. Prediction models are developed for the test flows to predict
test points continuously by using the inputs from the previous test flows. Machine
learning models presented in this thesis helps in the prediction of test point values
before placing the radio unit in all testing stations. The developed prediction model
helps in finding the measured value and their dependencies for failure. The model
presented in this thesis helps to predict the future test point value and reduces the
unit time to market. In addition, to obtain a more intuitive insight Graphical User
Interface is built up to access the prediction model.
Beskrivning
Ämne/nyckelord
Prediction model , Regressor , Radio unit , test flows , test points , Data , Features , Model , Modelling