Menu
Home
Log in / Register
 
Home arrow Computer Science arrow Behavioral Modelling and Predistortion of Wideband Wireless Transmitters
Source

Conclusions

In this chapter, practical aspects for power amplifier modeling, including characterization and identification techniques, were presented. It was shown that the power amplifier behavior is significantly affected by the signal statistical characteristics. Therefore, for better model accuracy, it is preferable to characterize the power amplifier using a modulated signal with similar characteristics to the model that will be used for the power amplifier. While such a characterization method achieves better modeling accuracy than using a continuous wave or multi-tone characterization, the use of a modulated signal in the characterization process requires involving more complicated identification techniques compared to a continuous wave characterization.

Two different classes of identification techniques were presented. The first class consists of moving average algorithms used to separate the static nonlinearity from the dynamic behavior of the power amplifier. These algorithms are used to identify memoryless models including look-up table based models or equation based models. The moving average algorithms are also used in the identification of box oriented models that include static nonlinear box(es) within the model.

The second class of model identification is used for the identification of equation based models including the family of memory polynomial based models and their variations. Three different algorithms for the identification of the model coefficients were presented and compared in terms of quality of estimation quantified with the normalized mean squared of the residual error, the robustness of convergence, the computational complexity, and the speed of convergence.

 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Mathematics
Political science
Philosophy
Psychology
Religion
Sociology
Travel