Time-Delay Neural Network based Amplifier Model (AMP_TDNN) in AWR

AMP_TDNN is a Time-Delay Neural Network-based (TDNN) amplifier model that can handle short-term and long-term memory effects. AMP_TDNN is a TDNN-based nonlinear behavioral amplifier model that can handle short-term and long-term memory effects. AMP_TDNN uses a model file that includes all the parameters of a trained TDNN. The model file is created by running the Amplifier Model Generator wizard. Once you have an AMP_TDNN model along with the model file, you cannot change the behavior or properties of the model at the system level; you must use the wizard to generate a new model.

It has ID, FILE and DIR parameters.

FILE. This model file defines all AMP_TDNN model parameters. The file is generated by running the Amplifier Model Generator wizard.

DIR. The location in which the model file is saved and used by AMP_TDNN for system simulations. By default, this is the directory in which the project used to create the model file is saved. The model file needs to remain in this directory to run simulations using an AMP_TDNN model along with the model file. If you move the project to a different directory after creating the model file, and you do not move the model file, an error is issued when running simulations using the model file.

The data input is Complex Type and used for Output Signal Purpose.

Team SimTech

Cadence Design Systems