Page tree
Skip to end of metadata
Go to start of metadata


System Data Models


The System - Data Models palette contains two types of components: those which enable the simulation-based extraction of behavioral profiles from circuit models and those which perform behavioral modeling using these profiles. There is an inherent pairing of these two types of components because parameters used to define behavioral profiles in data-based representation are specific to modeling theories and therefore data model requirements. Data-based behavioral modeling is an important tool for the bottom-up verification process; thus, each extractor-data model pair of this palette is also known as a verification model extraction (VME) pair. The most common usage of a VME pair during RFIC and MMIC design flows is as follows:

  • Creation of a circuit level design with circuit level components and/or parametric behavioral components during the top-down design phase. Parameter-based behavioral models are located in the Filters , System - Amps & Mixers , and System - Mod/Demod palettes. These models are characterized by a few independent parameters such as frequency, power and load.
  • Creation of a Harmonic Balance-based dataset or data file profiles of these circuit level models using the VME extractor components from the System - Data Models palette. Use the VME data model counterparts to recreate the behavior of the circuit level models. Compare circuit- and data-based behavioral models; calibrate the data-based behavioral models prior to hardware prototyping the circuit.
  • Using measurement instruments to extract behavioral datasets or data files from hardware prototypes from the foundry. Using the calibrated VME data model to generate a hardware behavioral profile. Compare the hardware behavioral profile with the simulation-based behavioral profile to estimate the success of prototyping.

T he parameter-based behavioral models typically provide superior speed whereas data-based behavioral models typically provide superior accuracy. The differences between parameter - and data-based behavioral models justify the palette emphasis on ADS flow (separate palettes for parametric and data-based models) rather than functionality (separate palettes for amplifiers, mixers, modulators each containing both parametric and data-based models).

The BehavioralModels example suite was developed to promote the usability of the System-Data Models components. This family of example projects demonstrates the use and characteristics of most VME models. To access the example project from the ADS Main window click on the following sequence of menu options: File > Example Project > _BehavioralModels. Notes for each example project are available in the accompanying notes in Behavioral Model Examples.

In addition to the data-based behavioral model components, the System-Data Models palette includes Balun3Port and Balun4Port; these components are also accessible from the System-Passive palette.

Classification of ADS System Data Models

ADS provides data-based modeling for system-level VME components of the following categories:

  • Amplifiers: AmplifierH1H2, AmplifierP2D, AmplifierS2D, AmpLoadPull, VCA_Data
  • Mixers: MixerHBdata, MixIMT_Data
  • Modulators / Demodulators: IQ_Mod, IQ_Demod

Most data models accept customized dataset (*.ds) or data file (*.p2d, *.s2d, *.imt) inputs. These inputs are typically provided by data model extractors. Identified by the suffix Setup or setup, these extractors are prefabricated subcircuits that run a predetermined simulation with user-settable sweep values on the circuit level model of interest and report the input and output stimuli in the dataset or data file form. The user is encouraged to push into the extraction subcircuits and be familiar with the measurement metrics that are being reported to the dataset or data file.

Most system data models have their own data extractors; however, there are exceptions for the older models MixerIMT and MixerIMT2 for which data files must be manually generated or retrieved from a measurement instrument.

Data-Based Behavioral Modeling Components and Example Projects in ADS. summarizes the availability and scope of each VME model currently available in ADS. This section describes component parameters and usage in further detail. For a discussion of the use of the ADS data model components from an application point of view please refer to the series of Application Notes being released under the title Behavioral Modeling in ADS .

Data-Based Behavioral Modeling Components and Example Projects in ADS
Extractor Model Data Model Data Type Comments on Scope of Modeling
AmpH1H2_Setup AmpH1H2 .ds Models nonlinearity at odd and even harmonics using circuit-based data at fundamental and second harmonics. Refer to Examples > BehavioralModels > AmplH1H2_prj
AmplifierP2D_Setup AmplifierP2D .p2d Models nonlinearity and noise at fundamental frequency using circuit-based data at fundamental frequency. For circuit envelope simulations, modeling nonlinear distortion at fundamental frequency due to interference from spectral components within the envelope band. Refer to Examples > BehavioralModels > AmplifierP2D_prj
AmplifierS2D_Setup AmplifierS2D .s2d Models nonlinearity and noise at fundamental and odd-order harmonic frequencies using circuit-based data at fundamental frequency. Assumes odd-order polynomial fitting for harmonic calculations. Refer to Examples > BehavioralModels > AmplifierS2D_prj
LoadPullSetup AmpLoadPull .ds Models nonlinearity at fundamental frequency for various input and output impedance conditions using circuit-based data at single frequency for a range of source and load conditions. Refer to Examples > BehavioralModels > AmpLoadPull_prj
VCA_Setup VCA_Data .ds Models nonlinearity at fundamental frequency for various settings of control voltage using circuit-based data at single frequency for a range of bias conditions. Refer to Examples > BehavioralModels > VCA_prj
MixIMTA_Setup MixIMTB_Setup MixIMT_Data .imt Models nonlinearity at all mixing products for single (O-type and A-type) or multi (B-type) IMT files. Refer to Examples > BehavioralModels > MixIMT_pr j
MixerHBsetup MixerHBdata .ds Models nonlinearity at fundamental, mixing product frequencies at the mixer output (IF) due to 1tone input signals (RF and LO) based on 1-tone harmonic balance characterization of circuit level mixer. Refer to Examples > BehavioralModels > MixerHBdata_prj
(Manual extraction) MixerIMT †† .imt Models nonlinearity at fundamentals and mixing product frequencies at the mixer output (IF) based on 1-tone input signals and a table of mixing coefficients.
(Manual extraction) MixerIMT2 .imt Models nonlinearity at fundamentals and mixing product frequencies at the mixer output (IF) based on 1-tone input signals and a table of mixing coefficients.
IQ_Mod_Setup IQ_Mod_Data .ds Models large-signal in-phase and quadrature components during modulation Refer to Examples > BehavioralModels > IQ_Mod_prj
IQ_Demod_Setup IQ_Demod_Data .ds Models large-signal in-phase and quadrature components during demodulation Refer to Examples > BehavioralModels > IQ_Demod_prj
       This data model accepts only O-type IMT data. ††    This is an obsolete component.


Components

  • No labels