The active device X-parameters extracted from physics-based TCAD simulations are imported into the Keysight ADS environment following the techniques outlined in [1], leading to an efficient, yet physically sound, circuit-level nonlinear model, retaining a direct link with the device fabrication technology. In this work, the X parameter model is used to demonstrate that a fully statistical analysis of a nonlinear circuit as a function of technological spread of the active device can be efficiently carried out within ADS, with extremely reduced simulation time and excellent agreement with the original TCAD analysis. As a test case, we address the statistical analysis of a deep class AB amplifier as a function of the variations of both the active device doping and the circuit layout. The statistical distribution of the output power significantly depends on the power backoff and exhibits a marked skew, which cannot be fully addressed by standard sensitivity or linearized approaches. The effect of correlated or uncorrelated active device or layout variations is also investigated in a combined stage.
Physically-based statistical analysis of nonlinear circuits through X-parameters / Guerrieri, S. Donati; Bonani, F.; Ghione, G.. - STAMPA. - (2019), pp. 84-87. ((Intervento presentato al convegno 2019 European Microwave Integrated Circuits Conference (EuMIC) tenutosi a Paris (France) nel 29 September - 4 October.
Titolo: | Physically-based statistical analysis of nonlinear circuits through X-parameters |
Autori: | |
Data di pubblicazione: | 2019 |
Abstract: | The active device X-parameters extracted from physics-based TCAD simulations are imported into th...e Keysight ADS environment following the techniques outlined in [1], leading to an efficient, yet physically sound, circuit-level nonlinear model, retaining a direct link with the device fabrication technology. In this work, the X parameter model is used to demonstrate that a fully statistical analysis of a nonlinear circuit as a function of technological spread of the active device can be efficiently carried out within ADS, with extremely reduced simulation time and excellent agreement with the original TCAD analysis. As a test case, we address the statistical analysis of a deep class AB amplifier as a function of the variations of both the active device doping and the circuit layout. The statistical distribution of the output power significantly depends on the power backoff and exhibits a marked skew, which cannot be fully addressed by standard sensitivity or linearized approaches. The effect of correlated or uncorrelated active device or layout variations is also investigated in a combined stage. |
ISBN: | 978-2-87487-056-9 |
Appare nelle tipologie: | 4.1 Contributo in Atti di convegno |
File in questo prodotto:
File | Descrizione | Tipologia | Licenza | |
---|---|---|---|---|
EuMiC 19.pdf | Articolo principale | 2a Post-print versione editoriale / Version of Record | Non Pubblico - Accesso privato/ristretto | Administrator Richiedi una copia |
http://hdl.handle.net/11583/2769672