CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Integration of Python based nonlinear analysis of electro-acoustic resonators in ADS EDA platform

Estudiants que han llegit aquest projecte:

Director/a: COLLADO GÓMEZ, CARLOS

Departament: TSC

Títol: Integration of Python based nonlinear analysis of electro-acoustic resonators in ADS EDA platform

Data inici oferta: 30-01-2026     Data finalització oferta: 30-09-2026


Estudis d'assignació del projecte:
    MU MASTEAM 2015
Tipus: Individual
 
Lloc de realització: EETAC
 
Segon director/a extern: Marta González Rodríguez
 
Paraules clau:
Electro-acoustic filters, SAW, AWR, ADS, Python
 
Descripció del contingut i pla d'activitats:
The project proposed aims to develop a software tool (Python) to simulate the non-linearities of electro-acoustic resonators using the ADS circuit computing program. This project will consist in the continuation of the same project previously carried out using the AWR software.
The application to be developed must be easy to use and must allow:
' Non-linear analysis of resonator filters
' The creation and reading of circuits from ADS
' Dynamic adjustment of simulation parameters
' Comparison with other simulations and real measurements
' Customization of resonator characteristics
The CSC group will provide the student with samples of real electro-acoustic resonators to carry out this project
 
Overview (resum en anglès):
Electro-acoustic (EA) filters utilizing Surface Acoustic Wave (SAW) technology have become vital components in modern mobile communications and aerospace systems.
These filters consist of several EA resonators working under the piezoelectric effect. When an electric field is applied across the electrodes of the resonator, it causes a physical deformation that propagates as a mechanical wave. In comparison of electromagnetic (EM) waves, because the phase velocity of the acoustic waves is order of magnitudes smaller than the velocity of their EM counterparts, their wavelength is significantly shorter at a given frequency. This physical property allows engineers to design exceptionally high-performance filters with an incredibly small size.
Evaluating and predicting the nonlinear (NL) behavior of EA resonators is crucial, as their inherent nonlinearities generate harmonics that cause signal quality degradation. The characterization is especially crucial today given the strict performance standards demanded in most applications.
To address this issue, this project helps characterizing this type of resonators, more specifically, the NL behavior corresponding to the second and third harmonic frequencies, allowing the simulation of complex circuits including EM simulations of the paths connecting them.
By means of a co-simulation methodology between Python and the commercial software Keysight ADS, this work implements the nonlinear lumped model methodology developed by the UPC's CSC research group, while taking advantage of the commercial software's advanced circuit simulation capabilities. Essentially, the linear simulations of the complex circuit are performed in ADS, whereas the NL simulations are performed in Python.
The developed ADS-Python application allows designers to rapidly evaluate harmonic responses and inspect each individual resonator contribution to locate the most dominant sources of nonlinearities. Furthermore, this work introduces two structural expansions to the workflow: an automated layout geometry generation feature ready for ADS' Momentum (EM) simulator for co-simulation purposes, and a top-level tuning schematic that pairs Touchstone measurement files with a tunable Butterworth-Van Dyke resonator model, enabling users to fine-tune the acoustic parameters using ADS' native Tuning Tool.
Qorvo Inc., a leading company in the semiconductors and EA filter sector that collaborates with the CSC group at UPC, has provided a set of on-wafer EA structures of LT-SAW technology to be nonlinearly characterized. These EA structures have been analyzed on this project.
The developed framework is also connected with FILTRAW®, a software tool for synthesizing and designing EA filtering structures product of a previous collaboration of the group with the European Space Agency (ESA), closing the loop between the mathematical filter synthesis and the filter final simulation including EM layout simulations and the NL response of the whole circuit.

© CBLTIC Campus del Baix Llobregat - UPC