Theoretical and simulative modeling of noise and complex systems
The research group has been involved during the last 20 years in the pioneering development and implementation of efficient numerical approaches for the physics-based noise modeling of semiconductor devices, both in the case of stationary (small-signal noise) and cyclostationary (large-signal noise) operation. The research group was a world leader in the implementation of noise analysis in Technology CAD (TCAD) tools through Green's function techniques applied to discretized transport models.
The algorithms developed by the group are currently implemented in all the commercial physics-based device simulation tools, and in many academic codes. The experience gained was also exploited to develop compact noise models for MOSFETs and bipolar transistors.
Concerning complex systems, the harmonic balance technique has been exploited to find limit cycles of autonomous and forced nonlinear dynamical systems, and to assess their stability by means of Floquet perturbation theory. The main result is the identification of efficient and accurate frequency-domain methods for the detrmnation of the DFloquet multipliers of the limit cycle, thus allowing for a detailed assessment of circuit stability. The approach is currently applied to the study of stability properties of nonlinear circuits such as microwave power amplifiers.
Floquet theory is also the basis for a rigorous assessment of noise in oscillators: exploiting a novel theoretical analysis and the numerical algorithms previously discussed, the group recently proposed a complete nonlinear perturbative characterization of the oscillator noise spectrum, including phase and amplitude noise and their correlation.
The research was funded by several national (PRIN, FIRB and Azione Integrata Italia-Spagna) and european (ESPRIT and HCM) projects, and supported by a cooperation with SILVACO DATA SYSTEMS (USA).