SRP-Onderzoekszwaartepunt: Center for model-based system improvement - From Computer-Aided Engineering to Model Aided Engineering

Designing complex systems uses abstract models to systematically refine the subsystems. In this process, the mathematical models are ever increasing in complexity when going deeper into the subsystems, requiring more and more simulation time. A solution to this problem is the use of so- called meta models which capture the subcircuit’s behaviour using a cheap to evaluate model that is tuned using lengthy but accurate simulations. This is also known as space mapping as it is mapping one design space onto another one.

This program will extend and apply the space mapping towards statistics based (system identification) models to include noisy data. This enables the usage of state-of-the-art nonlinear analysis techniques which represent the set of excitations of a system (such as a communication system with complex modulated signals) as a stochastic process with a large number of well- chosen periodic excitations to determine the system’s nonlinear behaviour.

Including noisy measurements in the design flow can result in time-consuming and expensive measurements of the (sub)systems. The developed meta- and system identification modelling will then be used to combine measurements with different levels of accuracy (the multi-fidelity modelling) to reduce both the measurement time and the necessary (re)calibration time.

To develop and demonstrate the usefulness and general applicability of the unified multi-fidelity space-mapping-based design flow, this design flow will be applied to a broad variety of state-of- the-art electronic designs including

• single photon avalanche detector (SPAD) system design, realisation, measurement, and characterization using statistical models;
• 6G mm-wave (140GHz) MIMO transceiver design, realization, and measurement with the heterogeneous integrations of electronics and antenna design, requiring circuit-level, thermal
and EM-field simulations;
• measurement and calibration techniques required for measuring the designed transceivers over the wide variety of excitations and modulation standards that are specified by future wireless communication standards such as 6G.