The Aurorasat team is part of the SIMULIA EMAG group within the SIMULIA Brand of Dassault Systèmes. This group is in charge of providing electromagnetic simulation software that enables engineers to investigate the electromagnetic properties of components or whole systems efficiently. To achieve this, we use dedicated solvers which exploit specific numerical technologies for fast and accurate simulation. Such simulation can be applied throughout the entire design process; from the inception of an idea, through the synthesis of components to meet specifications, to the analysis of the electromagnetic performance under operational conditions.

In this context, the main goal of the Aurorasat team is to develop tools for the synthesis, design and analysis of Radio Frequency components. The team is located at 3DS Valencia offices inside the Technical University of Valencia.

Role Description & Responsibilities :

The goal of the internship is the creation of a library that enables fast model generation of different resonators in cavity and planar technologies by exploiting a Machine Learning algorithm. This library can be integrated later in the synthesis tools developed by the team, thus enabling the automatic design of new RF components.

For this purpose, the student will be involved in the following tasks:

• Selection of resonators to be studied and its definition parameters.

• Selection of the most suitable Machine Learning algorithm to create the modelling library.

• Creation of resonator models with CST Studio Suite that will be used as training/testing models for the Machine learning algorithm.

• Training and testing the Machine learning algorithm to generate the model library.

Automation of all required steps to carry out the training and its testing.

Qualifications :

Student on IT Engineer, Telecommunications Engineer or Physics.

The following skills are of consideration:

• Knowledge on Electromagnetism and Signal Processing Theory

• Programming skills (Pyhton/C++ /Matlab)

• Knowledge on machine learning and deep learning techniques