Antennas are a crucial part of space missions and are classically separated into different categories based on their purpose: positioning, receiving order and sending parameters from/to the ground, earth observation, radar, radioastronomy, ... Even though they all appeal to the same physical phenomenon, to translate power from conductive elements to vacuum/air (or conversely), they show a huge diversity of topologies to meet a wide variety of specifications.

The antenna might be designed with a wide “field-of-view” to help, through appropriate post-processing, detecting and localize jamming devices in the GNSS bands (NESS). It can also answer the need of satellite-communication and to illuminate a precise region on the earth with high and stable intensity while minimizing power outside. It can also be designed to probe the atmosphere on very specific frequency bands and localizations to understand how pollution spreads (C²OMODO) or to evaluate the evolution of forests (BIOMASS).

This emphasizes the needs of specific antennas, whose properties need to be accurately characterized to secure the success of these missions. These characterizations classically include the measurement of the antenna radiation pattern, which describes how the antenna can focus or receive power as a function of the direction of emission or observation. The complexity of that function depends on the electrical and geometrical sizes of the antenna. Furthermore, the radiation properties of the antenna are influenced by its host, a satellite for example. In this case, the complexity of the radiation pattern is dictated by the size of the whole system. As an example, characterizations have been done for the MMX mission in the case of the rover not being able to open its solar panels once landed on Phobos, to evaluate if it would still be able to communicate with the relay orbiting the Martian moon.

The difficulties of radiation pattern characterization are significant and ways to mitigate them are investigated for as long as antenna measurements have begun, from the 30s using planes to nowadays inside anechoic chambers of various topographies. In particular, evaluating the radiation pattern requires potentially high and even unreachable number of measurement points on a surface surrounding the antenna. In practice, a probe must move around this scanning surface to evaluate the transmitted signals before deducing the field radiated by the antenna or system under test. Recent developments in computer science, machine learning, and optimizations allowed for a wide panel of new techniques to reduce this number of required data. Many of them rely on physical properties of electromagnetic fields radiated by antennas (such as regularity or symmetries), expansion coefficients in appropriate function basis, equivalent current representations, and so on.

However, most of the available works have focused their interest on the spatial distribution of the sampling points (the position of the probe on the scanning surface) and the aforementioned techniques are applied for each measured frequency. Recent works have shown that the frequency axis holds some information that can be fostered to improve data reduction techniques, in accuracy or by allowing further undersamplings. In the case of active antennas, where the radiation pattern can be reconfigured (as in the well-known Starlink terminal to follow satellites movements to maximize data transmission), this axis could be replaced by the parameters used to achieve the reconfiguration.

The goal of this PhD thesis is to bring new solutions and post-processing methods to further build on non-spatial measurement axis (frequency, antenna configuration, …) to reduce reliably and efficiently the number of required data to achieve the characterization of antenna radiation patterns.

The thesis will take place in Rennes at the IETR laboratory with regular supervision from the CNES team. Measurements will be carried out in the IETR M²ARS platform.

=================

For more Information about the topics and the co-financial partner (found by the lab!); contact Directeur de thèse - Renaud.Loison@insa-rennes.fr

Then, prepare a resume, a recent transcript and a reference letter from your M2 supervisor/ engineering school director and you will be ready to apply online before March 13th, 2026 Midnight Paris time !