PhD Offer in the framework of the MSCA Doctoral Network "GLITTER": Advanced Bayesian analysis and optimisation

GNSS-R is a technique to carry out Earth observation based on reflections on the ground (or sea, or ice) of signals originating from GNSS (Global Navigation Satellite System) signals. The proposed project consists of educating a new generation of experts, at doctoral level, able to bring a qualitative leap to this technology. The scientific and technological goal consists of developing such systems based on a synchronized constellation of Cubesats. An important advantage of this arrangement is the very low cost of cubesats and the possibility to increase resolution based on beamforming from the satellites. Ground truth, as well as some of the methods, will originate from near-field radar technology. This will require further research on all segments of GNSS-R technology and beyond: launching and adjustment of cubesat formations, RF synchronization, interferometry between moving platforms, calibration of RF front-ends, ground testing making use of drones, cubesat systems, on-board
processing, data transfer and analysis, translation into ground truth and into predictions important for climate change studies and for optimal territory management. The project may also benefit to other technologies making use of interferometry, such as radioastronomy and phased array based communications. It is also expected to assist industry segments making use of GNSS signals, such as precision agriculture, forestry and sea and land management.

Training programme for recruited researchers:

GLITTER offers a rich inter-sector training program, with private and academic partners, recognized in all segments of satellite-based Earth observation at microwaves. The program has been designed taking into consideration the "triple i" aspects: international, inter-sectoral, and inter-disciplinary, thanks to the different competences brought by each partner of the network as well as lecturer outside of the consortium with complementary competences.

Each Doctoral Candidate (DC) will benefit from an individual and customized training program, established in the Personal Career Development Plan (PCDP) that offers a combination of research specific and transferable skills. First, the core training, often but not always local, with order of 5 ECTs on specialized topics, close to a given DC personal project, to allow them to acquire a deep understanding of their subject. Second, the network-wide training will be offered by the consortium during the whole project life cycle through 4 training schools and 3 workshops. The workshops will include practical system-based development., Please provide the following four items, all in pdf format:

• Your CV (1 to 2 pages in pdf format)


• A list of grades/diplomas, certificates


• A list of referees (three at least)


• A motivation letter


The four documents should be preferably assembled in a
• .zip file with a file name that starts with DC12.


"Commitment to Equality and Diversity:
The GLITTER consortium is committed to promoting equality, diversity, and inclusion in all aspects of the project. We particularly encourage applications from women, underrepresented minorities, and individuals with diverse backgrounds and perspectives, aiming to create an
inclusive research environment that reflects the diversity of our society. Our policies ensure equal opportunities for all, regardless of gender, ethnicity, disability, age, sexual orientation, or religion."

Selection Procedure:

A first selection will be made on basis of CV, list of grades, certificates, list of referees and motivation letter.

Potential candidates will be invited and interviewed by a selection committee. comprising the work package leaders, who will interact with the Coordination team and the director of the local graduate school. Candidates will be requested to present a short talk.


• MSc degree in Physics, Engineering, Mathematics, Computer science or related fields.


• Solid mathematical background, outstanding academic records, and excellent communication skills in oral and written English.


• Strong Python programming & computing skills


• Experience with Bayesian analysis, machine learning & high-performance computing is desirable but not essential., PhD or equivalent
  
  The next decade will see a revolution in satellite configuration and data processing, with the launch of large constellations of small satellites coincident with ever more demand for earth and sky monitoring. This PhD project aims to develop and apply the cutting edge of Bayesian analysis and machine learning to the optimisation of satellite configurations for GNSS-R. Combining the data science expertise of Dr Handley's Cambridge lab and PolyChord Ltd company with the expertise of the GLITTER network, the PhD candidate will build and deploy the next-generation of analysis tools for the future of satellite earth observation.


Projects will include:

• Developing tools for optimising cubesat configurations using PolyChord algorithm as a global optimiser under arbitrary nonlinear constraints


• Exploring the application of Bayesian parameter estimation, model comparison and tension quantification in GNSS-R problems


• Applying the cutting edge of Likelihood-free inference techniques for simulation-based predictions


• Exploring the use of machine learning emulators for enhancing the above processes in speed and accuracy


The PhD candidate will join a multidisciplinary team and will be supervised by Dr Will Handley alongside data scientists at PolyChord Ltd (a spin-out company from the University of Cambridge) and researchers from Dr Will Handley's group based in the Kavli Institute for Cosmology, University of Cambridge. The PhD candidate will benefit from the GLITTER events in order to further improve technical and complementary skills: four one-week training schools, and three workshops. In addition, the PhD candidate will interact with CSIC (Barcelona) and SYNTONY (Toulouse) through 3 month secondments working with Dr Cardellach and Dr Carrié.

PolyChord Ltd is a spin-out company (SME) from the University of Cambridge, transferring technology from academia to industry, with current applications including rail monitoring, sensor placement optimisation, protein folding, and battery design. It has a long-term interest in the application of it's know-how and IP to the space sector, and the PhD candidate will be a key part of this process.

Each host organisation will appoint the successful applicant under an employment contract with a very competitive salary. The fellow is expected to join their host organisations starting from 1^st October 2024 (estimated time). Additional funding for participation to courses, workshops, conferences, etc. is ensured.

The funds for this post are available for 36 months.

The Fellowship is offered in conjunction with a PhD position, subject to the Fellow satisfying the University's admission requirements for PhD Study. Therefore, candidates will need to submit a formal application to the University of Cambridge for a PhD in Physics.

EU eligibility criteria for candidates:

The Doctoral Candidate (DC) may be a national of a Member State, of an Associated Country or of any Third Country.

Researchers must be doctoral candidates (no doctoral degree at the date of recruitment)

The DC must not have resided or carried out main activity (work, studies, etc.) in the country of the host organisation for more than 12 months in the 3 years immediately prior to recruitment. (Compulsory national service, short stays such as holidays, and time spent as part of a procedure for obtaining refugee status under the Geneva Convention are not considered)