PostDoc - Multi-source information fusion for Semantic Communications
ref :2025-41510 | 13 Feb 2025
apply before : 30 Jun 2025
- 28 Chemin du Vieux Chêne, 38240 MEYLAN - France
about the role
Your role is to carry out post-doctoral work on ‘multi-source information fusion for semantic communications’ as part of the ANR COMSEMA project.
In the context of future networks, research is pointing to higher requirements in terms of throughput, latency, and capacity. At the same time, new societal objectives are being put forward, such as global coverage at low cost, respect for privacy and a smaller energy footprint. In this context, the subject of semantic communications is developing. The principle is to consider the nature of the data and only transmit information that is relevant to the purpose of the communication. This approach represents a paradigm shift compared with current communications, which measure their performance by binary or packet error rate metrics, rather than by the achievement of an objective on reception. By choosing this objective for optimization, semantic communications promise not only better compression of useful information, but also greater resistance to errors. To do this, current approaches [1,2] use deep learning techniques based on a single source to define the semantic projection space of the data. The embeddings therefore represent the semantics of the information. The semantic space must therefore be optimized to both compress the initial information and be robust to disturbances in the radio propagation channel. What happens when the semantic communication objective is based on several sources [3]?
The main technical hurdles to overcome are
- Automatic construction of embeddings. This involves extracting salient features and compressing the data to meet the communication objective, as well as being resistant to transmission errors and taking into account the propagation channel.
- Information fusion techniques: a priori, a posteriori or hybrid.
- The definition of metrics to assess the quality of embedding construction and the degree to which the communication objective is achieved.
- Comparison with standard transmission methods, which are not traditionally evaluated based on the communication objective.
The study will focus on multimedia content for the re-identification of individuals. A communication chain illustrating semantic communication is expected. The work carried out will be the subject of patents, publications, and presentations at conferences.
[1] D. Gunduz et al., Beyond Transmitting Bits: Context, Semantics, and Task-Oriented Communications.
[2] E. Calvanese et al., Goal-Oriented and Semantic Communication in 6G AI-Native Networks: The 6G-GOALS Approach.
[3] X. Luo et al., Multimodal and Multiuser Semantic Communications for Channel-Level Information Fusion.
about you
You have a PhD in computer science or signal processing, with a specialization in data science / telecommunications / applied mathematics.
You have the following skills:
- You have a very good knowledge of neural network machine learning, and more generally of machine learning techniques.
- You have a sound knowledge of radio signal processing.
- You have significant development experience in Python, in particular with the Pytorch library,
- You are methodical, curious, autonomous and motivated to tackle an open research topic.
- You have publications in international conferences.
additional information
Semantic communications open up a vast new field of study in the field of communications and promise to shake up the way we design our means of communication in a number of cases.
Although the optimization of embeddings for a given task is a common operation in machine learning, the addition of a communication system at the center challenges the way we define this projection space. Limited energy quantities, quantization and propagation channels are all constraints that impact on the design of an efficient latent space. What's more, we also need to optimize the way these embeddings are transmitted, with an end-to-end optimization approach.
By linking communications system performance to application performance, semantic communications not only challenge current communications models, but also pave the way for even more energy-efficient transmissions.
Finally, the successful candidate will benefit from all the Orange advantages: works council, profit-sharing and incentive schemes.
department
The ambition of the Innovation Division is to take Orange's innovation even further and strengthen its technological leadership, by mobilising our research capabilities to nurture responsible innovation that serves people, informs the Group's long-term strategic choices and influences the global digital ecosystem. We train the experts in today's and tomorrow's technologies, and ensure continuous improvement in the performance of our services and our efficiency. The Innovation division employs 6,000 people worldwide dedicated to research and innovation, including 740 researchers. With a global vision and a wide range of profiles (researchers, engineers, designers, developers, data scientists, sociologists, graphic designers, marketers, cybersecurity experts, etc.), the men and women of Orange Innovation listen to and serve countries, regions and business units to make Orange a trusted multiservice operator.
At Orange Innovation, you will be part of a research team with dual expertise in artificial intelligence and radio connectivity. You will be surrounded by researchers working on innovative and fundamental subjects for the future of telecommunications and in particular for 6G. This position is supervised by Dr. Grégoire Lefebvre at Orange Innovation in Meylan (38). This position is for one year (renewable twice).
contract
Post Doc
Only your skills matter
Regardless of your age, gender, origin, religion, sexual orientation, neuroatypia, disability or appearance, we encourage diversity within our teams because it is a strength for the collective and a vector of innovation. Orange Group is a disabled-friendly company: don't hesitate to tell us about your specific needs.
Similar offers
Orange SA
Orange Group
of our employees are proud to work for Orange
recommend Orange as a good place to work
is the candidate experience in France, in the category of companies with over 1,000 employees
Since 2011, Orange has GEEIS (Gender Equality European & International Standard) certification in some twenty countries