Moteur de recherche d'offres d'emploi CEA

Deep Neural Network testbed for 5G Radio Resource Management H/F

Vacancy details

General information


The French Alternative Energies and Atomic Energy Commission (CEA) is a key player in research, development and innovation in four main areas :
• defence and security,
• nuclear energy (fission and fusion),
• technological research for industry,
• fundamental research in the physical sciences and life sciences.

Drawing on its widely acknowledged expertise, and thanks to its 16000 technicians, engineers, researchers and staff, the CEA actively participates in collaborative projects with a large number of academic and industrial partners.

The CEA is established in ten centers spread throughout France



Description de l'unité

Within CEA (Alternative Energies and Atomic Energy French Commission), the Laboratory for Electronics & Information Technology (CEA LETI) works with companies to increase their competitiveness through technological innovation and transfers. CEA LETI is focused on micro- and nano-technologies and their applications, from wireless devices and systems to biology and healthcare or photonics. Within CEA LETI, the proposed internship will take place at the Wireless Broadband System Laboratory, involved in research activities related to 5G and beyond-5G radio communication systems. For the past few years, the laboratory has been particularly active in the AI field and its implementation in future communication systems. The lab is a member of the ETSI “Experiential Networked Intelligence” Industrial Standardization Group, focusing on the deployment of AI in future networks. It has strong expertise in cellular networks, resource management, heterogeneous networks, and optimization. The lab has lead or taken part in many R&D French, European, and international collaborative projects and has strong links with ICT industry. CEA LETI is also part of the multidisciplinary institute in artificial intelligence (MIAI), a French multidisciplinary project that involves a local network of well-known researchers working on AI.

Position description





Job title

Deep Neural Network testbed for 5G Radio Resource Management H/F


Internship offer in the domain of telecommunications, 5G radio resource management, and reinforcement learning.

Contract duration (months)


Job description

To support a multitude of services within a common framework, the fifth generation (5G) wireless networks will rely on a dense deployment of base stations that exploit multiple radio access technologies. One of the main challenges engineers have to face is the optimization of the Radio Resource Management (RRM) with respect to the dynamic characteristics of the radio environment.

Prior to starting the communication, to maximize the overall network performance while satisfying the quality of service of each user, it is necessary to properly associate user devices and base stations. In networks with a high density of users and access points, suboptimal associations may lead to excessive cell loads, increased interference, degraded network performance, and higher costs. Finding the optimal association is a combinatorial hard problem whose complexity grows exponentially with the number of base stations and user devices. Conventional solutions either are computationally too complex or induce large signaling overhead, which increases energy consumption and end-to-end latency.

To handle this complexity, we have developed an AI-based association framework that involves deep neural networks (DNNs), based on deep reinforcement learning, where each user independently learns and adapts its association policy, based on its own service requirements and in order to optimize the overall network performance.

In this context, we target to implement the proposed theoretical framework on a testbed to demonstrate its benefits. The intern will:

  1. Study and understand the concepts behind the proposed approach (DNN, 5G RRM, user association to network access points)
  2. Develop a small-scale wireless network demonstrator based on Android systems or Raspberry Pi.
  3. Deploy in the testbed the AI-assisted user association policy and design the communication protocol to enable information collection and exchange between the different network nodes.

The performance of the proposed implementation will be validated by comparison with state-of-the-art solutions.

Methods / Means

Python, Tensorflow

Applicant Profile

We are looking for a master/engineering student in his/her BAC+4/5 year.
The candidate must be passionate about telecommunications and be willing to work on a topic that involves both programming and mathematical/theoretical skills.

Knowledge of Python and Tensorflow is necessary; previous experiences with Android-Java programming are welcome.

Position localisation



Job location

France, Auvergne-Rhône-Alpes, Isère (38)