Networks, Joint Optimization and Source Coding (ROCS)


The themes of the Networks, Joint Optimization and Source Coding (ROCS) team cover a wide spectrum of activities, from network aspects to the application layer. Among the major current challenges of IoT, 5G, smart cities, energy efficiency, immersive video, real-time video transport, security or vehicle networks, our work focuses on two axes: network issues and improving video services. More specifically, in various contexts such as wireless, vehicular or 5G sensor networks, we model, optimize and offer new methods in the form of algorithms or protocols for problems such as routing, collection and dissemination of information, mobility management, energy saving. In addition, we are developing original estimation and distributed control approaches for networked systems. Improving video services involves improving the performance of standard video coders but also designing new emerging formats. For a better quality of user experience, we are developing new subjective visual quality assessment protocols as well as new objective methods for predicting this quality.

Axis 1


Our activities generally aims at fulfilling the agility and performance requirements of telecommunications networks. They are part of the 5G perspective, which advocates access anywhere and anytime.

  • Optimizing wireless networks: This involves taking into account the new uses of wireless networks for the Internet of Things, smart cities or industry 4.0. We address issues as diverse as communications in white areas, optimization of dense networks, the use of new physical features and their impact on the network control plan as well as security in the Internet of Things.
  • Virtualization, Slicing, SDN: We are also interested in slicing and in particular, in the allocation of hardware resources shared by different virtual networks under quality of service constraints and elastic traffic demand. The goal is to design demand models and control algorithms to optimize the use of physical resources by virtual resources.
  • Energy efficiency: Energy efficiency in networks is one of the major challenges identified for 5G. We are interested in the evaluation of the energy efficiency of techniques such as coordination or Massive MIMO in the context of heterogeneous networks. In addition, we are investigating the influence of these physical level techniques on the upper layers, the transport layer in particular, in order to integrate the time constraints, which may be related to mobility, coordination or variation of flows.
  • QoS of critical applications: Communication networks are increasingly used to run critical applications, such as autonomous vehicles, industry 4.0 or e-health. In this context, new protocols must be proposed for ad hoc wireless networks (sensor networks, vehicle networks) which allow compliance with strict quality of service constraints, without which it will not be possible to guarantee the proper functioning of critical applications.

Axis 2

Video coding and transmission

With the proliferation of video services, users are becoming more and more demanding regarding the quality of experience. In this context, new breakthrough technologies are emerging in order to provide a much more realistic and immersive rendering.

  • Immersive video: New breakthrough technologies are emerging to provide a much more realistic and immersive rendering, in order to create a feeling of physical presence in the scene. The so-called “light field” representations record all the light rays entering a camera. Alternatively, volumetric approaches such as point clouds store the 3D coordinates of objects and surfaces as well as their associated color. These representations require effective compression, as well as the development of new quality indicators to assess their performance.
  • Video coding based on deep learning techniques: Another important trend is the emergence of deep learning methods for video compression. These approaches make it possible to reflect on video coding architectures in a complete departure with existing standard schemes. For this purpose, we intend to use auto-encoders or generative adversarial networks (GAN) to optimize certain coding tools, such as spatial and temporal prediction, transform or entropy coding. In the longer term, it is also interesting to study fully optimized end-to-end video coding models from a large data collection.
  • Low latency linear video coding: In the context of a very low latency constraint, relevant for applications such as vehicle remote control or augmented reality, an alternative to conventional approaches is the use of linear video coders where all the operations carried out are linear. Therefore, the quality of the decoded video is proportional to the quality of the communication channel.

Head of the team

Véronique VEQUE

Professor – Université Paris-Saclay

Télécoms et réseaux – ROC

‭01 69 85 17 22

Bât. Breguet C4.10a