PhD – ” AI based network reconfiguration for QoS enhancement in 6G networks ” F/M

about the role

The next generation of mobile networks should enable the seamless operation of Industrial IoT systems, while reducing energy consumption.

In this context, ultra-reliable low latency communications (URLLC) are a key enabler for these applications that simultaneously require strict constraints in terms of low latency, a high level of reliability, and support for a massive number of connections. Therefore, meeting all these requirements represents a challenge for the future generation of mobile networks and requires learning methods for prediction and for adaptation of the network to the needs of the requested service in order to guarantee the necessary performance when and where is required.

The idea of the thesis is to design intelligent networks capable of transporting time critical information and adapting to the needs of applications. The objective is to propose the network configuration which achieves the best compromise between quality of service (QoS) and energy consumption.

In automotive and industrial automation applications, QoS predictive mechanisms help predict potential QoS changes. If the predicted QoS is not satisfactory, AI-based network reconfiguration is required to achieve the target QoS. We are interested at the RAN level in the integration of AI in the configuration of the radio interface (exploitation of multi-connectivity, allocation of radio resources) to improve QoS. In particular, reinforcement learning will be the preferred tool for learning the optimal configuration of the network. However, a big challenge arises: the strict constraint on reliability implies that packet loss is a rare event, which means slow learning convergence. A model-assisted approach will be proposed: prior knowledge contained in mathematical models is integrated into AI algorithms to reduce the amount of training required.

The PhD will rely on learning, optimization and performance evaluation techniques to develop algorithms to obtain optimal network configurations:

–State of the art of techniques that use prior knowledge of system models to accelerate learning convergence;

–Develop learning algorithms to wait for the target QoS that exploit network and traffic patterns in evaluating the reward that results from the action, meaning accelerating learning convergence;

–Provide optimal policies for URLLC traffic that minimize delays by guaranteeing QoS and energy constraints;

–Evaluation of the performance of the proposed policies.

about you

–Master in telecommunications or applied mathematics

–Learning skills(reinforcement learning, Markov decision processes, …)

–Skills in modeling networks and systems(probabilities, random processes, queues, …)

–In-depth knowledge in the field of fixed and mobile networks (architectures and protocols, virtualization, SDN, 5G, quality of service management, resource allocation, etc.)

–Programming and IT development skills (Python, Matlab, C / C ++, …)

–Fluency in written and oral English

additional information

The doctoral student will join Orange, an international group, expert in the information and communication technologies sector. In particular, he will join the MORE team made up of application and research engineers in applied mathematics and IA within the Data-AI department. This team is in charge of developing models and techniques to optimize the use of resources (in the broad sense) and to assess network performance. He/she will work in an international context and will have the opportunity to contribute to research projects with external partners from the academic and industrial environment.

department

Orange Innovation brings together the research and innovation activities and expertise of the Group’s entities and countries. We work every day to ensure that Orange is recognized as an innovative operator by its customers and we create value for the Group and the Brand in each of our projects. With 720 researchers, thousands of marketers, developers, designers and data analysts, it is the expertise of our 6,000 employees that fuels this ambition every day.

Orange Innovation anticipates technological breakthroughs and supports the Group’s countries and entities in making the best technological choices to meet the needs of our consumer and business customers.

Within Innovation, you will be integrated into a research team at the forefront of innovation and expertise on the networks of the future.

MORE (Mathematical Models for Optimization and Performance Evaluation) team is responsible for research in applied mathematics and AI, to develop models and techniques for resource optimization and infrastructure performance evaluation network. Research activities are carried out to identify technological weaknesses and design analytical tools for the design and planning of these infrastructures.

contract

Thesis

Read Full Description