Description
Date depot: 23 mars 2022
Titre: Network reconfiguration and management in 6G telecommunication networks
Directeur de thèse:
Serge FDIDA (LIP6)
Encadrant :
Thanasis KORAKIS (CERTH)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Systèmes et réseaux
Resumé: The evolution of the telecommunications infrastructure to the 5th generation has enabled key novelties, allowing the flexible management, deployment, and subsequent chaining of network components as software network functions. The cornerstone for these innovations is the wide adoption of softwarized components, across the entire stack, empowered by the disaggregation of services, and the adoption of a fully endto- end cloud-native model. When combining such features, fertile ground is created for the introduction of novel services enhancing the management of beyond 5G and 6G networks. Efforts like OpenRAN (O-RAN), Software-Defined RAN (SD-RAN) and FlexRAN take advantage of such functionalities, and build on them for providing the network operators with advanced functions for the on-the-fly network re-configuration and slicing. In this manner, the network can be dynamically and even autonomously adjusted in an end-to-end manner, based on the actual load that it is experiencing, allowing the transition to self-managed and organized 6G networks.
In addition, machine learning (ML) and artificial intelligence (AI) are two areas that show intense research activity and have several applications in the next generation of networking. The application of various machine learning techniques aims at faster decision making, by predicting accurately the behavior not only of the network but also the needs of services (Service-Aware Networks).
The combination of the above technologies contributes to the evolution of telecom networks in next-generation autonomous networks (Self Organized Networks), which promise high standards in quality of service and experience.
In this doctoral dissertation, all these open topics will be studied extensively and will be tested on real infrastructure, using appropriate software (e.g.OpenAirInterface, Docker, Kubernetes, etc). Specifically, the research objectives of the dissertation are:
– Study and design of next-generation telecommunication networks (Next Generation
- 6G networks).
– Study and design of telecommunication networks based on cloud computing (Cloud- Native Networks).
– Study and design of real or near real-time network controllers (Near-Realtime Network Controller).
– Application of machine learning / artificial intelligent techniques for the reconfiguration
of the network according to the network conditions and the load it receives.
– Design and evaluation of mechanisms for dynamic adjustment and sharing of resources
for applications/users running over the network (network slicing).
– Evaluation of the above network architectures and implementations in IoT devices
(e.g. UAVs).
– Study, design, and development of Edge Computing infrastructure and optimization
relocation of services.
The stages of the dissertation are planned to be the following:
1. Bibliographic search.
2. Design and development of modern telecommunication network architecture.
3. Deployment and reconfiguration of network functions in the Cloud (cloud-native)
based on existing tools.
4. Design and development of a real-time network controller (RAN Real-time / Near
Real-time controller placement and programming).
5. Design and implementation of artificial intelligence algorithms for decision-making
based on an supervised, unsupervised and reinforcement learning approaches.
6. Evaluation of the proposed algorithms and mechanisms through iterative experiments
in real and complex environments.
7. Thesis writing.
References
1. H. Fourati, R. Maaloul and L. Chaari, ”Self-Organizing Cellular Network Approaches Applied
to 5G Networks,” 2019 Global Information Infrastructure and Networking Symposium (GIIS),
2019, pp. 1-4, doi: 10.1109/GIIS48668.2019.9044964
2. Shafin, R. Liu, Lingjia Chandrasekhar, Vikram Chen, Hao Reed, Jeffrey Zhang, Jianzhong.
(2020). Artificial Intelligence-Enabled Cellular Networks: A Critical Path to Beyond-5G and
6G. IEEE Wireless Communications. PP. 1-6. 10.1109/MWC.001.1900323.
3. Peltonen, Ella Bennis, Mehdi Capobianco, Michele Debbah, mérouane Ding, Aaron Gil-
Castiñeira, Felipe Jurmu, Marko Karvonen, Teemu Kelanti, Markus Kliks, Adrian Leppänen,
Teemu Lovén, Lauri Mikkonen, Tommi Rao, Ashwin Samarakoon, Sumudu Seppänen, Kari
Sroka, Pawel Tarkoma, Sasu Yang, Tingting. (2020). 6G White Paper on Edge Intelligence
4. V. Yajnanarayana, H. Rydén and L. Hévizi, ”5G Handover using Reinforcement
Learning,” 2020 IEEE 3rd 5G World Forum (5GWF), 2020, pp. 349-354, doi:
10.1109/5GWF49715.2020.9221072.
5. Wu, Wen Zhou, Conghao Li, Mushu Wu, Huaqing Zhou, Haibo Zhang, Ning Xuemin,
Shen, Xuemin Zhuang, Weihua. (2021). AI-Native Network Slicing for 6G Networks.
Doctorant.e: Tsourdinis Theodoros