Description
Date depot: 1 janvier 1900
Titre: Méthodes de coordination décentralisées pour alignement de faisceaux et allocation de ressources pour la 5G
Directeur de thèse:
David GESBERT (Eurecom)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Non defini
Resumé:
Some challenges in 5G networks and beyond
5G networks will stand out from earlier 4G deployments in several important ways: (i) The need for very high throughput traffic delivery to a high density of high intensity internet users, (ii) The heterogeneity of devices that will download or upload data from/to the network, from classical human-controlled tablets and phones, to connected objects in use for smart cities and smart factories, to the connected vehicles of the upcoming intelligent transport networks (ITS)., (ii) The heterogeneity of performance metrics by which satisfaction will be measured across these diverse devices: From peak data rate, to latency, to reliability. These combine departures from 4G networks motivate a re-design of signal processing and information coding mechanisms in order to offer the required performance and adaptability.
Interestingly, the high density of devices (from things to tablet PCs to connected cars) represents both a challenge and an opportunity for system design as devices that are located closer to each other have greater chance for direct communication, hence allows for a cooperative behavior which has shown to be very useful for interference mitigation and resource allocation purposes [Gesbert2010].
Device-centric cooperation
While cooperation at the physical layer of wireless communication systems can be in principle handles in a centralized manner (i.e by concentrating all feedback data and computational capability at a unique point), this approach is inefficient as measurement data (e.g. channel feedback) typically have a short period of relevance (coherence time) and centralizing channel measurements can lead to outdated feedback which are of little use (except in certain specific scenarios). Moreover centralized coordination schemes ignore the computational and decision making capabilities of individual radio terminals/nodes.
In order to solve this problem, the cooperation methods will be considered in the form of decision making algorithms that are implemented at the devices themselves. The leading forms of cooperation envisioned for 5G will be considered:
• Beam selection and alignment [Heath2015] between transmitter and receiver (especially in mmwave bands)
• Cooperation between base stations for user scheduling
• Caching-based cooperation
In previous works, the device centric cooperation in wireless networks has been formulated as a team decision problem and some resource allocation solutions have been proposed as [kerret2013] More recently, an algorithm to efficiently design a joint precoder being robust to the imperfect information at the nodes has been proposed in [kerret2016]. Yet, the proposed solution is highly limited by the complexity and is therefore not adapted to the dense scenario formed by the factory setting.
However we expect to be able to build upon similar principle used in [kerret2016] in order to develop robust cooperation mechanisms, where robustness is defined with respect to the lack of shared knowledge of the propagation channel coefficients at the two (or more) cooperating nodes.
Links with classical team decision theory methods will also be studied[Radner1962], which are known for a long time, in particular in the control community, and are widely recognized as difficult problems. Yet, the strong development of the computation capabilities in the past decade has opened up new avenues for solving such difficult problems. In particular, Stochastic Optimization [Shapiro2014] and Machine Learning methods [Rasmussen2006] provide now very efficient tools for solving a wide range of problems.
Most of the work by the candidate will be performed within the context of 5G networks and realistic system evaluations which account for basic 5G design parameters will be carried out [Nokia2016].
Work plan
Year 1
• Bibliographic work relative to 5G cooperation methods
• Bibliographic work relative to team decisional methods
• 2 or 3 technical courses
• Creation of the MATLAB simulator and simulations of state-of-the-art schemes
• Development of the first Team Decision methods in the context of beam alignement
Year 2
• Development device centric coopeation communication approaches for 5G
• Analysis of the performance
• Presentation of results at IEEE international conferences (1 or 2, IEEE)
• Preparation of first journal paper (IEEE Transaction).
Year 3
• Selection and design of final practical schemes for the 5G mobile communication system (Beam alignment, scheduling, possibly caching)
• System level evaluation
• 1 or 2 IEEE conference paper,1 additional journal paper (IEEE Transaction).
• Writing of PhD dissertation
•
Bibliography
[Gesbert2010] “Multi-Cell MIMO Cooperative Networks: A New Look at Interference”, D. Gesbert, S. Hanly, H. Huang, S. Shamai Shitz, O. Simeone and W. Yu, IEEE Journal on Selected Areas in Communications, 2010.
[Kerret2013] “CSI sharing strategies for transmitter cooperation in w
Doctorant.e: Maschietti Flavio