Analytical performance analysis of wireless systems is a challenging task in complex propagation environment, i.e. including fading, shadowing and path loss. In particular, cellular operators are interested in outage probability in cellular environments meaning the packet error rate which can they offer with a certain reliability. Finding closed-form expressions of packet error outage taking into account transmission chain technology (MIMO, coding, modulation), channel (fading, shadowing, interference) is challenging and can avoid to perform long and tedious system simulation.

Resource allocation, channel access and routing in large sensor networks are still open problems. Sensor networks are generally constrained in energy and have to ensure an end-to-end reliability. I am interested in multi-objective optimization problems as the tradeoff between delay-energy-reliability in wireless-sensor networks. This problem can be tackled in a cross-layer way and can lead to propose new protocols for this kind of networks.

The idea that subscribers could cooperate in order to transmit their informations in a wireless system was introduced at the begining of the millenium with the seminal work of Dohler on virtual antenna array and Laneman in distributed space-time block codes. A huge research effort has been granted from academic and industrial world to provide performance bounds (e.g. capacity) of cooperative information transmission and is still a hot topic in wireless communication. My work in this field is related to resource allocation in multi-user cellular systems, when users can cooperate between them or with fixed relays in the network in order to save energy.

Recently, I have been interested in the orbital angular momentum (OAM) property of the EM field. This property, associated with the non-equiphase distribution in the space of the electric field can be used to create orthogonal channels for the transmission of the information. This property can be generated using a circular antenna array at the transmission side [Thide2007]. In free space propagation condition, it has been proved that OAM modes were a subset of MIMO decomposition [Edfors2012]. However, OAM is a physical observable in its own right, clearly independent from polarization. If it is true that it does not create another degree of freedom (at least with current way to generate and detect OAM), it remains nevertheless interesting for potential applications, e.g. RADAR, free space optical communications, RF wireless communications with particular pre-coding.

CURRENT RESEARCH PROJECTS

Industrial projet gathering several companies in the video broadcast industry. Cloudcast aims at deporting main baseband processing of a digital video chain in a cloud with only RF components remaining in transmitters. The bottleneck of this concept is the huge I/Q data rate requires on the fronthaul links. For economical reasons, optical fibers cannot not always be used for the fronthaul links. Hence, rate compression is needed in order to limit the I/Q rate on the fronthaul. Optimal compression schemes jointly designed along channel coding in order to be robust to errors on IP network are sought in this project.

Bilateral project with Orange Labs Grenoble, France, funding a PhD student. The projet aims at defining signal processing techniques enabling massive non-orthogonal access for IoT networks.

Project granted by the french research institution (ANR). The aim of the project is to provide achievable regions and converse for dense wireless networks of type Internet of Things. The main characteristics for these networks are the high number of connected objects, the intermittent nature of communications and the small packet size. While the latter invalidates the use of the asymptotic Shannon capacity as a performance indicator, the impulsive nature of interference also invalidates the Gaussian assumption of interference. This project aims at characterizing the fundamental performances, in terms of error probability, latency and nodes density of future IoT networks.

FORMER RESEARCH PROJECTS

Project granted by the french research institution (ANR) via the « investissements d’avenir » initiative. The aim of the project is to set the energy consumption of networks proportional to their loads.

Industrial project with orange labs Rennes. The aim of the project is to study promising communication techniques likely appealing to improve the energy efficiency of communication schemes like time reversal and massive MIMO, HARQ techniques.

Regional project granted by the European University of Brittany. The aim of the project is to investigate different distributed coded scheme in relay networks and their influence on the tradeoff between packet delivery and energy consumption with HARQ as well as resource allocation in multi-user cooperative networks.

National project granted by the French institution ANR. The goal of the project is to study algorithms from PHY to Network layer for heterogeneous ad-hoc networks for civil security applications. In a crisis scenario like chemical accident or seismic disaster, the intervention groups have to coordinate the rescue without any available communication infrastructures, potentially destroyed in the disaster. The rescues have to communicate video, data, voice leading to heteregeneous data flows in the network with different priority and QoS requirements. This important particularity leads to propose new signal processing algorithms as well as routing and resource allocation protocols.

In this project, the impact of accurate PHY layer models for higher network layers has been addressed. In particular, protocols exploiting multi-antenna transceivers for efficient resource allocation algorithms have been addressed.

The aim of the project was to propose a proof of concept of software radio terminals and their impacts on the deployment of future wireless systems.

COLLABORATIONS

Orbital Angular Momentum

• Christian Brousseau (IETR / Université Rennes 1)

Cognitive Radio and Machine Learning

• Christophe Moy (Supélec Rennes)

Cooperative Communications

• Olivier Berder (IRISA / ENSSAT)
• Charlotte Langlais (Lab-STICC / ENSTB)

Performance analysis in wireless communication systems

• Mischa Dohler (King’s College London)
• Jean-Marie Gorce (CITI / INSA Lyon)

Resource allocation in TH-UWB systems

• Inbar Fijalkow (ETIS / ENSEA)
• Charly Poulliat (IRIT / ENSEEIHT)

STUDENTS

Former PhD students

• Navikkumar Modi (2014-2017): Machine learning and Statistical Signal Processing for Green Cognitive Radio. Research engineer at Brussels Airport, Brussels, Belgium.
• Ahmad Mahbubul Alam (2014-2017): Energy Efficiency Scaling Laws for Wireless Networks. Research engineer at Siradel, Rennes, France.
• Hiba Bawab (2011-2015): Technology convergence for multimedia broadcast in LTE and DVB systems. Research engineer at Nokia, Lannion, France.
• Samih Abdul-Nabi (2012-2015): Centralized and distributed address correlated network protocols. Associate Professor at Lebanese International University
• Mohamad Maaz (2010-2013): Resource allocations for energy-efficient cellular networks.

PhD students

• Dadja Anade Akpo (2017-2020): Fundamental bounds in finite block length regime and dense networks.
• Wissal Ben Ameur (2018-2021): Multiple access mechanisms for massive IoT deployment over 5G networks.
• Qiong Liu (2018-2021): Spatial and Energy Efficiencies for Networks with Variable Load.
• Mohammadreza Mardani (2019-2022): Correlated spatial point processes for efficient resource allocations in dense and randomly deployed networks.
• Aya Shehata (2019-2022): Data rate compression for virtualized baseband processing in broadcasting applications.

Master students

• Alvaro Artigas Gil (2015): Techniques de transmission de l’information à l’aide d’une onde EM portant un moment angulaire orbital.
• Abdullah Haskou (2013 – Currently in PhD at IETR, Univ. Rennes 1): Detection of the OAM order in electromagnetic waves.
• Mohamed Bouzian (2013): Machine learning and statistical signal processing for green radio.
• Mohamad Kanj (2013 – Currently in PhD at IRT-BCOM): Energy-efficiency scaling laws for wireless networks.
• Marwa Chami (2012 – Currently in PhD at CNAM Paris): Ambiguity suppression in SAR imaging.
• Soha Farhat (2011): Optimization of the Number of Relays in a Cooperative Wireless Network.