Section: New Results

Agile Radio Resource Sharing

This axis addresses the challenges relative to the network perspective of software radio. While the two other axes work on the design of the software radio nodes, we focus herein on their coexistence in a multi-user communications perspective. We are first interested in theoretical limits of some reference scenarios where trade-offs between spectral efficiency, energy efficiency, stability and/or fairness are analyzed. Our research activities are further driven by applicative frameworks. We focused on radio access networks with new results on energy efficiency-spectral efficiency trade-off in LTE networks and multi-band CSMA strategies in Wifi networks. We also studied pure random access and success probabilities for the challenging ultra-narrow band (UNB) technology of SigFox. Lot of efforts has been put on body area networks [8] with deep studies on positioning strategies and distributed decisions and information gathering. As mentioned above, our research follows three objectives:

• Establishing theoretical limits of cooperative wireless networks in the network information theory framework.

• Designing MAC procedures, coding and signal processing techniques for optimal transmissions (e.g. interference alignment).

• Developing distributed mechanisms for distributed decision at layer 1 and 2, using game theory, consensus and graph modeling.

Theoretical limits from information theory

The group strengthened his activities from a formal perspective in the framework of network information theory as initiated with the recruitment of Samir Perlaza and the sabbatical of Jean-Marie Gorce at Princeton University in the group of Prof. H. Vincent Poor. The first scenario is devoted to cellular networks with a random distribution on base stations. The main contribution concerns the broadcast channel (BC) generalized to a continuum of users. The second scenario concerns the interference channel (IC) and the main contribution is relative to the characterization of the Nash stable region for the interference channel with noisy feedback.

Broadcast channel with a continuum of users in a typical cell

The theoretical Energy efficiency-Spectral efficiency Pareto optimal front in a typical cell has been evaluated by associating stochastic geometry (Poisson point processes, PPP) and information theory.[21] , the broadcast channel is extended to a continuum of users. We derived the theoretical uniform achievable rate with superposition coding principles. We show the potential gain of superposition coding techniques compared to the conventional time sharing. These results are however limited to Gaussian channels and the extension to the vector Gaussian channel is still under investigation. The PPP modeling for multi-cells has been also introduced as well as the price of interference management.

Interference Channel with feedback

The decentralized interference channel (DIC) with noisy feedback has been analyzed. In [31] , all the rate-pairs that are achievable at a Nash equilibrium (NE) in the two-user linear deterministic symmetric decentralized interference channel (LD-S-DIC) with noisy feedback are identified. A second result provides closed form expressions for the PoA, which allows the full characterization of the reduction of the sum rate due to the anarchic behavior of all transmitter-receiver pairs. The price of anarchy (PoA) and the price of stability (PoS) of the game in which transmit-receiver pairs seek an optimal individual transmission rate are fully characterized in [9] . In particular, it is shown that in all interference regimes, there always exists at least one Pareto optimal Nash equilibrium (NE).

Coding, signal processing and MAC procedures for optimal transmissions

Implementation

While theoretical studies provide interesting insights about potential gain and limits of cognitive networks, the achievable efficiency may depend on practical issues related to quantization, synchronization and real-time processing limits. We developed the CortexLab facility offering a reproducible environment for fostering the validation of cooperative communication schemes. The first demo has been presented at the Infocom conference [28] and also at the Melbourne Greentouch meeting. We also contributed to the implementation and analysis of a cognitive transceiver for opportunistic networks [ref Maso JWCN]. The work first focused on a previously introduced dynamic spectrum access (DSA) - cognitive radio (CR) solution for primary-secondary coexistence in opportunistic orthogonal frequency division multiplexing (OFDM) networks, called cognitive interference alignment (CIA). The implementation is based on software-defined radio (SDR) and uses GNU Radio and the universal software radio peripheral (USRP) as the implementation toolkit. The proposed flexible transceiver architecture allows efficient on-the-fly reconfigurations of the physical layer into OFDM, CIA or a combination of both.

Interference alignment

In the framework of Greentouch, we studied interference alignment as a mean for improving the EE-SE tradeoff in cellular networks [43] . We combined theoretical studies with stochastic geometry and simulations to show the potential interest. We are also developing a demo with CorteXlab enhancing the IA capability from a real perspective.

Multiband MAC

In collaboration with CEA-Leti, we studied MAC strategies for multiband systems. The main idea is based on exploiting the multiband system as a slotted Aloha channel for the RTS/CTS initiation but keeping the total band as a whole for data transmission. We proved that this strategy outperforms classical approaches [39] , [40] , [30] .

MAC for localization

In the context of the ANR Cormoran project, we account for radiolocation experiments aiming at both indoor navigation and mobility detection applications for Wireless Body Area Networks (WBAN) [7] . We also studied the relation between the MAC protocol and ranging techniques for localization. The impact of mobility on the distance estimation between 2 nodes of a Wireless Body Area Network (WBAN) by comparing the Two-Way Ranging (2WR) and Three-Way Ranging (3WR) protocols has been proposed in [23] . We also investigated the impact of mobility on the Motion Capture applications [22] .

random access in Ultra-narrow band networks

Ultra narrow band (UNB) transmission is a very promising technology for low-throughput wireless sensor networks. This technology has already been deployed and has proved to be ultra-efficient for point-to-point communications in terms of power-efficiency, and coverage area. We studied the scalability of UNB for a multi-point to point network. In particular, we proposed a new multiple access scheme: random frequency division multiple access (R-FDMA) and studied the impact of the induced interference on the system performance in terms of bit error rate and outage probability [20] . We also analyzed the system performance in terms of bit error rate and outage probability [37] .

Distributed decision mechanisms

Distributed decisions appear in many situations in the wireless world. Resource allocation, power management or relaying techniques are all expecting distributed decisions. To avoid strong coordination, distributed mechanisms inspired e.g. by game theory or consensus algorithms are appealing. Some of the results obtained below also rely on information theory but with a more important focus on algorithms and decision processes when several pairs of wireless transceivers are willing to simultaneously transmit in the same environment.

Cognitive radio networks

The problem of joint channel selection and power control is analyzed in the context of multiple-channel clustered ad-hoc networks in[ref Rose [3] , i.e., decentralized networks in which radio devices are arranged into groups (clusters) and each cluster is managed by a central controller (CC). The problem is modeled by a game in normal form in which the corresponding utility functions are designed for making some of the Nash equilibria (NE) to coincide with the solutions to a global network optimization problem. A second scenario has been considered where multiple source-destination pairs communicate with each other via an energy harvesting relay [5] . The focus was put on the relay's strategies to distribute the harvested energy among the multiple users and their impact on the system performance. Specifically, a non-cooperative strategy that uses the energy harvested from the i-th source as the relay transmission power to the i-th destination is considered first. An auction based power allocation scheme is also proposed to achieve a better tradeoff between system performance and complexity.

Distributed decisions and consensus in MANETs

In the large research area of wireless body area networks, cooperative applications involving several users is attracting strong interests. This cooperation may target a simple information exchange or even some cooperative decision such as swarm coordination. We considered in [26] such a swarm of users moving in a common direction and we are interested in the mechanisms allowing to propagate and share some common information. We extend and improve a previous algorithm derived as a max-consensus approach. We describe a complete experimental setup deployed during a real bike race with 200 runners.