Section: New Results

Flexible Radio Front-End

The contributions of members of this axis are mainly on four topics: Wake-Up Radio,Full-Duplex transceivers, SDR Gateways for Urban Networks, and Channel Estimation. In the global concept of enhancing wireless communications, those four topics are complimentary, addressing the reduction of energy consumption, the increase of throughput and/or flexibility of the transmission and the performance evaluation.

The last decades have been really hungry in new ways to reduce energy consumption. That is especially true when talking about wireless sensor networks in general and home multimedia networks in particular, since electrical energy consumption is the bottleneck of the network. One of the most energy-consuming functional block of an equipment is the radio front end, and methods to switch it off during the time intervals where it is not active must be implemented. This study [10] has proposed a wake-up radio circuit which is capable of both addressing and waking up not only a more efficient but also more energy-consuming radio front end. By using a frequency footprint to differentiate each sensor, awaking all the sensors except for the one of interest is avoided. The particularity of the proposed wake-up receiver is that the decision is taken in the radio-frequency part and no baseband treatment is needed. The global evaluation in theory and in simulation was performed, and a first testbed of this technology was fabricated.

This work studies [8] a Full-Duplex Dual-Band (FDDB) OFDM radio architecture that enables the radio transceiver to be more flexible and provides a viable radio link capacity gain. A simple but practical I/Q imbalance estimation and compensation method, based on the frequency-flat-fading behavior of the self-interference channel, is proposed. The performance of the proposed I/Q imbalance compensation method is evaluated by system level simulations conducted with ADS and Matlab. The co-simulation results show that the proposed radio transceiver could potentially increase the physical layer transmission rate by four times compared to the conventional radio link at the cost of tolerable loss of BER performance. The I/Q imbalance compensation method can effectively compensate both high and low I/Q imbalance without the problem of algorithm convergence. Application of this technique for physical layer security has already been proposed.

The technologies employed in urban sensor networks are permanently evolving, and thus the gateways of these networks have to be regularly upgraded. The existing method to do so is to stack-up receivers dedicated to one communication protocol. However, this implies to have to replace the gateway every time a new protocol is added to the network. A more practical way to do this is to perform a digitization of the full band and to perform digitally the signal processing, as done in Software-Defined Radio (SDR). The main hard point in doing this is the dynamic range of the signals: indeed the signals are emitted with very different features because of the various propagation conditions. It has been proved that the difference of power between two signals can be so important that no existing Analog-to-Digital Converter (ADC) is able to properly digitize the signals. We propose a solution to reduce the dynamic range of signals before digital conversion. In this study [28] , the assumption is made that there is one strong signal, and several weak signals. This assumption is made from the existing urban sensor networks topology. A receiver architecture with two branches is proposed with a “Coarse Digitization Path” (CDP) and a “Fine Digitization Path” (FDP). The CDP allows to digitize the strong signal and to get data on it that is used to reconfigure the FDP. The FDP then uses a notch filter to attenuate the strong signal (and then to reduce the dynamic range of the signals) and digitizes the rest of the band. Another way to relax these specifications on ADCs is an analog processing, such as companding, that should be performed before digitization. The companding technique is usually employed on one signal (and not on multiple signals that are only separated on the frequency domain). This work [36] , [29] studies three companding laws to test their efficiency in relaxing the digitization constraints with multiple signals. A µ-law, a Piecewise-Linear (PL) law and a Piecewise-Linear, Constant Gain with Offsets (PLCGO) law are tested. We have described how to use a PLCGO approach to reduce ADC's complexity, and two implementations of the compressing law are proposed.

In modern mobile telecommunications, shadow fading has to be modeled by a two-dimensional (2D) correlated random variable since shadow fading may present both cross-correlation and spatial correlation due to the presence of similar obstacles during the propagation. In our study, 2D correlated random shadowing is generated based on the multi-resolution frequency domain ParFlow (MR-FDPF) model. The MR-FDPF model is a 2D deterministic radio propagation model, so a 2D deterministic shadowing can be firstly extracted from it. Then, a 2D correlated random shadowing can be generated by considering the extracted 2D deterministic shadowing to be a realization of it. Moreover, based on the generated 2D correlated random shadowing, a complete 2D semi-deterministic path loss model can be proposed. The proposed methodology [5] can be implemented into system-level simulators where it will be very useful due to its ability to generate realistic shadow fading.

[23] presents the first implementation on software defined radio nodes in the large scale testbed CorteXlab of a radio link estimation technique based on OFDM transmissions. The purpose of this large scale testbed is to offer to the whole scientific community an open tool to test new techniques for multiuser , cooperative and cognitive radio networks in a controlled environment. As the experimentation room was defined in order to offer reproducible measurements, it is important to be able to characterize each radio link between all transceivers. Therefore, we present here the development of a channel sounder directly implemented on the software radio nodes. This paper presents the first implementation on software defined radio nodes in the large scale testbed called CorteXlab of a radio link estimation technique based on OFDM transmissions. The purpose of this large scale testbed is to offer to the whole scientific community an open tool to test new techniques for multiuser , cooperative and cognitive radio networks in a controlled environment. As the experimentation room was defined in order to offer reproducible measurements, it is important to be able to characterize each radio link between all transceivers. Therefore, we proposed the development of a channel sounder directly implemented on the software radio nodes.