PhD students' day — CITI lab


A parallel unbalanced digitization architecture to reduce the dynamic range of multiple signals

by Mathieu Vallérian

Download slides
Picture of Mathieu Vallérian

Details

Socrate team
3rd year

Keywords

Radio front-end
Multiple signals digization
Near-far problem

Advisors

Guillaume Villemaud
Tanguy Risset

Partner

Orange Labs


The technologies employed in urban sensor networks are permanently evolving, and the gateways of these networks have to be regularly upgraded. In order to facilitate this upgrade, the signal processing should be done as much as possible digitally, as it is done in Software-Defined Radio. The propagation conditions strongly differ when considering several applications. Then, the signals can be received by the gateway with very different features, mainly in terms of spectral occupancy or of power level. Digitizing two signals together with a high difference of power may be problematic since the required ADC resolution varies with the power ratio of the signals.

A previous work showed that under realistic conditions, the signals dynamic range can be up to 100 dB, leading to a resolution of 21 bits. As digitizing the signals using this resolution is not affordable with today’s ADCs, this works aims at reducing the dynamic range of incoming signals using a two-antennas architecture. The basic idea of this architecture is to digitize separately the strong and weak signals on two branches. The strong signal is digitized on the first branch and is then attenuated on the second branch, allowing the digitization of the weaker signals.


Acknowledgements


CITI lab   INSA Lyon   Inria   Rhone Alpes   Rhone Alpes ARC   Rhone Alpes ARC 7   LeFabShop   Pulsalys