Vous êtes ici

Energy efficient context discovery in pervasive computing using ambient backscatter communications

Equipe et encadrants
Département / Equipe: 
Site Web Equipe: 
Directeur de thèse
Frédéric Weis
Co-directeur(s), co-encadrant(s)
Angeliki Kritikakou
Paul Couderc
Olivier Sentieys
NomAdresse e-mailTéléphone
+33 299 84 72 92
+ 33 2 99 84 71 85
Sujet de thèse

Teams : CAIRN et EASE

Context and topic

While pervasive computing technologies are becoming mainstream thanks to cheap and ubiquitous wireless communications, ease of interactions with these systems is still hampered by environment discovery and initialization issues: in particular, radio interfaces are not always active due to energy cost, and security context often imposes high overhead handshake/authentification protocols. An alternative approach is to use dedicated technologies for environment discovery, beside regular networking.

Environment discovery in IoT is typically implemented at different levels: lower level communication layers provides devices discovery and basic network function bootstrapping, while higher layers provide services and/or device capabilities discovery. This separation is welcomed from a separation of concerns and technology independance pespective, but also raises issues when discovery happen in dynamic or mobile context, because application level data exchange are often directly related to discovery or MAC-level interactions; for example, a location-dependant application may have to signal an event when two mobile devices are getting into the range of each other, and this condition may only last for a short timeframe.

Unfortunately, in dynamic environments discovery is an energy hungry process as it has to rely on some kind of continuous announcement and channel listening protocol, which are inherently power inefficient.

However, recent advancements in wireless communications have shown a way to avoid the energy cost associated with continuous communications : ambient backscatter consists in creating a side channel leveraged on existing, ambient, signals by modulating their reflections. This process is well known for RFID, but ambient backscattering apply the principle on ambient signals from Wi-Fi, 3G/4G, FM radios, DVB etc. [1,2]

These advancements open a wide range of perspectives: fully passive data communications with objects are possible, enabling self description with embedded data, zero-conf protocols, wake-up mechanisms for Wi-Fi or BLE, etc. it is also especially interesting for discovery protocols and opportunistic communications.

It has been shown that implementing the backscatter interface is possible with digital logic, without relying on an analog RF frontend: an FPGA implementation of passive Wi-Fi has been demonstrated by [1]. We propose to investigate efficient designs for this type of modules, in terms of logic complexity and energy efficiency, specifically to support discovery protocols with passive chips. As ranging properties are important in this context, in particular for security issues, various type of ambient backscatter signals could be considered.


The candidate for this Ph. D research should hold a CS master degree, and is expected to have a strong background in hardware architecture design and FPGA programming. Familiarity with wireless communication and RF is also expected. 


[1] Passive Wi-Fi: Bringing Low Power to Wi-Fi Transmissions.
B. Kellogg, V. Talla, S. Gollakota, JR Smith - NSDI, 2016

[2] FM Backscatter: Enabling Connected Cities and Smart Fabrics.
A. Wang, V. Iyer, V. Talla, JR Smith, S Gollakota - NSDI, 2017

Mots clés: 
pervasive computing, energy harvesting, backscatter communication, FPGA
IRISA - Campus universitaire de Beaulieu, Rennes