The availability of positional information has become an integral part of our daily lives, and is critically relied upon by industry and governments. In cases where traditional GPS-based positioning fails, new solutions have been proposed, relying on cooperation among peer devices. In theory, such cooperative positioning has great promise to provide position context in GPS-challenged environments. Contrary to GPS, much of the research on cooperative positioning has relied on highly idealized assumptions, resulting in elegant, yet highly non-robust algorithms. The goal of this research proposal is to (i) tackle several such robustness issues in a rigorous and systematic fashion, based on Bayesian inference; (ii) implement the resulting novel algorithms on a prototype testbed. Our theoretical work will allow networks to operate in unknown environments, in the presence of limited false information. Our research can be extended to other problems in robust Bayesian inference, and network processing. For the purpose of validation, our theoretical work will be complemented by a proof-of-concept testbed, which will be the first of its kind. This testbed will enable commercial opportunities and novel technologies.
Biträdande professor vid Chalmers, Elektroteknik, Kommunikationssystem, informationsteori och antenner, Kommunikationssystem
Finansierar Chalmers deltagande under 2011–2013