Opportunistic localization scheme based on linear matrix inequality

Localization and Tracking are very interesting functionalities that can benefit a number of applications. Despite the large number of algorithms and technologies that have been proposed in this context, the literature still lacks a widely accepted solution, capable of cutting a tradeoff between service quality (i.e., localization accuracy) and device/architecture cost and complexity. In this paper, we tackle the problem from a different and rather new perspective: we investigate how the localization accuracy of nodes can be ameliorated by opportunistically exchanging localization information among nodes that occasionally happen to be in proximity. To this end, we define a simple though accuratemodel of the opportunistic interaction and we develop a technique based on Linear Matrix Inequalities and barycentric algorithm in order to localize a user node that is completely unaware of its position. We analyse the opportunistic localization performance for different settings of the design parameters as duty–cycle and error modeling taking into account accuracy, degradation in time, correlation among consecutive estimation