Wideband cooperative localization via belief condensation

Abstract

The combination of wideband transmission and cooperative techniques can enable accurate and reliable wireless localization. However, due to the nonlinear/non-Gaussian behavior in the system and measurement models, current cooperative localization algorithms either cannot capture such behavior or suffer from high computational complexity. In this paper, we establish a general framework for designing cooperative localization algorithms based on belief condensation (BC), which is an iterative process that accurately approximates the intractable distribution of nodes' positions arising in cooperation by amenable parametric ones. Our simulation results show that the proposed method can obtain high accuracies, approaching the theoretical benchmark, with an affordable complexity.