Robust Distributed Estimation for Localization in Lossy Sensor Networks

摘要: Abstract: In this paper we address the problem of fault resilient estimation for large-scale systems, where measurements are possibly corrupted due to faults low-cost sensors. As a toy application, consider localization in Sensor Networks (SN). We propose distributed solution based on recently developed generalized descent algorithm. To cope with real-world applications, algorithm is suitable an asynchronous implementation and numerically robust non ideal communications, i.e., packet-losses. Under mild assumptions, theoretical convergence shown. The compared ADMM-based state estimation.