Self-Organizing Distributed State Estimators

• Published in: Intelligent Sensor Networks pp. 483 - 513
• Main Authors: Sijs, Joris, Papp, Zoltan
• Format: Book Chapter
• Language: English
• Published: CRC Press 2013
• Edition: 1
• Subjects: Reconfiguration Process; Fusion Method; Global State Vector; Local Measurements; Sensor Networks; Large Area Processes; Local Sampling Time; Sensor Nodes; Sampling Instants; Local Estimate; Wireless Sensor Networks; Management Layer; Runtime Reconfiguration; Neighboring Nodes; Distributed Kalman Filtering; Critical Energy Level; Self-organization Sensor Network; Cognitive Wireless Sensor Networks; Intelligent Sensor Networks; Improved Estimation Result; KF; Power Consumption; Distributed State Estimation; Estimation Error
• ISBN: 1138199745; 9781439892817; 1439892814; 9781138199743
• DOI: 10.1201/b14300-25

This chapter investigates a self-organization sensor network with the purpose of estimating the state vector of large-area processes. Distributed solutions for signal processing techniques are important for establishing large-scale monitoring and control applications. Runtime reconfigurability has a thorough impact on system design, implementation, testing/validation, and deployment. The presented research focuses on the widespread signal processing method known as state estimation with Kalman filtering in particular. Computational demand is related to the algorithms performed in sensor networks for processing the measurements. The main advantage of exchanging local estimates is that measurement information spreads through the entire network, even under the condition that nodes exchange data only once per sampling instant. The design challenge of any embedded system is to realize given functionalities, in this case the ones of the local estimation algorithm, on a given hardware platform while satisfying a set of nonfunctional requirements, such as response times, dependability, power efficiency.