An Optimal Scanning Sensor Activation Policy for Parameter Estimation of Distributed Systems
A technique is proposed to solve an optimal node activation problem in sensor networks whose measurements are supposed to be used to estimate unknown parameters of the underlying process model in the form of a partial differential equation. Given a partit
- PDF / 682,717 Bytes
- 36 Pages / 439.36 x 666.15 pts Page_size
- 54 Downloads / 245 Views
Abstract A technique is proposed to solve an optimal node activation problem in sensor networks whose measurements are supposed to be used to estimate unknown parameters of the underlying process model in the form of a partial differential equation. Given a partition of the observation horizon into a finite number of consecutive intervals, the problem is set up to select nodes which will be active over each interval while the others will remain dormant such that the log-determinant of the resulting Fisher information matrix associated with the estimated parameters is maximized. The search for the optimal solution is performed using the branch-andbound method in which an extremely simple and efficient technique is employed to produce an upper bound to the maximum objective function. Its idea consists in solving a relaxed problem through the application of a simplicial decomposition algorithm in which the restricted master problem is solved using a multiplicative algorithm for D-optimal design. The performance evaluation of the technique is additionally presented by means of simulations.
1 Introduction 1.1 Distributed Sensor Networks and Sensor Management Distributed sensor networks have recently become an important research area regarding spatio-temporal phenomena [10, 11, 15, 35, 72, 99]. This is because dramatic progress in hardware, sensor and wireless networking technologies enables large-scale deployment of superior data acquisition systems with adjusting
D. Uci´nski () Institute of Control and Computation Engineering, University of Zielona G´ora, ul. Podg´orna 50, 65–246 Zielona G´ora, Poland e-mail: [email protected] H.G. Bock et al. (eds.), Model Based Parameter Estimation, Contributions in Mathematical and Computational Sciences 4, DOI 10.1007/978-3-642-30367-8 4, © Springer-Verlag Berlin Heidelberg 2013
89
90
D. Uci´nski
resolutions. A sensor network may comprise thousands of inexpensive, miniature and low-power sensor nodes that can be deployed throughout a physical space and connect through a multi-hop wireless network, providing dense sensing close to physical phenomena. The nodes process and locally communicate the collected information, as well as coordinate actions with each other. Sensor networks have recently come into prominence because they hold the potential to revolutionalize a wide spectrum of both civilian and military applications, including monitoring microclimates and wildlife habitats, tracking chemical plumes, traffic surveillance, industrial and manufacturing automation, building and structures monitoring, and many others. What makes sensor networks so attractive is their miniaturization, low cost, low power radio and autonomous ad hoc connectivity, which basically eliminates the need for any human intervention. The design, implementation and operation for a sensor network requires the confluence of many disciplines, including signal processing, networking and protocols, embedded systems, information management and distributed algorithms. There are, however, a number of
Data Loading...
