Paper Info
Title
Parameter Estimation Method for Time-Variant Target Object Using Randomly Deployed Sensors and Its Application to Participatory Sensing
Abstract
We propose a method for estimating the size, perimeter length, and location of a time-variant target object moving in a monitored area. This method uses only binary information from sensors of which locations are unknown, where the binary information includes whether each sensor has detected the target object or not. Analysis based on integral geometry provides the relationship between the number of sensors detecting the target object and the target object parameters to be estimated. Because this relationship is linear, a linear filter, such as the Kalman filter, is applicable to estimate parameters if we can assume that the dynamics of the parameters are linear. As a concrete example, the size, shape, and location of an active thunder area is estimated. The model discussed in this paper is applicable as a model of participatory sensing.
Year
DOI
Venue
2015
10.1109/TMC.2014.2347037
IEEE Transactions on Mobile Computing
Keywords
Field
DocType
Sensors,Estimation,Kalman filters,Shape,Nonhomogeneous media,Geometry,Wireless sensor networks
Computer vision,Thunder,Linear filter,Computer science,Kalman filter,Geometric probability,Artificial intelligence,Estimation theory,Integral geometry,Participatory sensing,Wireless sensor network
Journal
Volume
Issue
ISSN
14
6
1536-1233
Citations 
PageRank 
References 
6
0.44
16
Authors
2
Name
Order
Citations
PageRank
Hiroshi Saito116416.39
Shigeo Shioda216526.69