DocumentCode
2771897
Title
Fundamental Limits of Wideband Cooperative Localization via Fisher Information
Author
Shen, Yuan ; Wymeersch, Henk ; Win, Moe Z.
Author_Institution
Lab. for Inf. & Decision Syst., Massachusetts Inst. of Technol., Cambridge, MA
fYear
2007
fDate
11-15 March 2007
Firstpage
3951
Lastpage
3955
Abstract
Determination of position accuracy for geolocation is a fundamental issue in wireless sensor networks. In a dense obstacle environment, anchors (or base stations) may not be able to provide sufficient localization information to agents because of radio blockage or limited range. In such cases, cooperation among agents (or mobile stations) can be very helpful. In this paper, we develop a model for cooperative localization based on time-of-arrival (TOA) ranging information and derive the position error bound (PEB) for agents in the network using information inequality. Equivalent Fisher information (EFI), which has been applied in the single agent localization case (Sen and Win, 2007), is employed to characterize the localization accuracy. From analysis, we also show that anchors and agents are essentially equivalent in our unified cooperative localization model.
Keywords
eigenvalues and eigenfunctions; mobile agents; mobile computing; time-of-arrival estimation; wireless sensor networks; agent localization; cooperative localization model; equivalent Fisher information; geolocation; information inequality; position error bound; time-of-arrival ranging; wideband cooperative localization; wireless sensor networks; Bandwidth; Base stations; Communications Society; Cramer-Rao bounds; Eigenvalues and eigenfunctions; Global Positioning System; Peer to peer computing; Pollution measurement; Wideband; Wireless sensor networks;
fLanguage
English
Publisher
ieee
Conference_Titel
Wireless Communications and Networking Conference, 2007.WCNC 2007. IEEE
Conference_Location
Kowloon
ISSN
1525-3511
Print_ISBN
1-4244-0658-7
Electronic_ISBN
1525-3511
Type
conf
DOI
10.1109/WCNC.2007.722
Filename
4224967
Link To Document