This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Relative Location in Wireless Networks Neal Patwari and Robert J. ODea Florida Research Lab Motorola Labs 8000 West Sunrise Blvd, Rm 2141 Plantation, FL 33322 [N.Patwari, Bob.ODea]@Motorola.com Yanwei Wang Dept. of Electrical & Computer Eng. University of Florida P.O. Box 116200 Gainesville, FL 32611-6200 email@example.com Abstract In ad-hoc networks, location estimation must be de- signed for mobility and zero-configuration. A peer-to- peer relative location system uses pair-wise range esti- mates made between devices and their neighbors. De- vices are not required to be in range of fixed base sta- tions, instead, a few known-location devices in the net- work allow the remaining devices to calculate their loca- tion using a maximum-likelihood (ML) method derived in this paper. This paper presents simulations using both a standard channel model and actual indoor chan- nel measurements for verification. Both simulation and measurements show that a peer-to-peer relative location system can provide accurate location estimation using received signal strength (RSS) as a ranging method. 1. Introduction In many proposed applications for wireless peer-to- peer and ad-hoc networks, knowing the location of the devices in the network is key. For ad-hoc networking, researchers have proposed using location information for routing purposes . For military, police, or fire- man radio networks, knowing the precise location of each person with a radio can be critical. In offices and in warehouses, object location and tracking applica- tions are possible with large-scale ad-hoc networks of wireless tags. Finally, for wireless sensor networks that have a variety of home, industrial, and agricultural ap- plications, knowledge of sensor location is critical. Motorola has introduced the concept of NeuRFon T M systems to describe a wireless sen- sor network in which distributed RF devices operate in analogy to human neurons. These systems are composed of devices that sense, process, transceive, and act in a distributed, low power network. Devices communicate with neighboring devices to pass around, condense, and make decisions based on information they have collected. NeuRFon T M devices, to be fault tolerant, are deployed more densely than necessary in the environment of interest. Location information in these systems will be critical both for identification, information fusion, and localized reactions to stimuli. The location of a sensor may replace ID numbers as the means for addressing sensors . 1.1. Exisiting Positioning Systems The Global Positioning System (GPS) has been sug- gested as a means to obtain location information in ad-hoc networks . For outdoor applications in which device density is low, and cost is not a major concern, GPS is a viable option. However, adding GPS capabil- ity to each device in a dense network is expensive. Fur- thermore, achieving high accuracy from GPS requires use of differential techniques....
View Full Document
This note was uploaded on 05/15/2010 for the course NETWORKING optimaizat taught by Professor Eamir during the Spring '10 term at University of Advancing Technology.
- Spring '10