Hierarchical Cellula Multihop Networks分层多跳蜂窝网络.pptVIP

by hlwu, HSCC Localization by RFID “LANDMARC: Indoor Location Sensing Using Active RFID”, PerCom’03 RFID Background The current (2003) RFID system does not provide the transmission strength of tags to readers. Readers only report the power level of the tag being detected. Goal LANDMARC: LocAtioN iDentification based on dynaMic Active Rfid Calibration A location sensing prototype system that uses RFID technology for in-building use. Improves localization accuracy by utilizing the concept of reference tags. LANDMARC Approach (I) In the sensing field: n readers m fixed tags (i.e., reference tags) u tracking tags (i.e., moving objects) Readers are all configured with continuous mode and a detection-range of 1-8 level. Signal Strength Vector of a tracking tag: Signal Strength Vector of a fixed tag: LANDMARC Approach (II) Definition: Euclidian distance between a tracking tag and a reference tag j : Definition: The location relationship vector between the m reference tags and the tracking tag: LANDMARC Approach (III) The placement of the unknown tag is surrounded by k most likely reference tags. Weighted location (x, y) of the unknown tag: where weights are defined as Experimental Environment Spider System: Reader: 308MHz, has an 802.11b interface to communicate, detection range of 150 feet, detect up to 500 tags in 7.5 seconds, 8 different power levels Active tag: read range of 150 feet and can be increased to 1000 feet with the addition of a special antenna, send signal in random with an average of 7.5 seconds Performance metric: Location estimation error Experiment Environment Effect of k (Number of Nearest Neighbors) Environmental Factors: Daytime vs. Night Effect of n (Number of Readers) Effect of Placement of Reference Tags The accuracy is improved with a higher reference tag density. However, it is not as great as expected. Conclusions A new RFID-based location system. Major advantages: No need for a large number of expensive RFID readers The environmental