Use of Area Layout Information for RSSI-Based Indoor Target Tracking Methods

Daisuke ANZAI  Kentaro YANAGIHARA  Kyesan LEE  Shinsuke HARA  

Publication
IEICE TRANSACTIONS on Communications   Vol.E94-B   No.7   pp.1924-1932
Publication Date: 2011/07/01
Online ISSN: 1745-1345
DOI: 10.1587/transcom.E94.B.1924
Print ISSN: 0916-8516
Type of Manuscript: PAPER
Category: Network
Keyword: 
indoor target tracking,  maximum likelihood estimation,  maximum a posteriori estimation,  infinite impulse response filter,  particle filter,  

Full Text: PDF>>
Buy this Article




Summary: 
For an indoor area where a target node is tracked with anchor nodes, we can calculate the priori probability density functions (pdfs) on the distances between the target and anchor nodes by using its shape, three-dimensional sizes and anchor nodes locations. We call it “the area layout information (ALI)” and apply it for two indoor target tracking methods with received signal strength indication (RSSI) assuming a square location estimation area. First, we introduce the ALI to a target tracking method which tracks a target using the weighted sum of its past-to-present locations by a simple infinite impulse response (IIR) low pass filter. Second, we show that the ALI is applicable to a target tracking method with a particle filter where the motion of the target is nonlinearly modelled. The performances of the two tracking methods are evaluated by not only computer simulations but also experiments. The results demonstrate that the use of ALI can successfully improve the location estimation performance of both target tracking methods, without huge increase of computational complexity.