Localization of RFID Tags Using Stochastic Tunneling

Basheer, Mohammed Rana; Jagannathan, S.

doi:10.1109/tmc.2012.80

Cited by 14 publications

(1 citation statement)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Currently available indoor positioning systems include cellular network-based systems [ 8 , 9 , 10 , 11 ], single frequency network-based systems [ 12 , 13 , 14 , 15 ], ultra-wide band (UWB)-based systems [ 16 ], wireless local area network (WLAN)-based systems [ 17 , 18 ], radio frequency identification (RFID)-based systems [ 19 , 20 ], dead-reckoning-based systems [ 21 ], Bluetooth indoor positioning systems [ 22 ] and pseudolite-based systems [ 23 ]. Among them, the received signal strength (RSS)-based WLAN indoor positioning system is the most popular technique employed [ 24 , 25 , 26 , 27 ].…”

Section: Introductionmentioning

confidence: 99%

An Improved Compressive Sensing and Received Signal Strength-Based Target Localization Algorithm with Unknown Target Population for Wireless Local Area Networks

Yan

Chen

et al. 2017

Sensors

View full text Add to dashboard Cite

In this paper a two-phase compressive sensing (CS) and received signal strength (RSS)-based target localization approach is proposed to improve position accuracy by dealing with the unknown target population and the effect of grid dimensions on position error. In the coarse localization phase, by formulating target localization as a sparse signal recovery problem, grids with recovery vector components greater than a threshold are chosen as the candidate target grids. In the fine localization phase, by partitioning each candidate grid, the target position in a grid is iteratively refined by using the minimum residual error rule and the least-squares technique. When all the candidate target grids are iteratively partitioned and the measurement matrix is updated, the recovery vector is re-estimated. Threshold-based detection is employed again to determine the target grids and hence the target population. As a consequence, both the target population and the position estimation accuracy can be significantly improved. Simulation results demonstrate that the proposed approach achieves the best accuracy among all the algorithms compared.

show abstract