TY - GEN
T1 - A novel real time method of signal strength based indoor localization
AU - Ye, Letian
AU - Geng, Zhi
AU - Xue, Lingzhou
AU - Liu, Zhihai
PY - 2007
Y1 - 2007
N2 - Localization using wireless signal is a hot field now, and the real time indoor localization is a difficult problem for its complex and sensitive to the environment. This paper proposes a method based on grid to convert global to local. Based on the Markov random field, we convert efficiently signals between different environments and achieve high precision and fast speed. The paper also discusses influence of multiple signals to location precision, explains that multiple sets of signal can be used greatly to improve localization precision. To reduce the number of supervised grids in learning data required by the grid-matching algorithm, this paper presents a method which combines the grid matching and the signal strength model. First the position is localized by the grid-matching method and then its location is refined by using the signal strength model in the local area.
AB - Localization using wireless signal is a hot field now, and the real time indoor localization is a difficult problem for its complex and sensitive to the environment. This paper proposes a method based on grid to convert global to local. Based on the Markov random field, we convert efficiently signals between different environments and achieve high precision and fast speed. The paper also discusses influence of multiple signals to location precision, explains that multiple sets of signal can be used greatly to improve localization precision. To reduce the number of supervised grids in learning data required by the grid-matching algorithm, this paper presents a method which combines the grid matching and the signal strength model. First the position is localized by the grid-matching method and then its location is refined by using the signal strength model in the local area.
UR - http://www.scopus.com/inward/record.url?scp=38049058487&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=38049058487&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-74472-6_55
DO - 10.1007/978-3-540-74472-6_55
M3 - Conference contribution
AN - SCOPUS:38049058487
SN - 9783540744689
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 678
EP - 688
BT - Computational Science and Its Applications - ICCSA 2007 - International Conference, Proceedings
PB - Springer Verlag
T2 - International Conference on Computational Science and its Applications, ICCSA 2007
Y2 - 26 August 2007 through 29 August 2007
ER -