IPSJ SIG Technical Report Vol.2013-HCI-155 No.5 Vol.2013-UBI-40 No /11/5 Boosting Wi-Fi Wi-Fi Wi-Fi Wi-Fi Wi-Fi Boosted Wi-Fi,,, Boostin

Transcription

1 Boosting Wi-Fi 1 1 Wi-Fi Wi-Fi Wi-Fi Wi-Fi Boosted Wi-Fi,,, Boosting Robust Wi-Fi Indoor Positioning Method using Boosting Taniuchi Daisuke 1 Maekawa Takuya 1 Suzuki Jun Kishino Yasue Abstract: Recently, many indoor positioning techniques based on Wi-Fi signals have been studied. Wi-Fi fingerprinting technique, which is one of the most popular and practical method, makes use of the Wi-Fi received signal strength (RSS) information collected at several indoor places in advance to construct an indoor positioning model. However, changing environmental dynamics, i.e., layout changes and moving or removal of WiFi access points, may cause the instability of Wi-Fi based positioning methods. In this work, we try to cope with the instability with a boosted positioning estimator consists of several weak estimators. Each weak estimator uses only the signals from some randomly selected APs. Even when signal strength from a specific AP may change, some weak estimators that do not employ the AP are not affected by the change. In our proposed method, we track a user s coordinates with the particle filter and we evaluate each weak estimator s prediction by using the particle filter outputs. That is, we find weak estimators that are not affected by the AP by comparing the predictions and coordinates estimated by the particle filter based on the past coordinate history. Our boosted estimator computes final estimation based on the trustworthy weak estimators. Keywords: Wi-Fi fingerprinting, signal change, particle filter, boosting 1 Graduate School of Information Science and Technology, Osaka University, Suita, Osaka, Japan NTT NTT Communication Science Laboratories 1. GPS Wi-Fi LAN GPS c 013 Information Processing Society of Japan 1

2 GPS A-GPS Wi-Fi Wi-Fi [1], [] Wi-Fi Wi-Fi Boosting Wi-Fi Wi-Fi 3 4. Wi-Fi.1 Wi-Fi Wi-Fi Wi-Fi Naive BayesSVMk knn: k-nearest Neighbor algorithmgmm: Gaussian Mixture Model [3] Wi-Fi k. Rai [4] Zee Zee Zee Robertson [5] FootSLAM FeetSLAM [6]Ferris [7] WiFi GP-LVM(Gaussian process latent variable models)[8] WiFi WiFi-SLAM Wi-Fi.3 Wi-Fi Wi-Fi c 013 Information Processing Society of Japan

3 S. Chen [9] Y.C. Chen [10] Yin [11] Wi-Fi Wi-Fi Wi-Fi Pan [1] Wi-Fi Manifold co-regularization Wi-Fi (1) Wi-Fi () Wi-Fi PDRPedestrian dead-reckoning Wi-Fi Wi-Fi [13] Wi-Fi Fig. 1 1 Outline of proposed method 3. Wi-Fi x n f(x, n) = i f i(x i, µ i,n, σ i,n ) i x i f i (x i, µ i,n, σ i,n ) = 1 πσ i,n exp ( (x i µ i,n ) σ i,n µ i,n σ i,n n i f(x, n) top-k 3.3 t ) c 013 Information Processing Society of Japan 3

4 t [14] 3 m t+1 m t+1 wt+1 i t p t p i t+1 = Ap t + w A p t p t 1 Fig. Weighting of boosted position estimator and particles w 0 1 p t p i t+1 i 1 5 Wi-Fi Wi-Fi m t+1 p i t+1 m t+1 wt+1 i = N(p i t+1 m t+1 ) 4/ t n Wi-Fi t w t n t 1 w t 1 wt n = λwt 1 n + (1 λ)w t n λ 0 < λ < 1 t 3.5 Wi-Fi Wi-Fi Wi-Fi Wi-Fi c 013 Information Processing Society of Japan 4

5 0 Wi-Fi 3. Wi-Fi 3.6 Wi-Fi Wi-Fi Wi-Fi Wi-Fi 8 ( 1 ) Google Galaxy Nexus ( ) Wi-Fi Boosted ( 3 ) 16 Wi-Fi Boosted 3 GMM Fig m 16.3m Floor plan of experimental environment9.8m 16.3m GMM GMM AP Boosted 5 k k = 3 [ 1] [ ] Wi-Fi 4. 1 GMM 4.80m GMM 3.75m GMM GMM c 013 Information Processing Society of Japan 5

6 m.5 [ ] Gmm [] Gmm m [ ] [] 4 GMM 1 Fig. 4 Transision of accuracies related to GMM method and our methods GMM GMM 4.37m m 3.69m GMM 1 5. Wi-Fi Boosted Boosted 5 Fig. 5 GMM Transitions of accuracies related to GMM method and our methods when we remove APs in one room at 15th day [1] P. Bahl and V. N. Padmanabhan. RADAR: An inbuilding RF based user location and tracking system, Proc. of Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM 000), volume, pp (000). [] A. LaMarca, Y. Chawathe, S. Consolvo, J. Hightower, I. Smith, J. Scott, T. Sohn, J. Howard, J. Hughes, F. Potter and others: Place lab: Device Positioning Using Radio Beacons in the Wild, Proc. of Pervasive 005, pp (005). [3] S. Saha, K. Chaudhuri, D. Sanghi and P. Bhagwat: Location determination of a mobile device using IEEE 80.11b access point signals, Proc. of Wireless Communications and Networking Conference (WCNC 003), pp (003). [4] A. Rai, K. K. Chintalapudi, V. N. Padmanabhan and R. Sen: Zee: Zero-effort crowdsourcing for indoor localization, Proc. of the 18th Annual International Conference on Mobile Computing and Networking (MobiCom 01), pp (01). [5] P. Robertson, M. Angermann, and B. Krach, Simultaneous localization and mapping for pedestrians using only foot-mounted inertial sensors, Proc. of 11th International Conference on Ubiquitous Computing (Ubi- Comp 009), pp (009). [6] P. Robertson, M. G. Puyol and M. Angermann: Collab- c 013 Information Processing Society of Japan 6

7 orative Pedestrian Mapping of Buildings Using Inertial Sensors and FootSLAM, Proc. of the 6th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 011), pp (011). [7] B. Ferris, D. Fox and N. D. Lawrence: WiFi-SLAM Using Gaussian Process Latent Variable Models, Proc. of the twentieth International Joint Conferences on Artificial Intelligence (IJCAI-07), pp (007). [8] N. D. Lawrence: Gaussian process models for visualisation of high dimensional data, Proc. of Advances in Neural Information Processing Systems (NIPS), volume 16 (004). [9] S. Chen, Y. Chen, and W. Trappe: Exploiting environmental properties for wireless localization and location aware applications, Proc. of Sixth Annual IEEE International Conference on Pervasive Computing and Communications (PerCom 008), pp (008). [10] Y. C. Chen, J. R. Chiang, H. h. Chu, P. Huang, and A. W. Tsui: Sensor-Assisted Wi-Fi Indoor Location System for Adapting to Environmental Dynamics, Proc. of the 8th ACM International Symposium on Modeling, Analysis and Simulation of Wireless and Mobile Systems (MSWiM 05), pp (005). [11] J. Yin, Q. Yang and L. Ni: Adaptive temporal radio maps for indoor location estimation, Proc. of Third IEEE International Conference on Pervasive Computing and Communications (PerCom 005), pp (005). [1] S. J. Pan, J. T. Kwok, Q. Yang and J. J. Pan: Adaptive localization in a dynamic WiFi environment through multi-view learning, Proc. of AAAI 07, Vol., pp (007). [13] and :, (013). [14] A. Doucet, N. De Freitas and N. Gordon: Sequential Monte Carlo methods in practice, Springer Verlag (001). [15] J. C. Stein: Indoor Radio WLAN Performance Part II: Range Performance In a Dense Office Environment, Intersil Corporation (1998). [16] A. P. Jardosh, E. M. B. Royer, K. C. Almeroth and S. Suri: Real-world environment models for mobile network evaluation, IEEE Journal on Selected Areas in Communications, vol. 3, pp (005). c 013 Information Processing Society of Japan 7