The Hybrid HMM for RSS-Based Localization in Wireless Sensor Networks

Abstract:

We proposea method of RSS-base localization in WSN (Wireless Sensor Network), called Hybrid HMM, to improve the stabilityof node localization basedon RSS(Received Signal Strength).This model utilizesHMM(Hidden Markov Model) to takeinto account the time factor when receiving the RSS sequence, andconverts the action of ranging into an operationof classification.For the received RSS used for localization,our Hybrid HMMwill compare it withthe preset RSS threshold value, and put the result into one of two categories for subsequent processing: If the received value is higher than the threshold value, the distance value will be drawn from the signal propagation model. If lower, the information will be obtained from a trained HMM. Experimental results show that the Hybrid HMM method can greatly improve the localization accuracy.

• Periodical: Applied Mechanics and Materials (Volumes 347-350)
• Pages: 796-802
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.347-350.796
• Online since: August 2013
• Authors: Yan Hong Zang, Jin Song Wang, Lin Ling, Pei Zhong Lu
• Keywords: Hidden Markov Model (HMM), Localization, Received Signal Strength, Wireless Sensor Network (WSN)

References

[1] K. Stone, T. Camp, A survey of distance-based wireless sensor network localization techniques, International Journal of Pervasive Computing and Communications, vol. 8, pp.158-183, (2012).

DOI: 10.1108/17427371211245373

[2] D. Macii, F. Trenti, P. Pivato, A robust wireless proximity detection technique based on RSS and TOF measurements, Measurements and Netwroking Proceedings, 2011 IEEE International Workshop, pp.31-36, (2011).

DOI: 10.1109/iwmn.2011.6088482

[3] M. Soleimanifar, M. Lu, I. Nikolaidis,S. Lee, A robust positioning architecture for construction resources localization using wireless sensor networks, the 2011 Winter Simulation Conference, pp.3357-3567, (2011).

DOI: 10.1109/wsc.2011.6148050

[4] Y. Zhao, H. Zhou, M. L, A novel overlap area matching algorithm based on location fingerprinting in wireless networks", Wireless communications, Networking and Mobile Computing, 2009, WiCOM, 09. 5th International Conference, pp.1-5, (2009).

DOI: 10.1109/wicom.2009.5305359

[5] A. Coluccia, F. Ricciato, Maximum Likelihood trajectory estimation of a mobile node from RSS measuremets, Wireless on-demand Network Systems and Services(WONS), 2012 9th Annual Conference, pp.151-158, (2012).

DOI: 10.1109/wons.2012.6152222

[6] J. Krumm, E. Horvitz, LOCADIO: Inferring Motion and Location from Wi-Fi Signal Strengths, First Annual International Conference on Mobile and Ubiquitous Systems: Networking and Services, Boston, MA, USA, pp.4-13, (2004).

DOI: 10.1109/mobiq.2004.1331705

[7] C. Morelli, M. Nicoli, Hidden Markov Models for Radio Localization in Mixed LOS/NLOS Conditions, IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 55, NO. 4, pp.1525-1542, APRIL (2007).

DOI: 10.1109/tsp.2006.889978

[8] V. Zheng, E. Xiang, Q. Yang,D. Shen, Transferring Localization Models Over Time, Twenty-Third AAAI Conference on Areificial Intelligence, pp.1421-1426, (2008).

[9] R. Arthi, K. Murugan, Localization in Wireless Sensor Networks by Hidden Markov Model, Advanced Computing (ICoAC), 2010 second International Conference, pp.14-18, (2010).

DOI: 10.1109/icoac.2010.5725355

[10] http: /www. comp. leeds. ac. uk/roger/HiddenMarkovModels/html_dev/main. html.