Automation of localization methods for wireless WLAN networks

Description

Localization in WLAN networks is finding the position of a mobile node, using the properties of a wireless network (radio signal strength, propagation delay, information about the neighbours, etc.). Localization is used for various purposes ranging from personal navigation systems to finding the position of an emergency call. Location-aware services increase the value of existing wireless network infrastructures.

The known localization methods achieve a precision of the range from one to ten meters [1,5,6,8]. For instance, the Horus system achieves correct location estimations with a precision of up to one meter in the 90% of the cases [8]. Many localization algorithms use the Machine Learning principle. During the training phase they create a radio-map between received signal strengths and the corresponding positions. During the online phase, these data is used to estimate the position from the actual signal strengths.

A problem for these methods is the constriction and the actualization of the radio-map. In the cited works this is done manually by walking around in the environment and measuring the signal strength values at grid-points at 1 to 5 meters. This method is problematic, because it requires lots of manual effort. Moreover, the environment can change continuously. Some WLAN networks use the so called self-healing methods and automatically (re)adjust the transmission power of the Access Points for an optimal coverage [4]. These environmental changes require again manual effort and time to calibrate the knowledge about the environment.

This seminar topic considers methods and algorithms that automate WLAN localization techniques, or reduce the manual effort by their application. The scientific works [2,3,7] and the commercial product of Cisco [5] can be used as a starting point. All references are available on request.

This topic requires knowledge in the following fields: wireless networks, probabilistic theory, Hidden Markov Models (HMM), and regression analysis.

References

1. P. Bahl and V. N. Padmanabhan. RADAR: An In-Building RF-Based User Location and Tracking System. IEEE Infocom, 2000.
2. Ezekiel S. Bhasker, Steven W. Brown, William G. Griswold. Employing User Feedback for Fast, Accurate, Low-Maintenance Geolocationing. In Second IEEE International Conference on Pervasive Computing and Communications (PerCom'04), 2004.
3. Xiaoyong Chai and Qiang Yang. Reducing the Calibration Effort for Location Estimation Using Unlabeled Samples. In Proceedings of IEEE International Conference on Pervasive Computing and Communications (PerCom 2005)
4. CiscoWorks Wireless LAN Solution Engine
5. Cisco 2700 Series Wireless Location Appliance
6. Ekahau Positioning Engine
7. Jie Yin, Qiang Yang, Lionel Ni. Adaptive Temporal Radio Maps for Indoor Location Estimation. In Proceedings of IEEE International Conference on Pervasive Computing and Communications (PerCom 2005)
8. Moustafa Youssef and Ashok Agrawala. The Horus WLAN Location Determination System. In MobiSys '05: Proceedings of the 3rd international conference on Mobile systems, applications, and services, 2005. ACM Press.

Contact

Svilen Ivanov