Loading...

Accurate and Low-Cost Location Estimation using Machine Learning Techniques in Wireless Sensor Networks

Afzal, Samira | 2011

405 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: English
  3. Document No: 41736 (52)
  4. University: Sharif University of Technology, International Campus, Kish Island
  5. Department: Science and Engineering
  6. Advisor(s): Beigy, Hamid
  7. Abstract:
  8. Wireless sensor networks have a wide range of applications in the world. In most of the applications, collected data is not usable without the knowledge about the localization of events. There are two approaches to specifying the location of a sensor: using hardware solutions such as GPS, which is an expensive solution, and using the localization algorithm. Therefore, localization has an important role in sensor networks. Most of the current localization algorithms are non-adaptive and proposed for fixed wireless sensor networks. Recently, adaptive localization algorithms have been considered because of their simple implementation, fast result and low computation overhead for each node. In addition, learning concepts is more suitable for mobile networks and target tracking. Most of the adaptive algorithms for mobile networks are used in received signal strength. Signal strength doesn’t need extra hardware or time synchronization. But it has less accurate result because of signal reflection or obstacle existence or humidity. Therefore, these algorithms are not suitable for outdoors and large scale networks. In another view, most of the proposed adaptive and non-adaptive localization algorithms are considered two-dimensional network areas. However, many applications need three-dimensional localization algorithms. In this thesis, we propose an adaptive localization algorithm for mobile large scale networks, which is suitable to deploy in three-dimensional network area. The experimental results show more than 90 percent accuracy for the proposed algorithm
  9. Keywords:
  10. Wireless Sensor Network ; Machine Learning ; Sensors ; Localization

 Digital Object List

  • مكان يابي دقيق و كم هزينه با استفاده از تكنيك هاي يادگيري ماشين
  •   view

 Bookmark

No TOC