Non line of sight localization in multipath environment

• Main Author: Maung, Thein Toe.
• Other Authors: Tan Soon Yim
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/40891
• Institution: Nanyang Technological University

The main focus of this project is to improve the accuracy of the estimation of the mobile device in multipath environment. In general, the conventional localization schemes only work when there is Line of Sight (LOS) between the reference devices (RDs) and the mobile device (MD). Furthermore, the conventional localization schemes require at least 3 RDs for TOA localization schemes and 2 RDs for AOA localization schemes while our proposed localization schemes, which are Area of Possible Mobile Device (APMD) and APMD using the image of reference device, require only one RD. In this report, firstly the methodology of the Area of Possible Mobile Device (APMD) was proposed. APMD uses the bidirectional LOS Time of Arrival (TOA) and Angle of Arrival (AOA) from both RD and MD to track the location of the mobile device. In order to improve the accuracy of the estimation, AOA auto adjustment is also proposed. The performance of the proposed localization schemes were compared with conventional localization schemes. As the main objective of this report which is to improve the accuracy of the estimation of MD, APMD using the image of reference device was proposed. In multipath environment, since there is always a mixture of LOS and NLOS paths, instead of using single data from LOS, this methodology uses the bidirectional of LOS and NLOS data from both RD and MD to give better accuracy in estimation. In this method, from NLOS data, we first estimate the location of the image of the reference device, and there after the location of the mobile device MD is estimated from the estimated image of the reference device. Taking the average of the LOS APMD estimation and NLOS image of reference device estimation, the accuracy was improved significantly. The AOA auto adjustment for NLOS is also proposed in this section to get the better accuracy. By using MATLAB simulation, the accuracy of the proposed localization schemes were compared with conventional localization schemes. In general, all our proposed localization schemes are not only able to localize the MD using just one RD but also able to achieve better localization accuracy than the conventional localization schemes.