Degree Type


Date of Award


Degree Name

Master of Science


Electrical and Computer Engineering

First Advisor

Greg R. Luecke


Precise Indoor Localization is a major component of numerous location based applications and services which perform indoor guidance and object tracking. There are many existing solutions which address the localization issue, but most of them do not provide a fault tolerant solution. In this work, we have developed a fault tolerant statistical method which leverages the existing infrastructure by using the readily available Wi-Fi Access Points. Our proposed method can be applied to any environment which has a Wi-Fi coverage and we do not assume the knowledge of the placement of the Access Points or any physical layout. Initially we map the signal strengths and the corresponding positions to obtain the RF distribution of the region and this is the offline phase. We develop different fault tolerant models and use an Android application for monitoring various Access Points to provide the status of the Access Points in the environment. During the online phase, we measure the signal strength at distributed locations in the environment and then, depending on the status obtained from the application, we use the appropriate scheme to obtain the corresponding locations. In specific we use a Maximum Likelihood Estimator to obtain the position from the previously recorded RF map. Further, we provide 95% confidence intervals for the location obtained by using a Bootstrap method. Our method, compared to other deterministic methods is more accurate and fault tolerant. We also provide the experimental results which validate the accuracy of our method in obtaining the user location.


Copyright Owner

Anusha Chennaka



File Format


File Size

43 pages