Home > Publications > Doctoral dissertation & Master thesis
Domestic Conference
 
 
Title Range-free Localization: Mobile Beacon and Multi-hop Problems
Degree Ph.D.
Author Sang-Ho Lee
Advisor Kiseon Kim
Graduation Date 2012.08.24 File
    Date 2013-09-11 11:03
Localization is the process to find out the locations of sensor nodes which are randomly deployed over the sensing area, to specify "where" a certain event takes place. The locations of sensor nodes are important to identify where the reported events happen, in order to cope with the events timely. In addition, location information assists environmental monitoring, location-based routing, coverage, forwarding inquiries, etc. Localization schemes can be divided into range-free and range-based schemes. Range-free schemes can localize sensor nodes by estimating distance information without power-consuming and expensive devices for measuring distance or angle information, while range-based schemes measure distance and angle information. Since range-free schemes do not require any additional hardware for measurement, range-free schemes could be a good approach to localize sensor nodes with respect to the constraints of low cost and low energy consumption. Range-free schemes can be divided into schemes with static beacons and a mobile beacon according to the type of beacon nodes. In the case of schemes with static beacons, they can be divided into one-hop schemes and multi-hop schemes. In one-hop schemes, a sensor node is connected to beacon nodes with a one-hop connection. In multi-hop schemes, a sensor node is connected to beacon nodes with a multi-hop connection. Due to the mobility of a mobile beacon, the schemes with a mobile beacon belong to one-hop schemes.Our focuses are mobile beacon and multi-hop problems. Even though there are many issues for localization, we investigated the improvement of the localization accuracy in mobile beacon and multi-hop problems, because the accuracy of the location is the most important and critical issue in localization. To improve the localization accuracy for range-free schemes with a mobile beacon, we propose a localization scheme using geometric constraints and propose static path planning of a mobile beacon. To improve the localization accuracy for range-free multi-hop schemes, we propose a localization scheme with distance refinement by considering the characteristics of a multi-hop and utilizing node density within the communication range.As the first contribution, we improve the localization accuracy for range-free schemes with a mobile beacon. Localization schemes using a mobile beacon have similar effects as the use of many static beacons in terms of improving localization accuracy. Specifically, the localization scheme with mobile beacons proposed by Ssu et al. has fine-grained accuracy, scalability, and power efficiency without requiring measured distance or angle information. However, this scheme often has large location errors in ill-conditioned cases. To improve the localization accuracy in Ssus scheme, we propose a localization scheme that estimates sensor location from possible areas by using geometric constraints. During simulations, the proposed scheme was shown to provide higher localization accuracy than Ssus scheme and other schemes using a mobile beacon.As the second contribution, we propose static path planning of a mobile beacon to improve the localization accuracy for our proposed scheme with a mobile beacon based on geometric constraints. Since the path of a mobile beacon directly affects the localization accuracy, it is needed to plan a static path for improving the accuracy. We derives the average values of the maximum location errors and the worst location error according to the static path and determines the static path to minimize the average values of the maximum location errors. Simulation results show that the determined static path provides higher localization accuracy than other static paths.As the third contribution, we compare performance between DV-Hop and Amorphous since the DV-Hop and Amorphous schemes are typical range-free multi-hop schemes and they have distinctive features caused by their different approaches in multi-hop distance estimation. Through the comparison between the DV-Hop and Amorphous schemes, we can know that Amorphous is better than DV-Hop when beacon nodes are randomly deployed. For the reason, Amorphous is more appropriate for such applications as earthquake and battlefield environments, where it is difficult to manually deploy beacon nodes. Then, we propose a specific algorithm, to improve the Amorphous scheme, including estimating each hop distance independently as a function of the specific number of neighbor nodes of each sensor node and considering the characteristic of the last hop. We derive the estimated hop progress as a function of the number of neighbor nodes of each sensor node by analyzing the relationship between a forward hop progress and the number of one-hop neighbor nodes.
목록
 
 
광주과학기술원 한·러 MT-IT 융합기술연구센터 광주과학기술원정보통신공학부