Development an accurate and stable range-free localization scheme for anisotropic wireless sensor networks

Han, Fengrong (2022) Development an accurate and stable range-free localization scheme for anisotropic wireless sensor networks. PhD thesis, Universiti Malaysia Pahang (Contributors, Thesis advisor: Mohamed Abdelaziz, Izzeldin Ibrahim).

07.Development an accurate and stable range-free localization scheme for anisotropic wireless sensor networks.pdf - Accepted Version

Download (394kB) | Preview


With the high-speed development of wireless radio technology, numerous sensor nodes are integrated into wireless sensor networks, which has promoted plentiful location-based applications that are successfully applied in various fields, such as monitoring natural disasters and post-disaster rescue. Location information is an integral part of wireless sensor networks, without location information, all received data will lose meaning. However, the current localization scheme is based on equipped GPS on every node, which is not cost-efficient and not suitable for large-scale wireless sensor networks and outdoor environments. To address this problem, research scholars have proposed a rangefree localization scheme which only depends on network connectivity. Nevertheless, as the representative range-free localization scheme, Distance Vector-Hop (DV-Hop) localization algorithm demonstrates extremely poor localization accuracy under anisotropic wireless sensor networks. The previous works assumed that the network environment is evenly and uniformly distributed, ignored anisotropic factors in a real setting. Besides, most research academics improved the localization accuracy to a certain degree, but at expense of high communication overhead and computational complexity, which cannot meet the requirements of high-precision applications for anisotropic wireless sensor networks. Hence, finding a fast, accurate, and strong solution to solve the range-free localization problem is still a big challenge. Accordingly, this study aspires to bridge the research gap by exploring a new DV-Hop algorithm to build a fast, costefficient, strong range-free localization scheme. This study developed an optimized variation of the DV-Hop localization algorithm for anisotropic wireless sensor networks. To address the poor localization accuracy problem in irregular C-shaped network topology, it adopts an efficient Grew Wolf Optimizer instead of the least-squares method. The dynamic communication range is introduced to refine hop between anchor nodes, and new parameters are recommended to optimize network protocol to balance energy cost in the initial step. Besides, the weighted coefficient and centroid algorithm is employed to reduce cumulative error by hop count and cut down computational complexity. The developed localization framework is separately validated and evaluated each optimized step under various evaluation criteria, in terms of accuracy, stability, and cost, etc. The results of EGWO-DV-Hop demonstrated superior localization accuracy under both topologies, the average localization error dropped up to 87.79% comparing with basic DV-Hop under C-shaped topology. The developed enhanced DWGWO-DVHop localization algorithm illustrated a favorable result with high accuracy and strong stability. The overall localization error is around 1.5m under C-shaped topology, while the traditional DV-Hop algorithm is large than 20m. Generally, the average localization error went down up to 93.35%, compared with DV-Hop. The localization accuracy and robustness of comparison indicated that the developed DWGWO-DV-Hop algorithm super outperforms the other classical range-free methods. It has the potential significance to be guided and applied in practical location-based applications for anisotropic wireless sensor networks.

Item Type: Thesis (PhD)
Additional Information: Thesis (Doctor of Philosophy) -- Universiti Malaysia Pahang – 2022, SV: DR. IZZELDIN IBRAHIM MOHAMED ABDELAZIZ, NO. CD: 13110
Uncontrolled Keywords: range-free localization scheme, anisotropic wireless sensor networks
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculty/Division: Institute of Postgraduate Studies
College of Engineering
Depositing User: Mr. Nik Ahmad Nasyrun Nik Abd Malik
Date Deposited: 07 Dec 2022 03:04
Last Modified: 01 Nov 2023 07:17
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item