Study On 3D Node Localization For Wireless Sensor Networks

The node localization is one of the most important research aspects in wireless sensor networks. However, most of the previous research limits localization to a two-dimensional (2D) space and does not provide enough information to infer three-dimensional (3D) location. But in the context of real-world applications, large quantities of sensor nodes are a physical impossibility to be airdropped into the area of absolute plane forming a ground WSN. The physical world is 3D scenario, e.g., deep sea, hill or wild nature. Due to the wireless channel attenuation, the precision of the node localization can not be protected. But the localization precision should forestall the deployment of wireless sensor networks to be predominated under some network parameters (such as node density, anchor node density and radio range). So the analysis of node localization is also an important technology for wireless sensor networks.Many localization methods in WSN have been proposed by far. These methods are usually aimed at how to locate unknown nodes or how to improve performance of localization under a certain density of anchor nodes. However, localization method for anchor nodes is limited to GPS or hand-placing method merely. Perhaps the simplest method of providing localization is to equip every anchor node with a GPS receiver in outdoor environments]. However, GPS technology presented a costly solution for localization in WSN, due to cost, form factor, energy consumption, and the requirement for a second radio. Another alternative is hand-placing each anchor node and manually recording its position. This is a tedious and error prone approach unsuitable for large sensor networks and many of the proposed WSN applications. This creates a demand for efficient and cost-effective localization discovery methods of anchor nodes in WSN. Therefore, the research contents of our paper are developed based on these problems. The main achievements and innovations in this dissertation focus on the three aspects as follows. Firstly, the weighted least squares method of node localization. A weighted least squares method is presented by using the ambient noise. The weighted least squares methods is presented in RSSI scenarios. Simulation results indicate that the weighted least squares method has excellent localization precision.Secondly, based on 2D Cramer-Rao analysis on condition that anchor nodes was placed in fix position for wireless sensor networks, the corresponding 3D Cramer-Rao analysis on condition that anchor nodes was distributed randomly are proposed in this thesis. With a matlab simulation platform, CRB was simulated respectively in two application scenarios, RSSI and TOA. The influence of anchor node density and radio range on the CRB was analyzed. Finally, A new localization method of anchor nodes was been proposed in wireless sensor networks. The magnetic dipole model was analysed. The obtained absolute coordinates method of anchor nodes was provided in the case of placing anchor nodes randomly under 2D flat. The result of simulation indicates the performance of the localization method. And the method can resolve easily the problem that new anchor nodes replace old anchor nodes.The dissertation consists of seven chapters.In chapter one, the introduction and the analysis of the development and the present situations of wireless sensor network are summarized. The theoretical significance and the practical values of purpose and contents of this dissertation are given.Chapter two introduces the basic methods of node localization for wireless sensor networks. Sum up all sorts of benchmark about localization, and summarize the previous localization method.In chapter three, the node localization of MLE for 3D wireless sensor networks。The triangular algorithm of 2D is applied to 3D system, and the MLE arithmetic is presented in the case of 3D wireless sensor networks。In chapter four, the weighted least squares method of node localization. A weighted least squares method is presented by using the ambient noise. The weighted least square methods is presented in RSSI scenario. Simulation results indicate that the weighted least squares method has excellent localization precision. In chapter five, 3D Cramer-Rao analysis for wireless sensor networks. Based on 2D Cramer-Rao analysis on condition that anchor nodes was placed in fix position for wireless sensor networks, the corresponding 3D Cramer-Rao analysis on condition that anchor nodes was distributed randomly are proposed in this thesis. With a matlab simulation platform, CRB was simulated respectively in two application scenarios, RSSI and TOA. The influence of anchor node density and radio range on the CRB was analyzed.In chapter six, study on the node localization for wireless sensor networks under magnetic dipole model. A new localization method of anchor nodes was been proposed in wireless sensor networks. The magnetic dipole model was analysed. The obtained absolute coordinates method of anchor nodes was provided in the case of placing anchor nodes randomly under 2D flat. The result of simulation indicates the performance of the localization method. And the method can resolve easily the problem that new anchor nodes replace old anchor nodes.In chapter seven, a brief summary of the dissertation is given, and future work is also expected.