Research On Non-Line-of-Sight Localization Technology For Wireless Sensor Network

Wireless sensor network is a kind of integrated intelligent information processing platform, including sensor technology, micro-electromechanical technology, modern networking and wireless communication technology. It has broad prospects of application and its development and application will have revolutionary consequences in every field of human life and production. Due to the sensor nodes' uncontrollability when arranged, the majority of nodes in the network location can not be determined in advance while much applications of the wireless sensor network require nodes'location information in the network to obtain the exact location of sources of information.Node localization is the basis of many applications of wireless sensor networks and positioning accuracy is one of its key performance indicators. Although, theoretically, distance measuring technology can improve the positioning accuracy, the measurement of the distance between the nodes can easily be affected by those factors such as multi-path, non-line of sight transmission, multi-access interference and so on. The distance measurement errors include errors of measuring noise and non-line of sight, of which NLOS error is the main source of error in location estimation. A major problem in Node precise positioning is the non-line of sight signal transmission. This page focuses on positioning of wireless sensor network node, being committed to overcoming the positioning error in the non-line environment and its transmition accumulation in the process of wireless sensor network' positioning. The principal tasks summarized as follows:Based on the analysis of non-line-error suppression technique, this page presents a new weighted residuals localization algorithm. Compared with typical Rwgh positioning algorithm, this algorithm does not take all possible combinations of the measured values into account, but choose the smallest residuals combination to compute at each step. So it not only reduce the computational volume, but also avoid the positioning accuracy being influenced by a large number of high residual location estimation. Additionally, this algorithm adopts improved LLS algorithm to calculate the intermediate node coordinates, thus improve the positioning accuracy. For the residual weighting algorithm requires at least four or more neighbors anchor nodes, WLS algorithm can only amend NLOS error through the empirical formula. This paper presents a localization algorithm based on distance geometry. The algorithm transforms position estimation into a constrained optimization problem. It introduces measurement of the distance scaling factor to revise the objective function, utilizes distance geometry theorem and the geometric relationship between nodes to construct distance error optimization constraint function and calibrate the distance measurements through the optimization to make them more close to the line of sight environment measurements and suppress NLOS error.In large-scale wireless sensor networks, in order to reduce costs, the percentage of anchor nodes will be reduced, but it will reduce the node's location coverage. And changing positioned ordinary nodes into anchor nodes will bring the error accumulation problem. So, it presents a new distributed positioning method of large-scale wireless sensor networks. This algorithm improves positioning coverage through a multi-hop distance measurement, sets the confidence factor threshold to inhibit the accumulation of errors, adopts combination of trilateration to obtain coordinates of the estimated value of the collection and weighted average method to improve positioning accuracy and simplify the calculation.The application of wireless sensor network in the Coal Mine Safety Monitoring System is discussed and its major task is data collection and transmission. Combined with festures of data collection and transmission, a minimum cost routing mechanism is designed to meet the reliability and dissemination of warning information and dissemination of conventional real-time information on energy-efficiency. In addition, for wireless signal transmission model can be affected by environment, it proposes RSSI ranging model parameters online teaching methods to improve the positioning accuracy of the mine workers.