Research On Node Localization Algorithm For Anisotropic Wireless Sensor Networks

Wireless sensor networks are always used in complex environments, even under the harsh environment that far away from human life. So the sensor nodes are usually randomly thrown by the aircraft, which makes the localization to be a technical difficult problem. Currently, most localization algorithms assume that the network topology is isotropic. However, due to the randomness of the nodes distribution, the complexity of the geographical shape, the unevenness of the node density and other reasons, the actual network topology often exhibits anisotropic characteristics. This makes the traditional localization algorithms’performance rapidly deteriorated, makes the positioning accuracy descended, and makes the network coverage reduced. Therefore, the research on node localization algorithms for anisotropic wireless sensor network with good positioning accuracy has some important implications.Through the study of anisotropic wireless sensor network, the algorithm of HRL (Hops and Range based Localization algorithm for anisotropic networks) is proposed in this thesis. The hops model and range model is proposed in the HRL algorithm. In the hops model, specific to the large error of traditional localization algorithms based on hops, the HRL hop threshold and the HRL hop distance algorithm is designed in this thesis. In the range model, specific to the anisotropic characteristic of signals in anisotropic wireless sensor network, the radio irregular model is designed as the signal propagation model. In order to optimize the ranging signals, the weighted filtering algorithm based on median absolute deviation is proposed. In the distance calculation stage, the signal propagation model parameters recognition algorithm is proposed, so that the sensor nodes can determine the parameter values of the signal propagation model according to their surrounding nodes. In order to improve the positioning accuracy, the anchor node selection strategy is proposed. In the anchor nodes selection stage, the calculation method of collinear degree is designed to exclude the collinear nodes. In the trilateral localization algorithm, an optimized method is given to solve the problem of three circles intersect in a region. Finally, a detailed design of the process of HRL algorithm is given.In this thesis, the HRL algorithm is realized by simulation experiment, the experimental results are analyzed in detail. The effectiveness of HRL algorithm is verified by the simulation results.