Research Of DV-Hop Localization Algorithm For Wireless Sensor Networks

Node positioning in wireless sensor network is the premise and foundation in many application areas, the research and application is of great significance. Localization algorithm is mainly divided into two categories including range-based localization algorithms and range-free localization algorithms. DV-Hop algorithm belongs to range-free localization algorithms. It has advantages such as convenient node deployment, no need for distance measurement and additional hardwares. Many researchers begin to pay highly attention on it. Traditional DV-Hop localization algorithm has low positioning accuracy. Based on the research of DV-Hop localization algorithm, this thesis proposes the improved algorithm to increase the positioning precision, enhance algorithm stability and practicability.Combined with weighted residual method, a new DV-Hop algorithm is proposed in this thesis. It weights the estimated coordinate residual value of unknown nodes. Considering the inaccuracy of hop count between nodes, the proposed method updates hop count based on Amorphous algorithm. According to simulation, this method can improve localization accuracy. And it is suitable for large-scale and randomly deployed wireless sensor networks. Residual weighting algorithm can reduce large error caused by inaccurate estimated distance between beacon node and unknown node.According to the characteristics of anisotropic sensor networks, a normalized collinearity DV-Hop algorithm for wireless sensor networks is proposed. It analyzes the DV-Hop algorithm error and collinear effect, and uses hop count threshold value and collinearity theory to select eligible beacon nodes for localization. For unknown node belongs to indeterminated beacon node area in one hop distance, the Voronoi diagram is used to segment unknown node area. Then simulation in isotropous and anisotropic networks verifies the advantages of high positioning accuracy and stable performance of the proposed method.In view of the indoor environment, in this thesis, the factors which influence DV-Hop localization precision are analized. Aiming at the condition with the path detour and obstacle existence, a minimum average hop distance threshold method is put forward to choose reliable beacon nodes to calculate average hop distance of the entire network. In order to estimate the unknown node area, the method of threshold is used to select reliable beacon to localization. Considering the lack in number of located beacon node, a reference node selection mechanism is proposed. Then simulation experiment under indoor environment and the comparision between RSSI algorithm and DV-Hop algorithm are carried out. Comparing with RSSI algorithm, the proposed method can obtain similar localization accuracy, and even has better environmental adaptability, especially in large degree of irregularity environment.