Research On Weighted Centroid Localization Algorithm In Wireless Sensor Network

According to whether to measure the distance between the sensor nodes,the localization algorithms in the wireless sensor network node positioning can be divided into two categories: Range-based localization algorithm and Range-free localization algorithm.Range-based localization algorithm has high localization accuracy,but the measuring distance may easily be influenced by the environment and the node energy,resulting in the decrease of algorithm performance.Range-free localization algorithm has low hardware cost and node energy consumption,but the localization accuracy is rather low.An improved weighted localization algorithm based on dynamic anchor nodes(IWLA-DAN)combines the above two categories of localization algorithms for positioning.The anchor nodes are selected by the algorithm according to the theory that the positioning has the least error when the triangle is trilaterally equal constructed by three anchor nodes which can communicate with the unknown node.Then the coordinates of unknown node can be calculated by weighted centroid localization algorithm.At the same time,the unknown node should be updated successively for the pseudo anchor node to participate in the positioning of the next unknown node.In this thesis,the algorithm is studied and improved,and the main contents are as follows.(1)In this thesis,the deficiency of IWLA-DAN algorithm is analyzed and an improved weighted centroid localization algorithm based on distance correction is proposed.Firstly,in order to reduce the ranging error caused by the factors such as environment,a ranging correction method is proposed.The modified node is set as the anchor node nearest to unknown node and the distance error factor between modified node and unknown node is used to modify the ranging values of other anchor nodes.Secondly,the new weights is given considering that the weights of IWLA-DAN algorithm is partial.When the unknown node is located within the triangles constructed by anchor nodes,the three angles of the triangles are taken as the weights instead of the original weights so as to reflect the influence of the triangles with different shapes to the result of positioning.At last,when there are invalid nodes(the unknown nodes which have the following characteristics: there are less than three anchor nodes that can communicate with the unknown node)in the network,an improved calculation method of the jump hop and average jump distance is proposed to search for anchor nodes assistance positioning.The simulation results show that the improved algorithm can obtain better positioning accuracy and stability than IWLA-DAN algorithm.(2)On the basis of the improved algorithm mentioned above,an improved weighted centroid localization algorithm based on virtual anchor nodes is proposed.When the unknown node is located in the triangle constructed by the anchor nodes,the coordinates of the unknown node are calculated.All unknown nodes that meet this condition are treated as virtual anchor nodes to participate in the positioning of the remaining unknown nodes.The simulation results show that the improved weighted centroid localization algorithm based on virtual anchor nodes can get higher accuracy for the location of invalid nodes and better overall positioning results.