A Novel Distance Estimation Model and Its Use to Node Localization

To improve the positioning accuracy of anisotropic wireless sensor networks (WSNs), a range-free localization algorithm based on polynomial approximation and differential evolution (LA-PADE) is proposed. Firstly, the discrete value of hop count is converted into a more accurate continuous value, reducing the error in hop count calculation. A polynomial is then used to approximate the relationship between hop count and distance among nodes. Finally, the Dierential Evolution (DE) algorithm is applied to obtain the globally optimal solution of objective function corresponding to the estimated position of unknown node, in which the weights of anchor nodes are introduced to embody their importance in the calculation of the coordinates of the unknown nodes. Simulation results show that the proposed algorithm has higher localization accuracy compared to others in different networks.