Base Station Placement Algorithm For Wireless Sensor Network Positioning System

The positioning performance of the wireless sensor network(WSN)positioning system becomes a bottleneck problem which limit its WSN application.So far,the breakthrough works have not happened.The previous works show that the localization performance mainly depends on the following factors: correct measurement parameter estimation,positioning algorithm selection,the geometrical relationship between the target node and the base station(BS).Therefore,for a given WSNs,the positioning performance is only related to the positioning method selection and the base station placement.New base station placement algorithm can improve the positioning performance dramatically.In this paper,the base station placement algorithm in WSN positioning system is studied,the main contributions are as follows:1.The optimal base station placement for single trajectory is proposed.Taking the minimum average of the geometric dilution of precision(GDOP)as the placement rule,the problem of BS placement can be formulated as the optimization model.The descending iterative algorithm is utilized to solve the optimization model which can ensure the accuracy and efficiency of the optimization processing.2.The optimal base station placement algorithm using linear weighted method is proposed for multiple trajectories.In order to obtain the optimal average GDOP of multiple trajectories,the judgment matrix method is adopted for weight parameter calculation.Then according to obtained weight parameters,the multi-objective optimization problem is transformed into a single-objective optimization problem by the linear weighted method.Simulation results show that the proposed algorithm is very suitable for multiple trajectories with different positioning performance requirements.3.The base station placement algorithm with max-min optimization method is proposed for multiple trajectories.The basic idea is utilized the max-min optimization method which belongs to the evaluation function optimization technique to solve the above established multi-objective optimization model.Simulation results illustrate that this algorithm also performs well in multiple trajectories environments.