Research On Multi-objective Positioning Model Based On Error Analysis

DV-Hop is an emblematic range-free positioning algorithm,which has been widely concerned by academia,because of its cost-effective simple and positioning method advantages.Recently,with the increasing demands of users for position accuracy,various improvement strategies based on DV-Hop algorithm have been proposed,and the weighted positioning method is one of the most popular improvements.Currently,there are two research issue of the weighted positioning model: the lack of convergence analysis and the neglect of the relationship between weights and errors.And the work of this research is carried out with such background.In idea circumstances,based on the multi-hop communication principle of sensor nodes in WSNs,the theoretical upper and lower bounds of the estimated distance are obtained when the hop count is n.Additionally,not only the estimated distance calculation model and error model are constructed,but also the error convergence is analyzed when the hop count is n.That is,when the hop value tends to ?,the model error converges to a constant.Finally,the error-oriented weight model is constructed,and the model is solved using GA.The simulation results show that the error-oriented weight model proposed by our work has lower positioning error.In actual communication,due to the limitation of economic cost,the number of nodes deployed in a given area cannot reach the theoretical assumption.It caused the loss of the communication distance of the node during the relay forwarding communication process.For this issue,we abandon the boundary definition method adopted in Chapter 2 for the estimated distance and introduce the concept of average upper boundary.By revising the estimated distance and boundary range when the hop count is n,an improved error-oriented weight model is constructed.The simulation results show that the improved error-oriented weight model constructed in this research has excellent positioning performance.At present,in the single-objective DV-Hop positioning model,improving the positioning accuracy mainly depends on designing efficient search strategies or algorithms.In order to discuss the influence of the positioning model on the error,this research analyzes the defects of the traditional single-objective positioning model and the theoretical average per hop distance of the sensor node,and proposes a multi-objective positioning model.It uses the theoretical average per hop distance to calculate the estimated distance in the second objective,and combines with the traditional objective model to form a multi-objective positioning model.Simulation experiments show that,compared with the single objective positioning model,the multi-objective positioning model has obvious positioning advantages within different network topologies.