Research On Localization Algorithm Based On Differential Evolution

Internet of Things(Io T)uses the agreed protocols to transmit and exchange information between various objects and realizes the functions of localization,intelligent identification,tracking and management.It is widely used in smart factories,intelligent robots,precision agriculture and other fields,and precise localization is the key to the application of the Io T services.Therefore,it is of great application value and practical significance to carry out the Io T localization research.The Io T obtains the ranging information and establishes the location equation.The intelligent optimization method is used to solve the equation to obtain the unknown node location.In the intelligent optimization solution method,the differential evolution(DE)localization algorithm is relatively simple and has good search performance,which has been widely concerned.However,it does not fully utilize the distance information between unknown nodes,which results in limited improvement in its localization performance.Therefore,it is necessary to improve the DE localization algorithm to improve its localization performance.In view of this problem,this paper launches the research.The main work is as follows:(1)Aiming at the problem that the distance information used by the traditional DE localization algorithm is incomplete which results in the limited improvement in localization performance,a DE cooperative localization algorithm is proposed.Firstly,the fitness function is improved to make full use of the distance information between the nodes to reduce the localization errors.Then,the trilateral measurement method is used to generate a good initial population to ensure that the DE cooperative localization algorithm has good localization performance.Finally,Lévy flight mode is introduced to reduce the probability of location results falling into local optimum.Simulation results show that DE cooperative localization algorithm can reduce the impact of ranging error on location accuracy and improve the localization performance.(2)The DE cooperative localization algorithm starts from any location and searches for the target location in the location area with a certain probability according to the solution mechanism of "survival of the fittest".However,the algorithm does not use the precise search method,and the location results obtained are not accurate enough,so iterative solution is needed to improve the accuracy.To solve this problem,a fusion localization algorithm based on natural gradient and DE cooperative is proposed.The location obtained by the DE cooperative localization algorithm is used as the initial iterative value,and the iterative solution is performed through natural gradient to improve the accuracy of the localization results.At the same time,the convergence of the localization results is guaranteed by adaptively adjusting the iteration step.Simulation results show that the fusion localization algorithm can further improve the accuracy of location results.(3)In order to test the actual location performance of localization algorithm,a localization platform based on differential evolution is designed using MATLAB software.The localization platform includes two parts: localization interface and localization algorithm,which not only allows users to customize localization parameters,but also calls actual measurement data to complete the calculation of location results and achieve location performance analysis.