Research On Ultra-wideband Indoor Positioning Based On Swarm Intelligence Algorithm

Ultra-wideband(UWB)positioning technology is one of the most promising indoor wireless positioning technologies for its typical centimeter-level positioning accuracy,ultra-high time resolution and strong multipath resolution.However,the indoor environment is generally complex and easily affected by the Non-line-of-sight(NLOS)environment.There are still some problems in the accuracy and stability of the positioning system based on UWB.In wireless sensor networks,determining the location of target individual is usually divided into two stages: ranging and location calculation.In this thesis,the wireless location system based on swarm intelligence algorithm in location calculation is deeply studied.We improved the sparrow search algorithm in swarm intelligence algorithm.Firstly,we adopt homogeneous design idea and use tent chaotic map to population initialization.Secondly,the“recognition mechanism” was proposed,which gives the individual sparrow recognition thinking and only changes when they recognizes the change,rather than blindly changing to the best individual.Finally,the elite individual disturbance mechanism based on Cauchy mutation was added and the out-of-bounds reset processing method was improved to enhance the performance of the algorithm.The function test results show that the improved Sparrow search algorithm has faster search speed and higher accuracy,which verifies the reliability of the algorithm.The improved sparrow search algorithm was used as the final location algorithm.The location algorithm based on analytical expressions is too rigid in the solution process,which is greatly influenced by data errors.However,recursive algorithms such as sparrow search algorithm have a process for solving the answer,which can make better use of redundant information and correct abnormal data.In addition,in order to make the solution closer to the application,the bounding box method was combined to improve the search efficiency,and a fitness function related to positioning was designed.The final simulation results show that the positioning accuracy of this algorithm was improved compared with the positioning algorithm based on analytical expressions.In order to verify the effectiveness of the algorithm,we have made a wireless sensor positioning device and written the program of the algorithm.Firstly,building a wireless sensor network based on UWB.Secondly,the factors such as the multipath effect,the NLOS error,the clock error,the reference base station error,and the equipment error in engineering practice are analyzed,and the corresponding error optimization scheme was given and applied.Finally,in two indoor environments,the static and dynamic tests of the target to be located were carried out.The final experimental results show that the positioning algorithm proposed in this thesis is feasible.