Research On Indoor Positioning Algorithm Based On RFID

People’s demand for indoor positioning is increasing with the development of the times.With the characteristics of low price,strong signal penetration and sustainable use,radio frequency identification(RFID)technology has become a research hotspot in the field of indoor positioning technology.There are numerous unstable factors in indoor environment,which makes the positioning accuracy of the system poor.In order to effectively improve the positioning accuracy of the system,this dissertation researches the basic RFID theory,the LANDMARC positioning system and its improvement,the Gray Wolf optimization algorithm and its improvement,and the combination of the improved Gray Wolf optimization algorithm and the LANDMARC positioning system,and conducts several simulation experiments and test experiments to compare and analyze each experiment.The main contents are as follows:The LANDMARC indoor system is studied,and in order to solve the problem of low positioning accuracy in the LANDMARC system,an improved LANDMARC algorithm is proposed.The algorithm obtains the number of nearest neighbor reference tags that can minimize the positioning error of the system by improving the k-nearest neighbor algorithm;judges the environment and location of the tags to be measured by the judgment mechanism;by improving the weight formula,the weight calculation can be more consistent with the transmission characteristics of the signal.After simulation and comparative analysis,the localization error of the improved algorithm is reduced by about 54.7% compared with that before the improvement.To deal with the problems of slow convergence speed,low convergence accuracy and insufficient population diversity in the late iteration of the gray wolf optimization algorithm,we propose to improve the gray wolf optimization algorithm,which introduces a nonlinear control factor based on power function to improve the convergence accuracy of the algorithm;uses an exponential factor-based position update strategy to improve the convergence speed of the algorithm;adds a multiple position update strategy to reduce the iterative process of the algorithm The algorithm uses an exponential factor-based position update strategy to improve the convergence speed,and incorporates a multiple position update strategy to reduce the randomness in the iteration process.After simulation comparison experiments,the improved gray wolf optimization algorithm has better convergence performance compared to the gray wolf optimization algorithm.To further reduce the localization error,the LANDMARC algorithm based on the improved Gray Wolf optimization is proposed to calculate the position of the tag to be measured by substituting the known variables such as the coordinate position of the reference tag and the reader,the signal strength value of the reference tag measured by the reader and the tag to be measured,which are known in advance or can be obtained by measurement,with the fast computational capability of the improved Gray Wolf optimization algorithm.Simulation experiments show that the localization error of the LANDMARC algorithm based on the improved Gray Wolf optimization is reduced by about 79.9% compared with that of the LANDMARC algorithm,and the excellent localization performance can be maintained in noisy or densely distributed environments with reference tags.The test platform is built in an indoor environment for testing experiments,and the information of the reference tag and the tag to be tested is collected by the test platform,and the improved LANDMARC algorithm and the LANDMARC algorithm based on the improved Gray Wolf optimization are used to test the positioning of the tag to be tested respectively,and are compared and analyzed with the existing algorithm.After the test and comparison,both the improved LANDMARC algorithm and the LANDMARC algorithm based on the improved Gray Wolf optimization can achieve high positioning accuracy and achieve the expected results.