Research On Indoor Fingerprint Map Location Algorithm Based On RSSI

In today's world,information technology is developing rapidly.With the in-depth development of the Internet,technologies such as the Internet of Things and 5G are gradually maturing,and location-based services will play an increasingly important role.As an application based on location services,indoor positioning gradually penetrates into all aspects of social life.However,indoor positioning technology still has many problems,such as low positioning accuracy,high positioning cost,and high power consumption,etc.In order to improve indoor positioning accuracy,optimize positioning cost,and seek balance between positioning accuracy and positioning cost,this paper will explore some aspects of indoor positioning node layout and positioning recognition algorithm.Indoor positioning based on zigbee received signal strength indication is receiving more and more attention due to low cost,low hardware power consumption,and easy implementation.In order to improve RSSI(Received Signal Strength Indication)accuracy and optimize node resource configuration,an equal-arc trilateral localization algorithm based on RSSI is proposed to improve the accuracy of measurement and improve the layout of beacon nodes.The algorithm analyzes the relationship between the different communication distance and the RSSI,and can choose the best communication distance to meet the requirements according to the different application scenes.And equal-arc trilateral distribution model is used to deal with the distribution of beacon nodes to ensure that the tracks of unknown nodes are always kept within the best communication distance,thus improving the measurement precision.The experimental results in the actual scene show that the equal-arc trilateral distribution proposed in this paper can improve the measurement accuracy of RSSI and improve the positioning performance.In order to improve the accuracy of location recognition,this paper proposes an indoor positioning method GA-SVR based on genetic algorithm(GA)optimization multiple parameters of support vector regression(SVR).The algorithm is divided into two stages: offline acquisition and online prediction.The offline acquisition relies on the layout of the equal-arc trilateral nodes to establish the fingerprint database.The online prediction is based on the training model for position prediction.Firstly,the all collected data is processed by Kalman filter,and then the penalty parameter C,radial basis function kernel width ? and loss function variable ? of support vector regression are optimized by genetic algorithm,so that the support vector regression reaches the best position prediction performance.The experimental results in the actual scene show that the proposed GA-SVR algorithm has better position prediction ability.In general,the equal-arc trilateral distribution and GA-SVR algorithm proposed in this paper have superior positioning performance,and its comprehensive positioning capability reaches 0.516 m,which has certain reference significance in indoor positioning algorithm.