The Research Of Indoor Positioning Algorithm Based On WiFi And Inertial Sensor

In the social background of rapidly developing information technology such as Internet of Things,big data,cloud computing,etc.the application of location-based services has gradually increased in the proportion of people's lives.Indoor positioning technology,as the core of LBS system,has important research significance.Accuracy,cost,time consumption,scalability and so on are important indexes for measuring the performance of the positioning system.A single positioning technology cannot meet the multi-dimensional requirements of positioning at the same time.WiFi-based indoor positioning technology has become the first solution due to WiFi technology's advantages of wide distribution in indoor environment,meter-level positioning accuracy,and low development cost;PDR technology based on inertial sensors has good short-term accuracy and features that are not easily affected by external environmental interference,which provide the basis for integration with other technologies.This article focuses on the key technologies of WiFi positioning and PDR positioning,designs a fusion positioning algorithm based on WiFi and inertial sensor,proposes an improved schemes from the three stages of WiFi position fingerprinting,PDR positioning,and fusion positioning.Finally,based on the designed algorithm,the empty nest elderly person monitoring system based on Android platform is designed and developed.The main work of the dissertation is as follows:By analyzing the characteristics of the WiFi signal,the reason for the fingerprint positioning error is clarified.Aiming at the large number of fingerprint matching reference points which are searched in the online phase,an affine propagation clustering algorithm is used to optimize the offline fingerprint database.In order to make the reference point clustering in the offline fingerprint database more uniform and reasonable,an improved signal distance fusion reference point physics is proposed as a clustering feature.In the online matching stage,an improved weighted K-nearest neighbor method based on the sampling intensity of the reference point signal is proposed to improve the positioning accuracy.In order to reduce the interference caused by the pseudo-peaks of acceleration during pedestrian walking on the pedometer and improve the accuracy of the step-rate detection,a step-frequency detection algorithm based on the threshold value of the acceleration peak-valley difference threshold was designed.In the process of pedestrians' movement,the result of WiFi positioning is discontinuous and the jumps are large.However,using PDR positioning results in cumulative errors.By analyzing the different characteristics of WiFi position fingerprinting and PDR positioning algorithm,the fusion strategy is proposed to perform PDR segment by segment and average continuous processing of Wi-Fi positioning results,using PDR to compensate for the regular updating of pedestrian location.In order to verify the effectiveness of the algorithm,a large amount of data was measured and verified.The experimental results show that the designed algorithm can effectively improve the positioning accuracy and reduce the computational complexity.Finally,combined with the indoor positioning algorithm based on WiFi and inertial sensor fusion proposed in this paper,the empty nest elderly care and care system is developed on the Android platform to complete the real-time positioning of the elderly location,indoor danger zone settings,danger zone residency overtime warning,topic template corpora pushing and other basic functional modules to achieve a multi-dimensional care for the security and emotions of empty nesters.