Research On Indoor Location Technology Based On Bluetooth Fingerprint And Inertial Sensor

In recent years,with the rise of the mobile Internet and O2 O industry model,the demand for location Based(Location Service,LBS)gradually turn from outdoor to indoor.However,the complexity of the indoor environment poses unprecedented challenges to the indoor positioning.Although there are many kinds of indoor positioning technology,there is no efficient,reliable and universal positioning technology.Low-power Bluetooth technology has the advantages of low cost,low power consumption and easy deployment,which is very popular for popularization and becomes a research hotspot of indoor location.Inertial navigation can provide continuous high-precision short-term positioning,but due to track accumulation error should not be used for a long time.And Bluetooth indoor positioning can provide better positioning accuracy and positioning error is stable.But it is not suitable for dynamic navigation.Positioning accuracy is not as good as inertia in a short time.The advantages and disadvantages of these two positioning techniques are highly complementary.The blended positioning via Bluetooth and inertial navigation can improve positioning performance and is more suitable for dynamic positioning and navigation.In this paper,the influence of distance factor,line-of-sight factor and human interference on the RSSI of Bluetooth is researched simply by collecting the real indoor Bluetooth RSSI(Received Signal Strength Indicator)data.As the Bluetooth location fingerprints do not need to establish the propagation loss model,easy to implement and low cost,this paper selects the location fingerprint location method,and adopts BP neural network as the fingerprint matching method.The simulation results show that the performance of BP neural network is better than traditional K Nearest Neighbor(KNN)and Weighted K Nearest Neighbor(WKNN).At present,the intelligent terminal has integrated Bluetooth module and inertial measurement unit(IMU).It's more universal to achieve the fusion positioning of Bluetooth and inertial navigation by using intelligent terminal.In this paper,using iPhone mobile phone to achieve and verify the walking dead reckoning and strapdown inertial navigation positioning technology.From the step count,step size estimation and walking direction to enhance the positioning effect of walking dead reckoning.For the walking scene,the extended Kalman filter algorithm based on the stationary detection is adopted to suppress the drift of the integral error to improve the positioning performance of the strapdown inertial navigation system.Considering the location power and localization performance,this paper chooses walking dead reckoning as the inertial navigation method,and achieves the fusion location with Bluetooth location fingerprint.In a short time,use the result of inertial navigation which is relatively reliable and accurate.In the long time,through the fusion strategy of the decision-making level,the error of the inertial navigation is corrected periodically.The simulation results of Matlab show that the fusion location scheme can give full play to the advantages of the two positioning techniques.Not only inhibit the accumulation of inertial navigation error,but also greatly enhance the indoor positioning accuracy and stability.Simulation results show that the average positioning accuracy of the fusion scheme is 0.62 m,compared with a single Bluetooth location fingerprint positioning technology,the positioning accuracy is improved by 42.1% and the probability of positioning error is less than 1 m is as high as 95%,which can satisfy the high accuracy of the positioning needs.The proposed localization scheme based on the combination of Bluetooth location fingerprint and pedestrian dead reckoning is feasible to improve the indoor positioning performance,and is more suitable for indoor real-time positioning and dynamic navigation.