Study On Indoor Localization Algorithm And System Technology Based On Bluetooth 4.0

With the wide application of Location-Based Service,indoor positioning has brought great convenience to people,so the indoor positioning will certainly be commercialized.However,the indoor environment is complex and changeable.Most of the existing indoor positioning cost is higher,and the positioning accuracy and coverage range are limited.Bluetooth 4.0 has the characteristics of low power consumption,low cost,easy deployment and long distance transmission,which is highly consistent with the trend of indoor positioning.Therefore,the propagation characteristics,positioning algorithm and positioning system design of Bluetooth 4.0 are studied in the thesis.Firstly,the Bluetooth signal is collected and the effect of different environmental factors on the propagation characteristics of Bluetooth 4.0 is analyzed in the indoor environment,and the optimized propagation model is fitted according to the measured Bluetooth 4.0 signal.At the same time,the Gauss filtering algorithm is used to preprocess the RSSI signal to improve the accuracy of RSSI measurement.Secondly,aiming at the problem of the larger workload of off-line database building in the location fingerprint matching algorithm,a fast virtual grid matching positioning algorithm based on Pearson correlation coefficient is proposed.First,the Bounding-Box method is used to determine the initial virtual grid area;Then the Pearson correlation coefficient set between the RSSI vector and the distance logarithm vector is calculated;Finally,k nearest neighbor coordinates with the correlation value close to-1 are selected and the unknown node position is determined by taking the correlation coefficient as the weight value.In order to improve the location rate,the mesh subdivision is carried out with the iterative subdivision algorithm.The simulation results show that the algorithm not only does not need to establish the fingerprint database,but also has high positioning accuracy and positioning rate in the Gauss environment.Thirdly,aiming at the problem that the NLOS error has a serious impact on the positioning in the non-line-of-sight environment,an improved PSO positioning algorithm based on feature point correction is proposed.A few feature points are selected and the fingerprint database is set up in the off-line phase.In the online positioning stage,the initial positoning estimation and the feature point fingerprint are used to fit the signal propagation model by the weighted least squares algorithm.and the difference correction is carried out to improve the accuracy of the pseudo range.The final positioning estimation is implemented by linearizing the fitness function of the PSO positioning algorithm.The simulation results show that the algorithm can effectively weaken the influence of NLOS error on positioning,achieve high precision positioning,and have strong anti-noise ability.Finally,the Bluetooth 4.0 positioning system is built in the indoor environment,the Bluetooth beacon is deployed and configured,and the AT command is used to control the Bluetooth 4.0 module to complete the acquisition of RSSI signals.Positioning experiment test is carried out the in the indoor Gauss noise environment and the NLOS environment respectively.The experimental results show that the positioning system has high positioning accuracy and good anti-noise performance,and the positioning requirement can be met in the actual environment.