| With the continuous development of social economy and the construction of urban and rural areas,the scope of indoor activities continues to expand,and the demand of indoor location service will show explosive growth.In the indoor environment,due to the influence of multipath effect and none-light-of-sight,the GPS signal is attenuated significantly and it is unable to meet the indoor high-precision positioning requirements.At present,there are many positioning technologies in the field of indoor positioning,yet there is still no absolute dominant technology.The i Beacon Bluetooth has attracted much attention since it was proposed because of the unique features such as low power consumption,low cost and easy portability.This paper analyzes domestic and international indoor positioning methods and uses fingerprint localization method to carry out related research.The proposed K-means fingerprint optimization algorithm can effectively balance the relationship between accuracy,stability and real-time performance.The main research content is as follows:(1)In this paper,we analyze the mainstream indoor positioning methods,and use the fingerprint positioning method to conduct related research.For the fault results caused by the surrounding environment when the data is collected,we use Pauta criterion to judge and reject the singular values.Utilizing Kalman filter to decrease the data fluctuation and obtain robust fingerprint information.(2)To obtain efficient and accurate positioning results,we need to collect fingerprint information of high-quality and high-density.To reduce manual sampling,we use inverse distance method to increase fingerprint density.The experimental results show that this method can effectively improve the positioning accuracy and increase the error probability within 0.5m.However,a large amount of fingerprinting data will prolong the matching operation time.In order to ensure efficient and accurate positioning results,the K-means clustering algorithm selects the highest similarity sub-library to match operation.(3)The K-means fingerprint optimization algorithm is proposed in this paper,which mainly solves the problems that the K-means algorithm needs to set the number of clusters and randomly assigns the initial clustering center,and the fixed K in the WKNN algorithm cannot flexibly deal with the relative position of the measured point and the neighboring points.Demonstrate the rationality and feasibility of the proposed algorithm from the perspective of theory and experiment.Results of the experiments show that the proposed optimization algorithm for Euclidean distance error controllable accuracy in the 2D and 3D are respectively 2.5m and 3m.Deviation within 0.5m is over 50%,and the calculation time is shortened by more than 50%.This optimization algorithm can effectively improve the accuracy,stability and real-time performance.(4)In this paper,we developed the indoor positioning service application software ILBS based on the actual requirements.Using MyEclipse2014 as the server development platform,it mainly completes the data interpolation and cluster of the fingerprint database.The results will be feedback to the client,make sure the client can obtain a highly efficient and stable positioning results.Using Android Studio as the client development platform,and application of the K-means fingerprint optimization algorithm to achieve indoor positioning.ILBS mainly complete the specific personal information,data collection,object data and positioning service function.The experimental results show that the proposed optimization algorithm and the designed ILBS can achieve the expected experimental results. |
PDF Full Text Request