Research On Wi-fi Indoor Location Algorithm Based On CFSFDP And MDS

Wi-Fi indoor positioning technology has benefited from its low-cost and widely-deployed features and is currently the mainstream solution in indoor location services.In recent years,it has become a hot spot in indoor positioning research.However,the off-line phase requires cumbersome data collection and requires constant updating of the fingerprint database to accommodate changes in the indoor environment.Based on existing indoor positioning technology.A crowdsource update fingerprint database method based on CFSFDP algorithm is proposed to ensure the continuous updating of the fingerprint database and improve the robustness of the positioning algorithm.An indoor positioning method based on multidimensional scale analysis technology is proposed to improve the positioning accuracy of the algorithm.Focused on the following aspects.First,a crowdsource update fingerprint database method based on CFSFDP algorithm is proposed.In order to avoid changes in the location of the fingerprint database due to changes in the indoor environment and access point(AP)locations over time,the impact of the fingerprint database failure on the positioning effect is improved.The crowdsource method is used to update the fingerprint database so that the user can enjoy the position.At the same time,the service participates in the maintenance of the positioning system.CFFSDP-based abnormal event identification method is used to determine the user's rationality of feedback data,and the feedback data is retained or discarded.With the constant iteration of the algorithm,the robustness of the positioning system is ensured.Secondly,an indoor positioning algorithm based on multidimensional scale(MDS)is proposed.The Received Signal Strength(RSS)vector at all reference points is collected in the offline phase and its distance matrix is calculated.In the online positioning stage,the distance matrix between the target point RSS and the reference point RSS is calculated,and then the multi-dimensional scale analysis algorithm is used to calculate the relative position coordinates,and then the coordinate conversion algorithm is used to calculate the coordinates of the point to be positioned,in order to avoid signal fluctuations on the positioning effect.In this paper,a four-parameter coordinate transformation method combining difference algorithm and particle swarm optimization was proposed.Finally,the simulation experiment and the actual environment experiment are designed.The feasibility and accuracy of the localization algorithm proposed in this paper are verified.Then the experimental results are analyzed.