The Optimization Technology Of Indoor Positioning Based On Location Fingerprinting

Indoor positioning technology means:Obtain location information of a mobile terminal in the indoor environment,and provide the information to mobile users or communications systems.Outdoor positioning technology represented by GPS has become more and more consummate. However, this technique can not be used to locate the position in the indoor environment. With the increasing demand for indoor positioning, high precision, low complexity indoor positioning technology has gradually become a hot spot point of special interest.Indoor positioning technology is mainly composed of three kinds of technologies which are dedicated equipment based positioning technology, Wi-Fi signal ranging based positioning technology and location fingerprint based technology. With easy implementation, low cost and high accuracy,Location fingerprint based technology has become a hot topic and the direction in the field of indoor positioning.But, the fingerprint positioning accuracy is not high enough,the main reason is due to the complexity and particularity of the indoor environment.Therefore, in this thesis we improve the accuracy of location fingerprinting positioning by using the methods of filtering the noise of fingerprint database and non-deployed AP joint positioning.First of all, we analyze the key factors which affect the accuracy of fingerprint position technology by introducing the characteristics of the strength of the received signal, the deployments of AP access points and the selection of the fingerprint algorithm’s parameter.By summarizing the studies that optimize the fingerprint positioning, we get a clear point that location fingerprint is mainly optimized by modulating the matching algorithm, preprocessing of fingerprint library and deploying AP access points to improve the positioning accuracy. The method proposed in this paper is to solve the last two problems.Secondly, in allusion to noise point problem in fingerprint library,we presents a neighborhood mean filter based fingerprint database noise filtering methods referencing the filtering method for image denoising in order to improve the accuracy of location fingerprinting methods.And by designing the simulation experiments, we demonstrate the feasibility and effectiveness of the proposed positioning scheme.Finally,contraposing the optimization of deployment and selection of AP access points, we propose a method of non-binding joint deployment AP positioning, to improve positioning accuracy.As for the problem of decreasing accuracy caused by abnormal alteration of AP access points,we bring in the AP adaptive mechanism to realize the adaptive AP abnormal perception and processing in the environment of non-deployment AP. Through the simulation experiments, we verify the feasibility and effectiveness of the proposed positioning scheme.