Indoor Positioning Based On RSS Fingerprinting

Location fingerprints based indoor positioning, which uses wireless AP Received Signal Strength(RSS), has become a popular research topic during the last a few years.The location fingerprinting technique connects received signal strength(RSS) with grids in the region of interest(ROI) through measuring signals from available APs without knowing their location in advance, and uses these characteristics to infer the location. At present, the location fingerprint positioning has attracted great attention of many researchers. This dissertation makes a deep and systematic study on indoor positioning technology based on RSS and analyzes to improve the location accuracy of the method. Traditional positioning methods based on pattern classification require a large number of classifiers and suffer from high computational burden. Therefore, a novel axial-decoupled method for indoor positioning, which computes the position on X- and Y-axis independently, is proposed in this paper. The purpose is to reduce the decision complexity while keeping high accuracy. The major contributions of this dissertation are:Firstly, give the study on the basis of WLAN indoor positioning system based on RSS. Analysis of the advantages and disadvantages of indoor positioning system based on WLAN, points out that the indoor positioning technology based on location fingerprint is widely used of indoor positioning.Secondly, introduces the related theory of the positioning technology. Briefly, described the composition and the topological structure of the WLAN in this article. On this basis, given the indoor positioning technology with more widely applied in the WLAN environment, and mainly describes the location fingerprint of indoor positioning principle. On this basis, given the current several more widely applied in the WLAN environment of indoor positioning technology, and describes the principles of indoor localization based on location fingerprint. Focused on the analysis and caparisoned of the localization fingerprints algorithm of representative, including nearest neighbor algorithm(NN), K-nearest neighbor algorithm(KNN) and naive Bayesian method. They represent deterministic and probabilistic localization algorithm.Thirdly, this paper implements the indoor positioning algorithm of least squares support vector machines(LS-SVMs) based on location fingerprint. Specific contents including: fingerprint based indoor positioning by LS-SVMs is first proposed. Then, the detailed LS-SVM training process of fingerprinting samples is given. Next, described in how to transfer the positioning problem to a multi-class classification problem, which is handled by one-against-one(OAO) and one-against-all(OAA) approach respectively. Simulation results show that the proposed method has higher accuracy and Small computational cost among the Support Vector Machines(SVMs) and the traditional k-nearest neighbors(K-NNs).Finally, this paper realizes the indoor positioning technique of axial decoupled based on RSS fingerprint. Specifically, the framework of fingerprinting localization based on axial decoupled is given firstly. Secondly, the decoupled training and positioning process by fingerprinting samples is described in detail. Lastly, pattern classifiers including the LS-SVM, SVM and K-NN are embedded in the proposed framework. Experimental results are included to demonstrate the effectiveness of the proposed axial-decoupled positioning method.