Di Wu

Feature extraction methods have been used to extract location features for indoor positioning in wireless local area networks. However, existing methods, such as linear discriminant analysis and principal component analysis, all suffer from the multimodal property of signal distribution. This paper proposes a novel method, based on enhanced local fisher discriminant analysis (LFDA). First, LFDA is proposed to extract discriminative location features. It maximizes between-class separability while preserving within-class local structure of signal space, thereby guaranteeing maximal discriminative information involved in positioning. Then, the generalization ability of LFDA is further enhanced using signal perturbation, which generates more number of representative training samples. Experimental results in realistic indoor environment show that, compared with previous feature extraction methods, the proposed method reduces the mean and standard deviation of positing error by 23.9% and 33.0%, respectively.

Heading estimation is a central problem for indoor pedestrian navigation using the pervasively available smartphone. For smartphones placed in a pocket, one of the most popular device positions, the essential challenges in heading estimation are the changing device coordinate system and the severe indoor magnetic perturbations. To address these challenges, we propose a novel heading estimation approach based on a rotation matrix and principal component analysis (PCA). Firstly, through a related rotation matrix, we project the acceleration signals into a reference coordinate system (RCS), where a more accurate estimation of the horizontal plane of the acceleration signal is obtained. Then, we utilize PCA over the horizontal plane of acceleration signals for local walking direction extraction. Finally, in order to translate the local walking direction into the global one, we develop a calibration process without requiring noisy compass readings. Besides, a turn detection algorithm is proposed to improve the heading estimation accuracy. Experimental results show that our approach outperforms the traditional uDirect and PCA-based approaches in terms of accuracy and feasibility.