Yubin Xu

Optimal KNN Positioning Algorithm via Theoretical Accuracy Criterion in WLAN Indoor Environment

This paper proposes the optimal K nearest neighbors (KNN) positioning algorithm via theoretical accuracy criterion (TAC) in wireless LAN (WLAN) indoor environment. As far as we know, although the KNN algorithm is widely utilized as one of the typical distance dependent positioning algorithms, the optimal selection of neighboring reference points (RPs) involved in KNN has not been significantly analyzed. Therefore, in order to fill this gap, the optimal KNN positioning algorithm based on the best TAC is introduced. And this algorithm is beneficial to construct the reliable WLAN indoor positioning system and provide the efficient location based services (LBSs). The relationship among theoretical expectation accuracy, unit interval of neighboring RPs and dimensions of target location region is also revealed. Furthermore, the feasibility and effectiveness of optimal KNN positioning algorithm are verified based on the experimental comparisons respectively in the regular office room, straight corridors, static positioning and dynamic tracking situations.

Indoor positioning via nonlinear discriminative feature extraction in wireless local area network

The essential challenge in wireless local area network (WLAN) positioning system is the highly uncertainty and nonlinearity of received signal strength (RSS). These properties degrade the positioning accuracy drastically, as well as increasing the data collection cost. To address this challenge, we propose the nonlinear discriminative feature extraction of RSS using kernel direct discriminant analysis (KDDA). KDDA extracts location features in a kernel space, where the nonlinear RSS patterns are well characterized and captured. By performing KDDA, the discriminative information contained in RSS is reorganized and maximally extracted, while redundant features or noise are discarded adaptively. Furthermore, unlike previous monolithic models, we employ a location clustering step to localize the feature extraction. This step effectively avoids the suboptimality caused by variability of RSS over physical space. After feature extraction in each subregion, the relationship between extracted features and physical locations is established by support vector regression (SVR). Experimental results show that the proposed approach obtains higher accuracy while reducing the data collection cost significantly.

Dynamic Radio Map Construction for WLAN Indoor Location

A novel indoor location algorithm based on dynamic Radio Maps construction in wireless local area network (WLAN) is proposed. The limitation of previous static Radio Map method is that reconstruction work must be taken to adapt the variation of indoor wireless environment. By taking received signal strength (RSS) values varying over time and space into account, a dynamic Radio Map is constructed to avoid this work and thus the cost for location system is reduced significantly. During offline phase, the relationship between RSS of calibration points and reference points is established by artificial neural network (ANN), and then during online phase, the real-time RSS values at reference points are predicted based on the RSS collected at calibration points in real time. Through the simulation of nearest neighbor (KNN) indoor location method for location accuracy, the feasibility and effectiveness of dynamic Radio Map construction based indoor location method is verified.

Received Signal Strength Recovery in Green WLAN Indoor Positioning System Using Singular Value Thresholding

Green WLAN is a promising technique for accessing future indoor Internet services. It is designed not only for high-speed data communication purposes but also for energy efficiency. The basic strategy of green WLAN is that all the access points are not always powered on, but rather work on-demand. Though powering off idle access points does not affect data communication, a serious asymmetric matching problem will arise in a WLAN indoor positioning system due to the fact the received signal strength (RSS) readings from the available access points are different in their offline and online phases. This asymmetry problem will no doubt invalidate the fingerprint algorithm used to estimate the mobile device location. Therefore, in this paper we propose a green WLAN indoor positioning system, which can recover RSS readings and achieve good localization performance based on singular value thresholding (SVT) theory. By solving the nuclear norm minimization problem, SVT recovers not only the radio map, but also online RSS readings from a sparse matrix by sensing only a fraction of the RSS readings. We have implemented the method in our lab and evaluated its performances. The experimental results indicate the proposed system could recover the RSS readings and achieve good localization performance.

Kalman/Map Filtering-Aided Fast Normalized Cross Correlation-Based Wi-Fi Fingerprinting Location Sensing

A Kalman/map filtering (KMF)-aided fast normalized cross correlation (FNCC)-based Wi-Fi fingerprinting location sensing system is proposed in this paper. Compared with conventional neighbor selection algorithms that calculate localization results with received signal strength (RSS) mean samples, the proposed FNCC algorithm makes use of all the on-line RSS samples and reference point RSS variations to achieve higher fingerprinting accuracy. The FNCC computes efficiently while maintaining the same accuracy as the basic normalized cross correlation. Additionally, a KMF is also proposed to process fingerprinting localization results. It employs a new map matching algorithm to nonlinearize the linear location prediction process of Kalman filtering (KF) that takes advantage of spatial proximities of consecutive localization results. With a calibration model integrated into an indoor map, the map matching algorithm corrects unreasonable prediction locations of the KF according to the building interior structure. Thus, more accurate prediction locations are obtained. Using these locations, the KMF considerably improves fingerprinting algorithm performance. Experimental results demonstrate that the FNCC algorithm with reduced computational complexity outperforms other neighbor selection algorithms and the KMF effectively improves location sensing accuracy by using indoor map information and spatial proximities of consecutive localization results.

KNN-FCM hybrid algorithm for indoor location in WLAN

As a fingerprint match method, k-nearest neighbors (KNN) has been widely applied for indoor location in Wireless Local Area Networks (WLAN), but its performance is sensitive to number of neighbors k and positions of reference points (RPs). So fuzzy c-means (FCM) clustering algorithm is applied to improve KNN, which is the KNN-FCM hybrid algorithm presented in this paper. In the proposed algorithm, through KNN, k RPs are firstly chosen as the data samples of FCM based on received signal strength (RSS). Then, the k RPs are classified into different clusters through FCM based on RSS and the position coordinates. According to the rules proposed in this paper, some RPs are reselected for indoor location in order to improve the location precision. Simulation results indicate that the proposed KNN-FCM hybrid algorithm generally outperforms KNN when the location error is less than 2m.

An Indoor Positioning Algorithm with Kernel Direct Discriminant Analysis

Location estimation based on received signal strength (RSS) in WLAN environment is an attractive method for indoor positioning system. Unfortunately, due to the explicit nonlinearity and uncertainty of RSS signal, the traditional approaches always fail to deliver good location accuracy. This paper presents a novel positioning algorithm with kernel direct discriminant analysis (KDDA). We deploy the KDDA to map the original RSS vectors into a kernel feature space for feature extraction. The experimental results show that the proposed algorithm leads to higher location accuracy over the traditional algorithms including weighted k-nearest neighbor, maximum likelihood and kernel method. The performance improvement can be attributed to that the nonlinear discriminative location information can be efficiently extracted, while the redundant location information is considered as noise and discarded adaptively.

Physical Distance vs. Signal Distance: An Analysis towards Better Location Fingerprinting

The laborious collection of location fingerprints, that could also potentially change with time, remains a hurdle towards the widespread deployment of indoor and campus area positioning using WiFi. In this paper, we present a preliminary analysis of the complicated relationship between distance in signal space, the physical distance and location errors towards better guidelines for fingerprint collection. We introduce the idea of entropy of location fingerprints and investigate the relationships between physical and signal distances, entropy, and expected errors with positioning using location fingerprinting. We present results with no access point and one access point and consider variations in the density of reference points and the standard deviation of signal strength to illustrate the issues.

WLAN indoor positioning algorithm based on sub-regions information gain theory

As a very popular positioning system, WLAN positioning attracts widely researches and investigations throughout the world. It implements the fingerprint technique to realize indoor navigation. The fingerprinting technique which employs the KNN algorithm has to make use of RSS (Received Signal Strength) from the Access Points (APs) without any classification. However, not all of the APs provide the contributions but interference, which will not only burden the positioning system but also results in poor positioning accuracy. To increase the positioning accuracy and decrease the computation cost of Wireless Local Area Network (WLAN), a novel algorithm is proposed by implementing the KNN algorithm and Information Gain Theory to bridge the gap between the Access Point selection and positioning accuracy. The experiment results indicates that, the positioning accuracy is improved by 4.76% within 2m, and meanwhile the time for positioning is decreased by 23.81%, which means the proposed algorithm successfully achieves higher positioning accuracy with less computational cost. Besides, it is also proved in this paper that contrary to the traditional concept, more APs do not always mean higher positioning accuracy; on the opposite, in our experimental environment, a relatively small scale of APs can achieve higher positioning accuracy than that of large scale of APs.

Intelligent AP selection for indoor positioning in wireless local area network

Indoor positioning system in wireless local area network (WLAN) has been receiving increasing interest in pervasive computing applications. To keep balance between energy consumption on client device and positioning accuracy, AP selection strategy is always proposed to select the most discriminant APs for positioning. In this paper, we propose an intelligent AP selection method based on joint location information gain. In contrast to traditional AP selection methods which measure the discriminant ability of APs independently, we consider it jointly. By considering the correlation of the discriminant ability between different APs, more accurate measure of the discriminant ability can be taken. Besides, support vector regression (SVR) positioning algorithm is combined to estimate the location. Experiments are carried in a realistic WLAN indoor environment. Experimental results show that, by using the intelligent AP selection method, the proposed positioning algorithm maintains a high-level accuracy while reducing the energy consumption on client device significantly.