Detecting and correcting WiFi positioning errors

Abstract

Recent advances in GPS andWiFi-based positioning technologies for mobile phones have triggered many locationbased services. However, GPS positioning quickly drains a phone's battery and cannot be used indoors. On the other hand, WiFi positioning provides energy-efficient indoor and outdoor positioning with reasonable accuracy. However, WiFi positioning sometimes makes large errors caused by various reasons, e.g., the movement of reference WiFi access points. In this paper we attempt to detect and correct such errors automatically by performing outlier detection in time series. So, we solve this problem by comparing a user's current measurement at time T with her coordinate point at time T predicted from her past coordinate history, and judging whether the current measurement is correct or not by computing the distance between the measurement location and the predicted location. However, it is difficult to predict the user's coordinates accurately with a single prediction method (predictor) because the user's context (e.g., migration speed and sparseness of past coordinates) greatly affects predictor performance. We thus design a context-aware error detection method by employing an ensemble of predictors that have different strengths and weaknesses.

Wi-Fi Fingerprint-Based Indoor Positioning: Recent Advances and Comparisons

“…Due to measurement noise (multipath effects [25]), the target may be mapped to a distant position of similar signal vectors [23], [24], [70]. Higher accuracy can be achieved if the location is estimated by jointly considering temporal or spatial observations:…”

Modern WLAN Fingerprinting Indoor Positioning Methods and Deployment Challenges

“…Radio Map Construction [55], [109], [116], [129]- [142] Outlier Detection [65], [109], [110], [143]- [152] Heterogeneous Devices [153]- [156] heads have the least distances are selected as the coarse location of the user.…”

“…Next, an overview of the schemes for the detection of outliers in the online measurements is provided. Some approaches focus on outlier detection and improve the localization performance of conventional methods [65], [143], [144]. A categorization of the recently proposed outlier detection schemes for WLAN localization is depicted in Fig.…”

Research on Fingerprinting Indoor Positioning Algorithm via Whitening RSS and Fuzzy C-Means Clustering

“…Furthermore, lots of RSS (received signal strength) values containing redundancy, especially, public APs (access points) are deployed densely in which these RSS values are influenced by original information from different Aps [5]. Therefore, reducing the computation complexity without lowering the positioning precision attracts more attention.…”