1 April 2021 in IEEE Internet of Things Journal
Recently, round-trip time (RTT) measured by a fine-timing measurement protocol has received great attention in the area of WiFi positioning. It provides an acceptable ranging accuracy in favorable environments when a line-of-sight (LOS) path exists. Otherwise, a signal is detoured along with non-LOS (NLOS) paths, making the resultant ranging results different from the ground truth, called an RTT bias, which is the main reason for poor positioning performance. To address it, we aim at leveraging the user mobility trajectory detected by a smartphone’s inertial measurement units, called pedestrian dead reckoning (PDR). Specifically, PDR provides the geographic relation among adjacent locations, guiding the resultant positioning estimates’ sequence not to deviate from the user trajectory. To this end, we describe their relations as multiple geometric equations, enabling us to render a novel positioning algorithm with acceptable accuracy. Depending on the mobility pattern being linear or arbitrary, we develop different algorithms divided into two phases. First, we can jointly estimate an RTT bias of each access point (AP) and the user’s step length by leveraging the geometric relation mentioned above. It enables us to construct a user’s relative trajectory defined on the concerned AP’s local coordinate system. Second, we align every AP’s relative trajectory into a single one, called trajectory alignment , equivalent to transformation to the global coordinate system. As a result, we can estimate the sequence of the user’s absolute locations from the aligned trajectory. Various field experiments extensively verify the proposed algorithm’s effectiveness that the average positioning error is approximately 0.369 (m) and 1.705 (m) in LOS and NLOS environments, respectively.
28 July 2023 in IEEE Sensors Journal
Pedestrian dead reckoning (PDR) is one key localization technique using inertial measurement unit (IMU) installed on a smartphone. However, IMU measurements are affected by the user’s unpredictable smartphone-carrying pattern, degrading the resulting localization accuracy and making its practical usage questionable. This paper aims to tackle the issue by proposing a novel technique called Waveform GuIded Transformation of IMU measurements (WAIT), which transforms real IMU measurements to error-free waveforms. WAIT is based on a Resnet-based auto-encoder architecture that is trained to generate the waveforms no matter what smartphone-carrying pattern the IMU measurement comes in. The resulting consistent waveform can be used as input for various elementary algorithms required for updating a user’s location and pedestrian trajectory, such as step count and heading direction estimation algorithms. Besides, WAIT’s architecture remains unchanged regardless of the number of concerned smartphone-carrying patterns, and there is no performance degradation. The effectiveness of WAIT is well verified by field experiments conducted with three different smartphone-carrying patterns and two users, showing that the average positioning error is reduced from 0.967 (m) to 0.433 (m) compared to benchmark PDR algorithm using a different DNN model for each smartphone-carrying pattern.
13 June 2024 in IEEE Journal on Selected Areas in Communications
A smartphone-based user mobility tracking could be effective in finding his/her location, while the unpredictable error therein due to low specification of built-in inertial measurement units (IMUs) rejects its standalone usage but demands the integration to another positioning technique like WiFi positioning. This paper aims to propose a novel integration technique using a graph neural network called Mobility-INduced Graph LEarning (MINGLE), which is designed based on two types of graphs made by capturing different user mobility features. Specifically, considering sequential measurement points (MPs) as nodes, a user’s regular mobility pattern allows us to connect neighbor MPs as edges, called time-driven mobility graph (TMG). Second, a user’s relatively straight transition at a constant pace when moving from one position to another can be captured by connecting the nodes on each path, called a direction-driven mobility graph (DMG). Then, we can design graph convolution network (GCN)-based cross-graph learning, where two different GCN models for TMG and DMG are jointly trained by feeding different input features created by WiFi RTTs yet sharing their weights. Besides, the loss function includes a mobility regularization term such that the differences between adjacent location estimates should be less variant due to the user’s stable moving pace. Noting that the regularization term does not require ground-truth location, MINGLE can be designed under semi- and self-supervised learning frameworks. The proposed MINGLE’s effectiveness is extensively verified through field experiments, showing a better positioning accuracy than benchmarks, say mean absolute errors (MAEs) being 1.510 (m) and 1.077 (m) for self- and semi-supervised learning cases, respectively.
20 September 2024 in IEEE Transactions on Mobile Computing
Due to the emergence of various wireless sensing technologies, numerous positioning algorithms have been introduced in the literature, categorized into geometry-driven positioning (GP) and data-driven positioning (DP). These approaches have respective limitations, e.g., a non-line-of-sight issue for GP and the lack of a high-dimensional and labeled dataset for DP, which could be complemented by integrating both methods. To this end, this paper aims to introduce a novel principle called combinatorial data augmentation (CDA), a catalyst for the two approaches’ seamless integration. Specifically, GP-based data samples augmented from different positioning element combinations are called preliminary estimated locations (PELs), which can be used as high-dimensional inputs for DP. We confirm the CDA’s effectiveness from field experiments based on WiFi round-trip times (RTTs) and inertial measurement units (IMUs) by designing several CDA-based positioning algorithms. First, we show that CDA offers various metrics quantifying each PEL’s reliability, thereby extracting important PELs for WiFi RTT positioning. Second, CDA helps compute the observation error covariance matrix of a Kalman filter for fusing two position estimates derived by WiFi RTTs and IMUs. Third, we use the important PELs and the above position estimate as the corresponding input feature and the real-time label for fingerprint-based positioning as a representative DP algorithm. It provides accurate and reliable positioning results, with an average positioning error of 1.58 (m) and a standard deviation of 0.90 (m).