Hongji Cao

This paper proposes a robust heading estimation algorithm for smartphones, which integrates the gradient descent (GD) approach and genetic algorithm (GA) to compute the heading angle of smartphones, and reduce the heading error accumulation induced by the errors of sensors. In the proposed algorithm, the accelerometer, magnetometer and gyroscope are used to estimate the heading angle. The sum of squared triaxial errors of geomagnetic strength and acceleration is regarded as the fitness of GA to improve the accuracy of heading angle and restrain the error accumulation. The angular velocity given by gyroscope is utilized to generate the initial point of GA. To improve the convergence rate and local search capability of GA, a search area adjustment strategy based on GD is designed to obtain the optimal search area for each iteration of GA. Experimental results show that the proposed algorithm can achieve better heading estimation accuracy on three smartphones and four orientations, with a mean heading error of 1.577° on Mate 20 smartphone.

Several Wireless Fidelity (WiFi) fingerprint datasets based on Received Signal Strength (RSS) have been shared for indoor localization. However, they can’t meet all the demands of WiFi RSS-based localization. A supplementary open dataset for WiFi indoor localization based on RSS, called as SODIndoorLoc, covering three buildings with multiple floors, is presented in this work. The dataset includes dense and uniformly distributed Reference Points (RPs) with the average distance between two adjacent RPs smaller than 1.2 m. Besides, the locations and channel information of pre-installed Access Points (APs) are summarized in the SODIndoorLoc. In addition, computer-aided design drawings of each floor are provided. The SODIndoorLoc supplies nine training and five testing sheets. Four standard machine learning algorithms and their variants (eight in total) are explored to evaluate positioning accuracy, and the best average positioning accuracy is about 2.3 m. Therefore, the SODIndoorLoc can be treated as a supplement to UJIIndoorLoc with a consistent format. The dataset can be used for clustering, classification, and regression to compare the performance of different indoor positioning applications based on WiFi RSS values, e.g., high-precision positioning, building, floor recognition, fine-grained scene identification, range model simulation, and rapid dataset construction.

Smartphone-based WiFi ranging using fine time measurement (FTM) is severely impacted by Non-line-of-sight (NLoS) environments, which causes significant positioning errors. To address this problem, we propose a novel WiFi FTM positioning (WFP) approach based on the geomagnetism and enhanced genetic algorithm (EGA), which can simultaneously execute WiFi localization and ranging compensation. Based on the distribution of the ranging error in NLoS environments, a semiparametric error model-based ranging compensation method is proposed. To construct the EGA searching model, geomagnetism-based positioning is adopted and fed to the EGA together with the measured WiFi ranging data and the ranging compensation method. During online localization, the EGA model dynamically compensates for the erroneous ranging data until it finds the optimal position. Experimental results show that the ranging and localization accuracy of this EGA-based WFP are 1.33 m and 1.64 m, being an improvement of 30.7% and 56.5% compared to the uncompensated ranging data and the trilateration algorithm using the weighted least square (WLS) method, respectively.

A density-based spatial clustering of applications with noise (DBSCAN) and three distances (TD) integrated Wi-Fi positioning algorithm was proposed, aiming to enhance the positioning accuracy and stability of fingerprinting by the dynamic selection of signal-domain distance to obtain reliable nearest reference points (RPs). Two stages were included in this algorithm. One was the offline stage, where the offline fingerprint database was constructed and the other was the online positioning stage. Three distances (Euclidean distance, Manhattan distance, and cosine distance), DBSCAN, and high-resolution distance selection principle were combined to obtain more reliable nearest RPs and optimal signal-domain distance in the online stage. Fused distance, the fusion of position-domain and signal-domain distances, was applied for DBSCAN to generate the clustering results, considering both the spatial structure and signal strength of RPs. Based on the principle that the higher resolution the distance, the more clusters will be obtained, the high-resolution distance was used to compute positioning results. The weighted K-nearest neighbor (WKNN) considering signal-domain distance selection was used to estimate positions. Two scenarios were selected as test areas; a complex-layout room (Scenario A) for post-graduates and a typical large indoor environment (Scenario B) covering 3200 m². In both Scenarios A and B, compared with support vector machine (SVM), Gaussian process regression (GPR) and rank algorithms, the improvement rates of positioning accuracy and stability of the proposed algorithm were up to 60.44 and 60.93%, respectively. Experimental results show that the proposed algorithm has a better positioning performance in complex and large indoor environments.

In the field of indoor positioning, Wi-Fi FTM is a new technology for realizing high-precision positioning. However, errors caused by clock drift and non-line-of-sight (NLOS) signals affect its positioning accuracy. When receiving NLOS signals, most existing positioning algorithms only delete these signals, which decreases the number of nodes and may decrease accuracy. To address this issue, this paper proposed an adapted weighted positioning method under the NLOS environment. First, this method includes a compensation model to decrease the error caused by clock drift and multipath. Additionally, it can evaluate ranging results and improving the positioning accuracy by assigning greater weight to better ranging results. To verify the effectiveness and feasibility of the proposed method, a positioning experiment is performed under an NLOS environment. The results show that our proposed method is suitable for positioning in a completely NLOS environment and effectively improves the positioning accuracy. Compared with the traditional least squares-based method and the inverse distance weighting-based positioning method, the mean error of the proposed method outperformed by approximately 30% and 20% respectively.

Since many Wi-Fi routers can currently transmit two-band signals, we aimed to study dual-band Wi-Fi to achieve better positioning results. Thus, this paper proposes a fingerprint positioning method for dual-band Wi-Fi based on Gaussian process regression (GPR) and the K-nearest neighbor (KNN) algorithm. In the offline stage, the received signal strength (RSS) measurements of the 2.4 GHz and 5 GHz signals at the reference points (RPs) are collected and normalized to generate the online dual-band fingerprint, a special fingerprint for dual-band Wi-Fi. Then, a dual-band fingerprint database, which is a dedicated fingerprint database for dual-band Wi-Fi, is built with the dual-band fingerprint and the corresponding RP coordinates. Each dual-band fingerprint constructs its positioning model with the GPR algorithm based on itself and its neighborhood fingerprints, and its corresponding RP coordinates are the label of this model. The neighborhood fingerprints are found by the spatial distances between RPs. In the online stage, the measured RSS values of dual-band Wi-Fi are used to generate the online dual-band fingerprint and the 5 GHz fingerprint. Due to the better stability of the 5 GHz signal, an initial position is solved with the 5 GHz fingerprint and the KNN algorithm. Then, the distances between the initial position and model labels are calculated to find a positioning model with the minimum distance, which is the optimal positioning model. Finally, the dual-band fingerprint is input into this model, and the output of this model is the final estimated position. To evaluate the proposed method, we selected two scenarios (A and B) as the test area. In scenario A, the mean error (ME) and root-mean-square error (RMSE) of the proposed method were 1.067 and 1.331 m, respectively. The ME and RMSE in scenario B were 1.432 and 1.712 m, respectively. The experimental results show that the proposed method can achieve a better positioning effect compared with the KNN, Rank, Coverage-area, and GPR algorithms.

Many indoor fingerprinting localization methods are based on signal-domain distances with large localization error and low stability. An improved fingerprinting localization method using a clustering algorithm and dynamic compensation was proposed. In the offline stage, the fingerprint database was built and clustered based on offline hybrid distance and an affinity propagation clustering algorithm. Furthermore, clusters were adjusted using transition regions and a given radius, as well as updating the corresponding position and fingerprint of the cluster centroid. In the online stage, the lost received signal strength (RSS) in the reference fingerprint would be dynamically compensated by using a minimum RSS value, rather than a fixed one. Online signal-domain distance was calculated for cluster identification based on RSS readings and compensated reference fingerprint. Then, K reference points with minimum online signal-domain distances were selected, and affinity propagation clustering was reused by position-domain distances to choose the position-concentrated sub-cluster for location estimation. Experimental results show that the proposed method outperforms state-of-the-art fingerprinting methods, with the mean error of 2.328 m, the root mean square error of 1.865 m and the maximum error of 10.722 m in a testbed of 3200 square meters. The improvement rates, in terms of accuracy and stability, are more than 21% and 13%, respectively.