An Online Radio Map Update Scheme for WiFi Fingerprint-Based Localization

Huang, Baoqi; Xu, Zhendong; Jia, Bing; Mao, Guoqiang

doi:10.1109/jiot.2019.2912808

Cited by 96 publications

(34 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A general DNN does not provide the probability of the predicted value for regression, although it provides the probability of each class for classification. However, several machine learning techniques can provide a probability distribution for regression tasks; for example, the Gaussian process regression (GPR) [36], [37]. By using the GPR or combining a DNN with the GPR, our proposed method is expected to be used for regression tasks.…”

Section: Discussionmentioning

confidence: 99%

Count Estimation With a Low-Accuracy Machine Learning Model

Sei

Ohsuga

2021

IEEE Internet Things J.

View full text Add to dashboard Cite

Many Internet-of-Things (IoT) systems use machine learning techniques, such as deep neural networks. IoT systems can predict attributes, such as age, sex, car speed, human walking speed, and types of animals, using machine learning techniques. Although the functionality of machine learning is undeniable, the prediction accuracy is not always high. When a machine learning model is used to recognize several objects in an object counting system, the estimated count will have a significant error because of the accumulation of the recognition error of each object. In this study, a count estimation method that uses a confusion matrix generated in the training phase was proposed. The proposed method consists of an iterative Bayesian technique with the confusion matrix for count estimation and mitigating over-iterations technique for reducing estimated errors. The proposed method can be used even for a low-accuracy machine learning model. Experiments with synthetic and real data sets were conducted to demonstrate the functionality of the proposed method. The estimation errors of the proposed method were reduced by 64.3% in average compared to the baseline method in the experiments.

show abstract

Section: Discussionmentioning

confidence: 99%

Count Estimation With a Low-Accuracy Machine Learning Model

Sei

Ohsuga

2021

IEEE Internet Things J.

View full text Add to dashboard Cite

show abstract

“…Huang et al [ 36 ] implemented a marginalized particle Gaussian process in combination with pedestrian dead reckoning and Wi-Fi based localization for alignment of measurements with the existing radio map. Results show that implemented solution can provide better localization performance than the Gaussian process regression approach; however, this requires significantly more processing power.…”

Section: Related Workmentioning

confidence: 99%

Algorithm for Dynamic Fingerprinting Radio Map Creation Using IMU Measurements

Brída

Machaj

Racko

et al. 2021

Sensors

View full text Add to dashboard Cite

While a vast number of location-based services appeared lately, indoor positioning solutions are developed to provide reliable position information in environments where traditionally used satellite-based positioning systems cannot provide access to accurate position estimates. Indoor positioning systems can be based on many technologies; however, radio networks and more precisely Wi-Fi networks seem to attract the attention of a majority of the research teams. The most widely used localization approach used in Wi-Fi-based systems is based on fingerprinting framework. Fingerprinting algorithms, however, require a radio map for position estimation. This paper will describe a solution for dynamic radio map creation, which is aimed to reduce the time required to build a radio map. The proposed solution is using measurements from IMUs (Inertial Measurement Units), which are processed with a particle filter dead reckoning algorithm. Reference points (RPs) generated by the implemented dead reckoning algorithm are then processed by the proposed reference point merging algorithm, in order to optimize the radio map size and merge similar RPs. The proposed solution was tested in a real-world environment and evaluated by the implementation of deterministic fingerprinting positioning algorithms, and the achieved results were compared with results achieved with a static radio map. The achieved results presented in the paper show that positioning algorithms achieved similar accuracy even with a dynamic map with a low density of reference points.

show abstract

“…WiFi-based indoor localization mainly includes two types: fingerprinting [12][13][14] and multilateral positioning [15]. Fingerprinting has two stages: the offline stage and the online stage.…”

Section: Introductionmentioning

confidence: 99%

WiFi RTT Indoor Positioning Method Based on Gaussian Process Regression for Harsh Environments

Cao

Wang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

A novel two-way ranging approach was introduced into the Wireless Fidelity (WiFi) standard, and its ranging accuracy reached one meter in a low multipath environment. However, in harsh environments due to multipath or non-line of sight (NLOS), the range measurement based on the WiFi round trip time (RTT) usually has low accuracy and cannot maintain the one-meter accuracy. Thus, this paper proposes an indoor positioning method based on Gaussian process regression (GPR) for harsh environments. There are two stages in the proposed method: construction of a positioning model and location estimation. In the model construction stage, based on known positions of access points (APs), we can determine the position coordinates of some ground points and the reference distances between them and the APs, and the offline ranging difference fingerprints can be generated by the reference distances, which means that there is no need to collect data. Gaussian process regression (GPR) utilizes offline ranging difference fingerprints based on the reference distance to establish a positioning model, and the particle swarm optimization (PSO) algorithm is employed to estimate the GPR hyperparameters. In the location estimation stage, the gathered actual range measurements generate the online ranging difference fingerprint, which is the input data of the positioning model. The output of the model is the estimated position of the smartphone. Experimental results show that the mean errors (MEs) of the proposed method and Least Squares (LS) algorithm are 1.097 and 3.484 meters, respectively, in a harsh environment, and the positioning accuracy of the proposed method improved by 68.5% compared with the LS algorithm.

show abstract

An Online Radio Map Update Scheme for WiFi Fingerprint-Based Localization

Cited by 96 publications

References 27 publications

Count Estimation With a Low-Accuracy Machine Learning Model

Count Estimation With a Low-Accuracy Machine Learning Model

Algorithm for Dynamic Fingerprinting Radio Map Creation Using IMU Measurements

WiFi RTT Indoor Positioning Method Based on Gaussian Process Regression for Harsh Environments

Contact Info

Product

Resources

About