Deep Learning-Based Indoor Localization Using Received Signal Strength and Channel State Information

Hsieh, Chaur‐Heh; Chen, Jen‐Yang; Nien, Bo-Hong

doi:10.1109/access.2019.2903487

Cited by 169 publications

(123 citation statements)

References 37 publications

Supporting

Mentioning

105

Contrasting

Order By: Relevance

“…In the case of non-moving objects, the multipath propagation is different depending on the location or the shape of objects where they exist between antennas. This characteristic is used for location estimation and object identification [6][7][8][9][10] among others. In these applications, the main classification algorithms for estimation, such as SVMs and convolutional neural networks (CNNs), are used.…”

Section: Related Studiesmentioning

confidence: 99%

“…In addition to the two types of eigenvalues, α R x T x ,k and β R x T x ,k , the maximum eigenvalues derived from the variance-covariance matrix composed of the CSI amplitude or the phase CSI phase in Equations (10) and (11) are used. This method to create the maximum eigenvalues is the same as that of [11].…”

Section: Calculation Of Feature Values and Human Flow Predictionmentioning

confidence: 99%

“…The CSI consists of frequency and space information, expresses the effect of multipath propagation, and corresponds to the space state between transmitting and receiving antennas. To date, CSI has been used for various studies such as human movement detection, human behavior classification, location estimation, object identification, human counting in a room, and human counting in a passageway [3,[5][6][7][8][9][10][11][12][13][14][15][16][17][18][19].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Wi-Fi CSI-Based Outdoor Human Flow Prediction Using a Support Vector Machine

Ogawa

Munetomo

2020

Sensors

View full text Add to dashboard Cite

This paper proposes a channel state information (CSI)-based prediction method of a human flow that includes activity. The objective of the paper is to predict a human flow in an outdoor road. This human flow prediction is useful for the prediction of the number of passing people and their activity without privacy issues as a result of the absence of any camera systems. In this paper, we assume seven types of activities: one, two, and three people walking; one, two, and three people running; and one person cycling. Since the CSI can effectively express the effect of multipath fading in wireless signals, we expected the CSI to predict the various activities. In our proposed method, the amplitude and phase components are extracted from the measured CSI. The feature values for machine learning are the mean and variance of the maximum eigenvalue derived from the auto-correlation matrix and variance–covariance matrix composed of the amplitude or phase components and the passing time of flow. Using these feature values, we evaluated the prediction accuracy by leave-one-out cross-validation with a linear support vector machine (SVM). As a result, the proposed method achieved the maximum prediction accuracy of 100% for each direction and 99.5% for two directions.

show abstract

Section: Related Studiesmentioning

confidence: 99%

Section: Calculation Of Feature Values and Human Flow Predictionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Wi-Fi CSI-Based Outdoor Human Flow Prediction Using a Support Vector Machine

Ogawa

Munetomo

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…However, the complexity of DNNs is also important since it affects the computational load and implementation cost of a system. Thus, in this paper we use prediction accuracy and network complexity, presented in our previous work [25], as the metrics for the evaluation and comparison of the deep neural networks. The major contributions of this work are summarized as follows:…”

Section: Introductionmentioning

confidence: 99%

Detection of Atrial Fibrillation Using 1D Convolutional Neural Network

Hsieh

Hwang

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

The automatic detection of atrial fibrillation (AF) is crucial for its association with the risk of embolic stroke. Most of the existing AF detection methods usually convert 1D time-series electrocardiogram (ECG) signal into 2D spectrogram to train a complex AF detection system, which results in heavy training computation and high implementation cost. This paper proposes an AF detection method based on an end-to-end 1D convolutional neural network (CNN) architecture to raise the detection accuracy and reduce network complexity. By investigating the impact of major components of a convolutional block on detection accuracy and using grid search to obtain optimal hyperparameters of the CNN, we develop a simple, yet effective 1D CNN. Since the dataset provided by PhysioNet Challenge 2017 contains ECG recordings with different lengths, we also propose a length normalization algorithm to generate equal-length records to meet the requirement of CNN. Experimental results and analysis indicate that our method of 1D CNN achieves an average F1 score of 78.2%, which has better detection accuracy with lower network complexity, as compared with the existing deep learning-based methods.

show abstract

“…Therefore, research often looks into applying remote sensing in this field. It has been shown that indoor localization [6,16,18], measuring physiological signals [11,13,17,20], human identification [2,12], and general human activity recognition/gesture detection [1,7,21,22] are achievable by using CSI. The performances of such systems is comparable to the existing wearable wireless sensor systems.…”

Section: Introductionmentioning

confidence: 99%

Scaling Activity Recognition Using Channel State Information Through Convolutional Neural Networks and Transfer Learning

Brinke

Meratnia

2019

Proceedings of the First International Workshop on Challenges in Artificial Intelligence and Machine Learning for Internet of T

View full text Add to dashboard Cite

Unobtrusive sensing is receiving much attention in recent years, as it is less obtrusive and more privacy-aware compared to other monitoring technologies. Human activity recognition is one of the fields in which unobtrusive sensing is heavily researched" as this is especially important in health care. In this regard, investigating WiFi signals, and more specifically 802.11n channel state information, is one of the more prominent research fields. However, there is a challenge in scaling it up. Transfer learning is rarely applied, and when applied, it is done on filtered/modified data or extracted features. This paper focuses on two aspects. First, convolutional networks are used across multiple participants, days and activities and analysis is done based on these results. Secondly, it looks into the possibility of applying transfer learning based on raw channel state information over multiple participants and activities over multiple days. Results show channel state information is accurate for single participants (F 1-score of 0.90), but sensitive to different participants and fluctuating WiFi signals over days (F 1-score of 0.25-0.35). Furthermore, results show both clustering and transfer learning can be applied to increase the performance to 0.80 when using minimal resources and retraining. CCS CONCEPTS • Computer systems organization → Sensor networks; • Humancentered computing → Ubiquitous and mobile computing; • Computing methodologies → Cross-validation.

show abstract

Deep Learning-Based Indoor Localization Using Received Signal Strength and Channel State Information

Cited by 169 publications

References 37 publications

Wi-Fi CSI-Based Outdoor Human Flow Prediction Using a Support Vector Machine

Wi-Fi CSI-Based Outdoor Human Flow Prediction Using a Support Vector Machine

Detection of Atrial Fibrillation Using 1D Convolutional Neural Network

Scaling Activity Recognition Using Channel State Information Through Convolutional Neural Networks and Transfer Learning

Contact Info

Product

Resources

About