DuLoc: Dual-Channel Convolutional Neural Network Based on Channel State Information for Indoor Localization

Song, Xin; Zhou, Yufeng; Qi, Haoyang; Qiu, Weipeng; Xue, Yusheng

doi:10.1109/jsen.2022.3160700

Cited by 11 publications

(6 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, the final map of Tx localization results can be presented to the user as a collision avoidance application. Moreover, recently, research on sensing technology using channel state information (CSI) that can be obtained from Wi-Fi radio waves was conducted [ 19 , 20 , 21 ]. The main disadvantages of Bluetooth and BLE are that they currently do not provide CSI [ 22 ].…”

Section: Discussionmentioning

confidence: 99%

Human-to-Human Position Estimation System Using RSSI in Outdoor Environment

Yamamoto

Yamaguchi

2022

Sensors

View full text Add to dashboard Cite

Methods to prevent collisions between people to avoid traffic accidents are receiving significant attention. To measure the position in the non-line-of-sight (NLOS) area, which cannot be directly visually recognized, position-measuring methods use wireless-communication-type GPS and propagation characteristics of radio signals, such as received signal strength indication (RSSI). However, conventional position estimation methods using RSSI require multiple receivers, which decreases the position estimation accuracy, owing to the presence of surrounding buildings. This study proposes a system to solve this challenge using a receiver and position estimation method based on RSSI MAP simulation and particle filter. Moreover, this study utilizes BLE peripheral/central functions capable of advertising as the transmitter/receiver. By using the advertising radio waves, our method provides a framework for estimating the position of unspecified transmitters. The effectiveness of the proposed system is evaluated in this study through simulations and experiments in actual environments. We obtained an error average of the distance to be 1.6 m from the simulations, which shows the precision of the proposed method. In the actual environment, the proposed method showed an error average of the distance to be 3.3 m. Furthermore, we evaluated the accuracy of the proposed method when both the transmitter and receiver are in motion, which can be considered as a moving person in the outdoor NLOS area. The result shows an error of 4.5 m. Consequently, we concluded that the accuracy was comparable when the transmitter is stationary and when it is moving. Compared with conventional path loss, the model can measure distances of 3 m to 10 m, whereas the proposed method can estimate the “position” with the same accuracy in an outdoor environment. In addition, it can be expected to be used as a collision avoidance system that confirms the presence of strangers in the NLOS area.

show abstract

Section: Discussionmentioning

confidence: 99%

Human-to-Human Position Estimation System Using RSSI in Outdoor Environment

Yamamoto

Yamaguchi

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…A pool of machine learning algorithms are applied to each dataset with the aim of seeing how the implementation of multiple devices can increase accuracy. The machine learning algorithms selected for this paper are Support Vector Machine (SVM) [6], K-Nearest-Neighbours (KNN) [7], Normal Linear Discriminant Analysis [8] and Random Forest [9]. 10-fold cross-validation was used on each of the three datasets.…”

Section: E Machine Learningmentioning

confidence: 99%

Contactless Healthcare Monitoring System Performance Analysis of Multiple Devices

Taylor,

Nioche,

Murray-Smith

et al. 2023

2023 IEEE MTT-S International Microwave Biomedical Conference (IMBioC)

View full text Add to dashboard Cite

Wearable sensors are often used to assist with the healthcare of elderly people within their own homes for detecting incidents such as falls. However, wearable sensors cause numerous issues: they are uncomfortable, they are often forgotten, or they simply run out of battery. One alternative to providing accurate fall detection systems is making use of ambient microwave signals to detect movements. WiFi in particular is a technology that is present in many homes and interestingly, the signal transmission is affected when humans move in between the path of transmission. However, the setup needed for using this as an effective identification method is unclear. This paper focuses here on the performance gain in using multiple devices as transmitters and receivers, instead of only one pair in a simple identification task. We report an increased accuracy in identifying a human subject sitting and standing over a pool of machine learning algorithms when the two Raspberry PI pairs of devices were used in comparison to a single pair of devices.

show abstract

“…Given its objectives and characteristics, it should not surprise the reader that CSI could be very useful when we want to capture changes in an environment. The wireless signals' sensitivity to people reflects variations in CSI that have shown themselves really helpful in indoor localization [31], gesture recognition [32], and user authentication [33].…”

Section: Channel State Informationmentioning

confidence: 99%

A Systematic Review of Wi-Fi and Machine Learning Integration with Topic Modeling Techniques

Atzeni

Bacciu

Mazzei

et al. 2022

Sensors

View full text Add to dashboard Cite

Wireless networks have drastically influenced our lifestyle, changing our workplaces and society. Among the variety of wireless technology, Wi-Fi surely plays a leading role, especially in local area networks. The spread of mobiles and tablets, and more recently, the advent of Internet of Things, have resulted in a multitude of Wi-Fi-enabled devices continuously sending data to the Internet and between each other. At the same time, Machine Learning has proven to be one of the most effective and versatile tools for the analysis of fast streaming data. This systematic review aims at studying the interaction between these technologies and how it has developed throughout their lifetimes. We used Scopus, Web of Science, and IEEE Xplore databases to retrieve paper abstracts and leveraged a topic modeling technique, namely, BERTopic, to analyze the resulting document corpus. After these steps, we inspected the obtained clusters and computed statistics to characterize and interpret the topics they refer to. Our results include both the applications of Wi-Fi sensing and the variety of Machine Learning algorithms used to tackle them. We also report how the Wi-Fi advances have affected sensing applications and the choice of the most suitable Machine Learning models.

show abstract

DuLoc: Dual-Channel Convolutional Neural Network Based on Channel State Information for Indoor Localization

Cited by 11 publications

References 37 publications

Human-to-Human Position Estimation System Using RSSI in Outdoor Environment

Human-to-Human Position Estimation System Using RSSI in Outdoor Environment

Contactless Healthcare Monitoring System Performance Analysis of Multiple Devices

A Systematic Review of Wi-Fi and Machine Learning Integration with Topic Modeling Techniques

Contact Info

Product

Resources

About