Synthetic Sensor Data Generation for Health Applications: A Supervised Deep Learning Approach

2018 ◽  
Author(s):  
Skyler Norgaard ◽  
Ramyar Saeedi ◽  
Keyvan Sasani ◽  
Assefaw H. Gebremedhin
Keyword(s):  
Deep Learning ◽  
Sensor Data ◽  
Learning Approach ◽  
Data Generation ◽  
Health Applications

Sensor Data-Driven UAV Anomaly Detection using Deep Learning Approach

2021 ◽  
Author(s):  
Julio Galvan ◽  
Ashok Raja ◽  
Yanyan Li ◽  
Jiawei Yuan
Keyword(s):  
Deep Learning ◽  
Anomaly Detection ◽  
Data Driven ◽  
Sensor Data ◽  
Learning Approach

A Deep Learning Approach for Speed Bump and Pothole Detection Using Sensor Data

2021 ◽  
pp. 73-85
Author(s):  
Bharani Ujjaini Kempaiah ◽  
Ruben John Mampilli ◽  
K. S. Goutham
Keyword(s):  
Deep Learning ◽  
Sensor Data ◽  
Learning Approach ◽  
Pothole Detection

scholarly journals SSIM—A Deep Learning Approach for Recovering Missing Time Series Sensor Data

2019 ◽  
Vol 6 (4) ◽  
pp. 6618-6628 ◽  
Author(s):  
Yi-Fan Zhang ◽  
Peter J. Thorburn ◽  
Wei Xiang ◽  
Peter Fitch
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Sensor Data ◽  
Learning Approach

scholarly journals Trustworthiness of Dynamic Moving Sensors for Secure Mobile Edge Computing

Computers ◽  
2018 ◽  
Vol 7 (4) ◽  
pp. 63 ◽  
Author(s):  
John Yoon
Keyword(s):  
Deep Learning ◽  
Data Analytics ◽  
Radical Change ◽  
Sensor Data ◽  
Learning Approach ◽  
Meta Data ◽  
Research Areas ◽  
Cloud Server ◽  
Active Research ◽  
Target Environment

Wireless sensor network is an emerging technology, and the collaboration of wireless sensors becomes one of the active research areas for utilizing sensor data. Various sensors collaborate to recognize the changes of a target environment, to identify, if any radical change occurs. For the accuracy improvement, the calibration of sensors has been discussed, and sensor data analytics are becoming popular in research and development. However, they are not satisfactorily efficient for the situations where sensor devices are dynamically moving, abruptly appearing, or disappearing. If the abrupt appearance of sensors is a zero-day attack, and the disappearance of sensors is an ill-functioning comrade, then sensor data analytics of untrusted sensors will result in an indecisive artifact. The predefined sensor requirements or meta-data-based sensor verification is not adaptive to identify dynamically moving sensors. This paper describes a deep-learning approach to verify the trustworthiness of sensors by considering the sensor data only. The proposed verification on sensors can be done without having to use meta-data about sensors or to request consultation from a cloud server. The contribution of this paper includes (1) quality preservation of sensor data for mining analytics. The sensor data are trained to identify their characteristics of outliers: whether they are attack outliers, or outlier-like abrupt changes in environments; and (2) authenticity verification of dynamically moving sensors, which was possible. Previous unknown sensors are also identified by deep-learning approach.

scholarly journals Recognition and Repetition Counting for Complex Physical Exercises with Deep Learning

Sensors ◽  
2019 ◽  
Vol 19 (3) ◽  
pp. 714 ◽  
Author(s):  
Andrea Soro ◽  
Gino Brunner ◽  
Simon Tanner ◽  
Roger Wattenhofer
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Activity Recognition ◽  
Classification Accuracy ◽  
Sensor Data ◽  
Learning Approach ◽  
Physical Exercises ◽  
End To End

Activity recognition using off-the-shelf smartwatches is an important problem in humanactivity recognition. In this paper, we present an end-to-end deep learning approach, able to provideprobability distributions over activities from raw sensor data. We apply our methods to 10 complexfull-body exercises typical in CrossFit, and achieve a classification accuracy of 99.96%. We additionallyshow that the same neural network used for exercise recognition can also be used in repetitioncounting. To the best of our knowledge, our approach to repetition counting is novel and performswell, counting correctly within an error of 1 repetitions in 91% of the performed sets.

Sign in / Sign up

Export Citation Format

Share Document