User activity detection for massive Internet of things with an improved residual convolutional neural network

2020 ◽  
Author(s):  
Xiaojiang Wu ◽  
Guobing Li ◽  
Guomei Zhang
Keyword(s):  
Neural Network ◽  
Internet Of Things ◽  
Convolutional Neural Network ◽  
Activity Detection ◽  
User Activity

scholarly journals User Activity Detection and Channel Estimation for Grant-Free Random Access in LEO Satellite-Enabled Internet of Things

2020 ◽  
Vol 7 (9) ◽  
pp. 8811-8825 ◽  
Author(s):  
Zhaoji Zhang ◽  
Ying Li ◽  
Chongwen Huang ◽  
Qinghua Guo ◽  
Lei Liu ◽  
...  
Keyword(s):  
Internet Of Things ◽  
Channel Estimation ◽  
Random Access ◽  
Activity Detection ◽  
User Activity ◽  
Leo Satellite

scholarly journals An Integrative Framework for Online Prognostic and Health Management Using Internet of Things and Convolutional Neural Network

Sensors ◽  
2019 ◽  
Vol 19 (10) ◽  
pp. 2338 ◽  
Author(s):  
Yuanju Qu ◽  
Xinguo Ming ◽  
Siqi Qiu ◽  
Maokuan Zheng ◽  
Zengtao Hou
Keyword(s):  
Neural Network ◽  
Big Data ◽  
Internet Of Things ◽  
Convolutional Neural Network ◽  
Health Management ◽  
Added Value ◽  
Time Data ◽  
Integrative Framework ◽  
Operational Level ◽  
Prognostic And Health Management

With the development of the internet of things (IoTs), big data, smart sensing technology, and cloud technology, the industry has entered a new stage of revolution. Traditional manufacturing enterprises are transforming into service-oriented manufacturing based on prognostic and health management (PHM). However, there is a lack of a systematic and comprehensive framework of PHM to create more added value. In this paper, the authors proposed an integrative framework to systematically solve the problem from three levels: Strategic level of PHM to create added value, tactical level of PHM to make the implementation route, and operational level of PHM in a detailed application. At the strategic level, the authors provided the innovative business model to create added value through the big data. Moreover, to monitor the equipment status, the health index (HI) based on a condition-based maintenance (CBM) method was proposed. At the tactical level, the authors provided the implementation route in application integration, analysis service, and visual management to satisfy the different stakeholders’ functional requirements through a convolutional neural network (CNN). At the operational level, the authors constructed a self-sensing network based on anti-inference and self-organizing Zigbee to capture the real-time data from the equipment group. Finally, the authors verified the feasibility of the framework in a real case from China.

scholarly journals Voice activity detection in noisy conditions using tiny convolutional neural network

Informatics ◽  
2020 ◽  
Vol 17 (2) ◽  
pp. 36-43
Author(s):  
R. S. Vashkevich ◽  
E. S. Azarov
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Model Performance ◽  
Voice Activity Detection ◽  
Detection Accuracy ◽  
Activity Detection ◽  
Detection Model ◽  
Proposed Model ◽  
Computational Resources ◽  
Voice Activity

The paper investigates the problem of voice activity detection from a noisy sound signal. An extremely compact convolutional neural network is proposed. The model has only 385 trainable parameters. Proposed model doesn’t require a lot of computational resources that allows to use it as part of the “internet of things” concept for compact low power devices. At the same time the model provides state of the art results in voice activity detection in terms of detection accuracy. The properties of the model are achieved by using a special convolutional layer that considers the harmonic structure of vocal speech. This layer also eliminates redundancy of the model because it has invariance to changes of fundamental frequency. The model performance is evaluated in various noise conditions with different signal-to-noise ratios. The results show that the proposed model provides higher accuracy compared to voice activity detection model from the WebRTC framework by Google.

Sign in / Sign up

Export Citation Format

Share Document