Explore chapters and articles related to this topic
Applications of IoT through Machines' Integration to Provide Smart Environment
Published in Nishu Gupta, Srinivas Kiran Gottapu, Rakesh Nayak, Anil Kumar Gupta, Mohammad Derawi, Jayden Khakurel, Human-Machine Interaction and IoT Applications for a Smarter World, 2023
Manikandan Jagarajan, Ramkumar Jayaraman, Amrita Laskar
Fog architecture makes use of end-device facilities (switches, routers, multiplexers, and so on) for computing, storage, and processing. Fog computing architecture is made up of physical and logical network components, software, and hardware that come together to form a wide network of interconnected devices. It is used to transport data from the data center to the server's edge. Fog is being implemented in many fields like e-health, smart traffic management, and military applications [9,10]. In edge computing, the processing of data is run at edge devices themselves or the nearby edge servers, which will be physically close to the end devices. Autonomous cars face recognition application with reduced response time [11] for several applications.
Deep-Learning-Empowered Edge Computing-Based IoT Frameworks
Published in Bharat Bhushan, Sudhir Kumar Sharma, Bhuvan Unhelkar, Muhammad Fazal Ijaz, Lamia Karim, Internet of Things, 2022
Mithra Venkatesan, Anju V. Kulkarni, Radhika Menon
IoT-based sensors and end devices are capable of generating volumes of data which can be processed by deep learning-based models. Edge computing involves operating the computing nodes near the edge device and is capable of performing high computation and has low latency requirements. This greatly enhances scalability, privacy and efficiency in terms of bandwidth. Chen and Ran [32] worked in providing insights on the intersection of edge computing and deep learning. The paper has provided comprehensive overview on how deep learning can be applied at the edge of the network along with methods that can be used for deep learning interference execution across different end devices, servers and cloud. Various popular deep learning models can be applied across many edge devices. The challenges involved in terms of performance of the system, different technologies of the network and privacy has also been discussed. The methodology towards increasing the deep learning interference and performing training distributing on edge devices was also detailed in the paper.
Industrial Internet of Things (IIoT)
Published in Chanchal Dey, Sunit Kumar Sen, Industrial Automation Technologies, 2020
Edge computing can be considered to be a category or subset of fog computing. Fog refers basically to the network that connects edge to the cloud. Edge computing essentially refers to the processing of data being done close to where it is created, i.e., the edge devices. Fog, on the other hand, refers to the hub of network connections that exists between the edge and the cloud. Hence, fog is more about the way data is processed and the manner in which it is transported from the source to its destination. The three computing paradigm along with the jobs they perform, are shown in Figure 7.15.
A secured and optimized deep recurrent neural network (DRNN) scheme for remote health monitoring system with edge computing
Published in Automatika, 2023
D. Pavithra, R. Nidhya, S. Shanthi, P. Priya
In the e-healthcare sector, many technologies are used such as machine learning, fog computing, cloud computing, Internet of Things are integrated to provide real-time solutions. Medical diagnosis along with artificial neural networks is one of the key research areas in the field of health care. A type of cloud computing termed “edge computing” uses data being gathered using an IoT network and by using various edge devices they are computed locally. Once the data processing is completed, for further storage and calculations, the data is moved to the cloud. Machine Learning, In several domains, including image recognition, deep learning techniques are frequently employed, in decision making, object recognition, and disease diagnosis. In the current scenario, the amount of data generated has become too large and the hidden patterns and abnormal patterns need to identify very efficiently. In the case of Remote monitoring system, the patient data should be analysed and an accurate result and diagnosis of the disease should be predicted. In order to predict the disease many algorithms are used such as K Nearest Neighbor, Convolutional neural network, Bayes Network, and Support vector machine. In this healthcare sector, the data is sensitive and need to be secured more efficiently starting from the Human body, Local Server, Communication network and then the Cloud Storage [6]. So Data needs to be highly secured and confidential. So in order to secure the data while reception and transmission many algorithms have been proposed by many researchers.
Tensor decomposition to compress convolutional layers in deep learning
Published in IISE Transactions, 2022
Yinan Wang, Weihong “Grace” Guo, Xiaowei Yue
Given the properties of our proposed CPAC-Conv layer, we can summarize its contributions to industrial applications into three aspects. First, edge computing is becoming increasingly used in manufacturing systems, in which the edge devices carry out a substantial amount of tasks but usually have limited computational resources. Compressing the model complexity will potentially reduce the computational requirements for Deep Neural Networks implementation in edge devices. Second, our proposed CPAC-Conv layer is an alternative to replace the original Conv layer. It is not limited to process image data. It can also be used to compress the Deep Neural Networks with Conv layers for processing videos and point clouds. Our proposed CPAC-Conv layer can be used as one alternative building block for designing novel deep learning methods. Lastly, the model with fewer parameters is more promising according to the Occam’s Razor principle. It can have better interpretability and help us to understand the relationship between the decomposed kernels and extracted features.
The Internet of Things for Logistics: Perspectives, Application Review, and Challenges
Published in IETE Technical Review, 2022
Hoa Tran-Dang, Nicolas Krommenacker, Patrick Charpentier, Dong-Seong Kim
To address these issues, fog and edge computing technologies are developed and widely deployed in the IoT systems. Fundamentally, these techniques imply a distributed data processing approach instead of centralized cloud computing. Accordingly, the edge computing enables the edge devices such as sensors, mobile phones to process the raw IoT data locally. Meanwhile, the fog computing extends the functionalities and services of cloud servers to the closer locations of data sources [110–112]. The network devices such as IoT GWs, routers, or switches can extend their functions to act as the fog devices [113–115] to run the cloud-like services such as virtual machines. Additionally, these devices can process the IoT data to create the intelligent IoT services [116–121]. For instances, as illustrated in Figure 8, a smart phone can serve as a fog node to create and display the 3D layout developed from the works [3,112].