Paradigms for Intelligent IOT Architecture
01 Jan 2020-pp 67-100
TL;DR: Agents are introduced in the IOT architecture which reacts intelligently using its learning capability and its characteristics improve the system performance.
Abstract: Recent researches in IOT bring out smartness with the basis of machine learning techniques. IOT architecture describes the gateways or fog, an analysis engine and an insight layer. These layers are embedded between the cloud and the edge devices. The insight layer employs various learning modules onto the data in the cloud. The fog layer is the most significant layer that improves the efficiency of IOT architecture. Cloud computing and Fog computing is mutually operated. Fog based application should address the issue of the data to be kept in the fog device and to identify the data present in the nearest fog device with the relevant search query. Learning helps to identify the data that are referred frequently and predict the data requirements of the near future. Agents are introduced in the IOT architecture which reacts intelligently using its learning capability and its characteristics improve the system performance.
Citations
More filters
06 May 2021
TL;DR: In this article, the authors proposed different machine learning based classification algorithms such as logistic regression, random forest, and Naive Bayes for handling the heavily imbalanced dataset and calculated the accuracy, precision, recall, f1 score, confusion matrix, and Roc-auc score.
Abstract: Credit card is the commonly used payment mode in the recent years. As the technology is developing, the number of fraud cases are also increasing and finally poses the need to develop a fraud detection algorithm to accurately find and eradicate the fraudulent activities. This research work proposes different machine learning based classification algorithms such as logistic regression, random forest, and Naive Bayes for handling the heavily imbalanced dataset. Finally, this research work will calculate the accuracy, precision, recall, f1 score, confusion matrix, and Roc-auc score.
14 citations
28 Jan 2021
TL;DR: In this paper, the authors used the convolution neural networks to detect and classify the class of cancer based on historical data of clinical images using CNN and achieved an accuracy of 80%.
Abstract: There is a necessary need for early detection of skin cancer and can prevent further spread in some cases of skin cancers, such as melanoma and focal cell carcinoma. Anyhow there are several factors that have bad impacts on the detection accuracy. In Recent times, the use of image processing and machine vision in the field of healthcare and medical applications is increasing at a greater phase. In this paper, we are using the Convolution neural networks to detect and classify the class of cancer based on historical data of clinical images using CNN.Some of our objectives through this research are ,to build a CNN model to detect skin cancer with an accuracy of >80% ,to keep the false negativity rate in the prediction to below 10%, to reach the precision of above 80% and do visualization on our Data. Simulation results show that the proposed method has superiority towards the other compared methods.
13 citations
TL;DR: In this article, the authors present a comprehensive survey of industrial IoT security and provide insight into today's industry countermeasure, current research proposals and ongoing challenges, and highlight the remaining open issues and challenges.
Abstract: The inherent complexities of Industrial Internet of Things (IIoT) architecture make its security and privacy issues becoming critically challenging. Numerous surveys have been published to review IoT security issues and challenges. The studies gave a general overview of IIoT security threats or a detailed analysis that explicitly focuses on specific technologies. However, recent studies fail to analyze the gap between security requirements of these technologies and their deployed countermeasure in the industry recently. Whether recent industry countermeasure is still adequate to address the security challenges of IIoT environment are questionable. This article presents a comprehensive survey of IIoT security and provides insight into today’s industry countermeasure, current research proposals and ongoing challenges. We classify IIoT technologies into the four-layer security architecture, examine the deployed countermeasure based on CIA+ security requirements, report the deficiencies of today’s countermeasure, and highlight the remaining open issues and challenges. As no single solution can fix the entire IIoT ecosystem, IIoT security architecture with a higher abstraction level using the bottom-up approach is needed. Moving towards a data-centric approach that assures data protection whenever and wherever it goes could potentially solve the challenges of industry deployment.
12 citations
28 Apr 2022
TL;DR: The aim of this paper is to address that face recognition with the best method Eigen Faces or Principal Component Analysis (PCA) algorithm basis on a training database as Olivetti dataset.
Abstract: Face recognition is one of the finest method or field to use for better accuracy and also security. Face is comprising of multidimensional structure with complexity and for that complexity we need much good worthy computing techniques for detection as well as for recognition. The aim of this paper is to address that face recognition with the best method Eigen Faces or Principal Component Analysis (PCA) algorithm basis. We also see face detection, recognition in this paper. We used a training database as Olivetti dataset.
9 citations
06 May 2021
TL;DR: In this paper, a convolutional neural network (CNN) algorithm was used to detect seven different emotions such as happy, sad, neutral, angry, surprise, fear and disgust.
Abstract: Facial emotion recognition is one of the most interesting research areas where many researchers are actively participating over the past few decades. This paper attempts to discuss about the application of emotion recognition where seven different emotions such as happy, sad, neutral, angry, surprise, fear and disgust are obtained. Humans can produce thousands of emotions in different situations which have different meanings, intensities and complexities. By using convolutional neural network (CNN) algorithm, an accuracy of about 89%has been achieved. It is the simplest way of all. For better results deep learning and neutral networks have been used. Our proposed deep learning model helps us in focusing important features in humans face to detect emotion using multiple datasets such as FER-2013 and image dataset.
9 citations
References
More filters
17 Aug 2012
TL;DR: This paper argues that the above characteristics make the Fog the appropriate platform for a number of critical Internet of Things services and applications, namely, Connected Vehicle, Smart Grid, Smart Cities, and, in general, Wireless Sensors and Actuators Networks (WSANs).
Abstract: Fog Computing extends the Cloud Computing paradigm to the edge of the network, thus enabling a new breed of applications and services. Defining characteristics of the Fog are: a) Low latency and location awareness; b) Wide-spread geographical distribution; c) Mobility; d) Very large number of nodes, e) Predominant role of wireless access, f) Strong presence of streaming and real time applications, g) Heterogeneity. In this paper we argue that the above characteristics make the Fog the appropriate platform for a number of critical Internet of Things (IoT) services and applications, namely, Connected Vehicle, Smart Grid, Smart Cities, and, in general, Wireless Sensors and Actuators Networks (WSANs).
4,440 citations
TL;DR: By sacrificing modest computation resources to save communication bandwidth and reduce transmission latency, fog computing can significantly improve the performance of cloud computing.
Abstract: Mobile users typically have high demand on localized and location-based information services. To always retrieve the localized data from the remote cloud, however, tends to be inefficient, which motivates fog computing. The fog computing, also known as edge computing, extends cloud computing by deploying localized computing facilities at the premise of users, which prestores cloud data and distributes to mobile users with fast-rate local connections. As such, fog computing introduces an intermediate fog layer between mobile users and cloud, and complements cloud computing toward low-latency high-rate services to mobile users. In this fundamental framework, it is important to study the interplay and cooperation between the edge (fog) and the core (cloud). In this paper, the tradeoff between power consumption and transmission delay in the fog-cloud computing system is investigated. We formulate a workload allocation problem which suggests the optimal workload allocations between fog and cloud toward the minimal power consumption with the constrained service delay. The problem is then tackled using an approximate approach by decomposing the primal problem into three subproblems of corresponding subsystems, which can be, respectively, solved. Finally, based on simulations and numerical results, we show that by sacrificing modest computation resources to save communication bandwidth and reduce transmission latency, fog computing can significantly improve the performance of cloud computing.
681 citations
TL;DR: A unified architectural model and a new taxonomy are presented, by comparing a large number of solutions to support the requirements of IoT applications that could not be met by today’s solutions.
184 citations
04 Jan 2018
TL;DR: A Fog-based IoT-Healthcare solution structure is introduced and the integration of Cloud-Fog services in interoperable Healthcare solutions extended upon the traditional Cloud-based structure is explored.
Abstract: The issue of utilizing Internet of Things (IoT) in Healthcare solutions relates to the problems of latency sensitivity, uneven data load, diverse user expectations and heterogeneity of the applications. Current explorations consider Cloud Computing as the base stone to create IoT-Enable solution. Nonetheless, this environment entails limitations in terms of multi-hop distance from the data source, geographical centralized architecture, economical aspects, etc. To address these limitations, there is a surge of solutions that apply Fog Computing as an approach to bring computing resources closer to the data sources. This approach is being fomented by the growing availability of powerful edge computing at lower cost and commercial developments in the area. Nonetheless, the implementation of Cloud-Fog interoperability and integration implies in complex coordination of applications and services and the demand for intelligent service orchestrations so that solutions can make the best use of distributed resources without compromising stability, quality of services, and security. In this paper, we introduce a Fog-based IoT-Healthcare solution structure and explore the integration of Cloud-Fog services in interoperable Healthcare solutions extended upon the traditional Cloud-based structure. The scenarios are evaluated through simulations using the iFogSim simulator and the results analyzed in relation to distributed computing, reduction of latency, optimization of data communication, and power consumption. The experimental results point towards improvement in instance cost, network delay and energy usage.
156 citations
TL;DR: A novel framework named EdgeGame is proposed to improve the cloud gaming experience by leveraging resources in the edge and offloads the computation-intensive rendering to the network edge instead, which can reduce network delay and bandwidth consumption greatly.
Abstract: With the development of 4G/5G technology and smart devices, more and more users begin to play games via their mobile devices. As a promising way to enable users to play any games, cloud gaming is proposed to stream game scene rendered remotely in the cloud with the format of video. However, it faces major challenges in terms of long delay and high network bandwidth. To this end, a novel framework named EdgeGame is proposed to improve the cloud gaming experience by leveraging resources in the edge. Compared to existing cloud gaming systems, EdgeGame offloads the computation-intensive rendering to the network edge instead, which can reduce network delay and bandwidth consumption greatly. Moreover, EdgeGame introduces deep reinforcement learning in the edge to adjust the video bitrates adaptively to accommodate the network dynamics. Finally, we implemented a prototype system and compared it with an existing cloud gaming system. The experiments show that EdgeGame can reduce the average network delay by 50 percent and improve user's QoE by 20 percent.
98 citations