Semi-supervised Transformer Network for Anomaly Detection in Cellular Internet of Things

Waleed Abd; Ibrahim

doi:https://doi.org/10.54216/IJWAC.040106

Journal of Intelligent Systems and Internet of Things | 2022

An Edge Intelligence Framework for Elegant Power Management in IoT-enabled Power Grids

The Internet of Things (IoT) is a concept that has the potential to attract new audiences in fields as diverse as manufacturing, healthcare, and more. IoT devices included into the sensor were the primary drivers of the massive data collection. To successfully combine, assess, and comprehend all programme objects, thus, self-adaptive algorithms based on AI are necessary. The proliferation of both massive datasets and resource-intensive IoT devices makes stringent power management essential. The proliferation of both massive datasets and resource-intensive Internet of Things devices makes stringent energy management essential. Combining IoT with AI-based techniques is crucial for equitable power distribution to compact mobile devices. To this end, we offer an efficient way to communicate between power utilities and end users by forecasting future power usage over short periods of time. Innovations include a revolutionary convolutional recurrent model for lightweight prediction method with low duration intricacy and minimum margins of error, as well as massive energy administration for edge devices via a centralised cloud-based data supervisory server. To maintain the power consumption and supply paradox efficiently, the suggested scheme has mobile nodes interact with a central remote server via an IoT network and then on to the corresponding power grid. We use a number of preparation methods to accommodate the varied electrical data, and then we construct a powerful decision-making engine for quick prediction on devices with limited resources.

groups

Irina V. Pustokhina mail -

Denis A. Pustokhin mail

link https://doi.org/10.54216/JISIoT.060204

Volume & Issue

Vol. Volume 6 / Iss. Issue 2

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2022

Federated Resistance Against Adversarial Attacks in Resource-constrained IoT

  Federated learning (FL) is a recently evolved distributed learning paradigm that gains increased research attention. To alleviate privacy concerns, FL fundamentally suggests that many entities can cooperatively train the machine/deep learning model by exchanging the learning parameters instead of raw data. Nevertheless, FL still exhibits inherent privacy problems caused by exposing the users’ data based on the training gradients. Besides, the unnoticeable adjustments on inputs done by adversarial attacks pose a critical security threat leading to damaging consequences on FL.  To tackle this problem, this study proposes an innovative Federated Deep Resistance (FDR) framework, to provide collaborative resistance against adversarial attacks from various sources in a Fog-assisted IIoT environment. The FDR is designed to enable fog nodes to cooperate to train the FDL model in a way that ensures that contributors have no access to the data of each other, where class probabilities are protected utilizing a private identifier generated for each class.  The FDR mainly emphasizes convolutional networks for image recognition from the Food-101 and CIFAR-100 datasets. The empirical results have revealed that FDR outperformed the state-of-the-art adversarial attacks resistance approaches with 5% of accuracy improvements.

groups

Mahmoud A. Zaher mail -

Heba H. Aly mail

link https://doi.org/10.54216/JISIoT.060205

Volume & Issue

Vol. Volume 6 / Iss. Issue 2

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2022

Automated System for Management of Exam Cell

These days, exam cell migration typically includes some manual computations and is primarily dependent on pen and paper.  The main objective of this extension is to bring it in a centralised manner.  By doing so, it will be possible to successfully supervise the actions taking place throughout an examination.  By entering their enrollment number, title, phone number, email address, semester, etc., the framework enables college or school students to register themselves with the system.  Typically accomplished by having students create their own unique points of interest for the exam cell to use as their login ID and password.

groups

Ajith R. mail -

Mercy Beullah mail

link https://doi.org/10.54216/JCHCI.040104

Volume & Issue

Vol. Volume 4 / Iss. Issue 1

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2022

Accident Detection System Using GPS and GSM by IOT

The primary goal of accident detection implementation is to reduce traffic accidents that result in the loss of priceless human life and other valuable items. Accident detection systems that use GPS and GSM save lives by shortening the time it takes for emergency personnel to reach the scene of an accident. We made the decision to recognise an automobile collision and notify the emergency personnel as well as the driver's main contacts. The product's main goal was to increase security for consumers and their families. GPS (Global Positioning System) and GSM are its foundations (Global system of Mobile communication).

groups

R. Manish mail -

M. Sumithra mail -

Lokhitha D. mail -

Mahalakshmi L. mail -

Durga V. mail -

Nirmala G. mail

link https://doi.org/10.54216/JCHCI.040105

Volume & Issue

Vol. Volume 4 / Iss. Issue 1

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2022

Smart Wheelchair-An Effective Transport for Handicapped and Aged Citizens

A wheelchair is a chair fitted with wheels. A survey says that around 132 million people use wheelchair around the world. But majority of them are dependent on others for their movements, especially people with some disorders. This dependent nature had hindered them from succeeding. To overcome this problem, they can use smart wheelchair, which is auto movable based on head tilt movements. It collects information from the patient through in built sensors and enhances the seating position. It is also designed with obstacle and fall detection system which reduces the chance of collision during their journey. This makes a physically challenged or dependent person as physically independent person. This wheelchair can also be used by aged people who lack motor skills. In this paper, we can review the art of smart wheelchair and the features of it.

groups

R. Venkatesan mail -

Gokul Santhosh Y. mail -

Sathya Preiya V. mail -

V. D. Ashok Kumar mail

link https://doi.org/10.54216/JCHCI.040201

Volume & Issue

Vol. Volume 4 / Iss. Issue 2

Details open_in_new

International Journal of Neutrosophic Science | 2023

A Review on Metaheuristic Algorithms with Neutrosophic Sets for Image Enhancement

Breast cancer has emerged as a major killer in recent years. With a yearly rate of about one million new cases, it is the most prevalent among women in the world's poorest countries. Grading of cellular images has emerged as a key prognostic factor during the past decade. Neutrosophic sets used to enhance medical images in the last decade. Neutrosophic sets can overcome the uncertainty and indeterminacy of information. In recent years, metaheuristics have integrated with neutrosophic sets. Because of their adaptability, simplicity, and task independence, metaheuristics have been extensively employed to tackle many difficult non-linear optimization problems. The purpose of this research is to investigate several approaches to image classification for breast cancer pictures. This includes the use of metaheuristics and neutrosophic sets for optimization and image enhancement. This research was undertaken to better understand the current state of the art in breast cancer identification from medical pictures and to provide insight into the difficulties that lie ahead. We hope that this will encourage academics to investigate hitherto understudied facets of breast cancer identification.

groups

M. A. El-Shorbagy mail -

Hossam A. Nabwey mail -

Mustafa Inc mail -

Mostafa M. A. Khater mail

link https://doi.org/10.54216/IJNS.200113

Volume & Issue

Vol. Volume 20 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2023

Metaheuristics and Neutrosophic Sets for COVID-19 Detection: A review study

The fast spread of COVID-19 has been a problem for several nations since February 2020. Computer-aided diagnostic technologies that are both effective and affordable are urgently needed to help ease the burden on healthcare systems. Researchers are delving further into the feasibility of using image analysis to detect COVID-19 in X-ray and CT-scan pictures of patients. In the past ten years, deep learning has surpassed every other method for classifying images. However, deep learning-based approaches' effectiveness is very sensitive to the design of the underlying deep neural network. In recent years, metaheuristics and neutrosophic sets have become more popular as a means of fine-tuning the structure of deep networks. Because of their adaptability, simplicity, and task dependence, metaheuristics have been extensively employed to tackle many difficult non-linear optimization problems. To correctly identify COVID-19 patients from their chest X-rays, the authors of this research made a review of a neurotrophic model and metaheuristics methods.

groups

M. A. El-Shorbagy mail -

Hossam A. Nabwey mail -

Mustafa Inc mail -

Mostafa M. A. Khater mail

link https://doi.org/10.54216/IJNS.200114

Volume & Issue

Vol. Volume 20 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2023

Impact analysis of Macroeconomic Variables on Stock Market using Neutrosophic Interval Valued Dependent Matrix Model

Macroeconomic factors in general are crucial for developing a country's economy. This analysis takes into account the chosen macroeconomic factors for the years 2015 to 2019 including inflation, interest rate, GDP, and GDP per capita. The present study considered the new method of Neutrosophic environment in terms of the Fuzzy CETD matrix to determine the impact of the stock market for a particular year. This article describes a technique for examining how macroeconomic factors affect the stock market in a specific year. The proposed method reveals that the impact of the stock market is higher in 2015 than in 2016.

groups

G. Kavitha mail -

S. Bhuvaneswari mail

link https://doi.org/10.54216/IJNS.200115

Volume & Issue

Vol. Volume 20 / Iss. Issue 1

Details open_in_new

Prospects for Applied Mathematics and Data Analysis | 2023

Aspects of Language Monadic Predicate Logic System plus Identity (LMPL+I)

In this paper, we devoted to study the language monadic predicate logic system plus identity (LMPLS+I) as extension of the language of propositional logic system (LPLS). I.e., ( (LMPLS+I), which it contains all the hereditary traits (or features) of   , furthermore that, we will add some new data information between relationship of object, subject and predicate. This is the task of monadic predicate logic system addressed.  As mentioned in pervious papers, the main task of system of logic is classifying between valid and invalid arguments, moreover, the central role the system of logic how distinguishes between the conclusions which follow from their premises of the arguments and   those do not follow from their premises. As a matter of fact, when we encounter some proofs that seem perceptually (or intuitively) sound, but we are -unable to prove their validity due to the inability language of propositional logic system (LPLS). Hence, it was necessary to uses the monadic predicate logic system (LMPLS+I) to overcome this problem. In this article, we study syntax, semantics and inference on language monadic predicate logic system plus identity (LMPLS+I) and investigation characteristics of arguments such valid \ invalid and types of formulas and relations between formulas like consistency and inconsistency sets.

groups

Adel Mohammed Al-Odhari mail

link https://doi.org/10.54216/PAMDA.010103

Volume & Issue

Vol. Volume 1 / Iss. Issue 1

Details open_in_new

International Journal of Wireless and Ad Hoc Communication | 2022

Semi-supervised Transformer Network for Anomaly Detection in Cellular Internet of Things

Because of the lightning-fast expansion of the Internet of Things (IoT) technologies, an enormous amount of data has been produced. This traffic can be mined for information that can be used to identify and avoid intrusions into IoT networks. Despite the significant efforts that have been put into labeling Internet of Things traffic records, the total number of labeled records is still quite low, which makes it more difficult to detect intrusions. This study introduces a semi-supervised deep learning approach for intrusion detection (S2T-Net), in which we propose a temporal transformer module to empower the model to learn valuable interactions in cellular data. An improved spatial transformer is presented to capture local representation in the cellular traffic flow. At the same time, a multilevel semi-supervised training technique is used to account for the consecutive structure of the IoT traffic information. In order to provide effective real-time threat intelligence, the suggested S2T-Net can be tightly coupled into a cellular IoT network. Last but not least, empirical assessments on two current databases (CIC-IDS2017 and CIC-IDS2018) show that S2T-Net boosts intrusion detection accuracy and resilience while retaining resource-efficient computing.

groups

Waleed Abd Elkhalik mail -

Ibrahim Elhenawy mail

link https://doi.org/10.54216/IJWAC.040106

Volume & Issue

Vol. Volume 4 / Iss. Issue 1

Details open_in_new