Theses supervised by Prof. Dr. Osman Nuri Uçan
79 theses · Altınbaş University
Advanced 3D face anti-spoofing system using hybrid deep neural network and optimization techniques
ace recognition technology is used everywhere, including mobile security, surveillance, and online payments, to authenticate a person's identity. Spoof faces, like printed photos, videos, 3D masks, and deep fake-generated faces, can mislead such systems. The goal of this research is to improve Face Anti-Spoofing (FAS) techniques to improve facial recognition security, accuracy, and efficiency. The study suggests a 3D Face Anti-Spoofing Model incorporating Dense Squeeze and Excitation Networks and a Neighbourhood-Aware Kernel Adaptation (NAKA) process for identification of fine details and textures on a person's face. A Lightweight Multi-Modal Deep Fusion Network is suggested for fusing different face data modalities such as RGB images, depth maps, and texture information. It is useful in face spoofing detection with higher accuracy even against advanced attacks. Deep Reinforcement Learning (DRL) is also used by the system to learn automatically and keep itself updated at all times, thus able to identify new ways of spoofing. The models are tested using standard datasets like CASIA-SURF, CelebA-Spoof, and GREAT-FASD-S based on a number of performance metrics such as accuracy, precision, recall, and error rates. Experiments demonstrate that models as proposed compare favorably against existing methods and are computationally lightweight for use in real-world applications. The study also confronts fundamental problems such as generalizable good performance over datasets, adversarial attacks resistance, and security vs. usability. Grounded on multi-modal fusion, vi attention, and reinforcement learning, the study gives contributions towards robust, light-weight, and explainable face anti-spoofing approaches. These can be applied in high-security applications like banking, border control, and mobile authentication. The outcome will help to introduce biometric security and protect against future face spoofing attacks of novel forms.
Backscatter-assisted non-orthogonal multiple access for ultra-massive machine-type communications in 6G networks
6G cellular networks will require to provide significantly higher system capacity and user data rates. This potential growth along with today's shortage of spectrum increases the need for new frequency spectrum. The new millimeter wave spectrum is emerging as a suitable candidate with a large amount of available bandwidth (around 60 GHz). The new spectrum places a new requirement for single element antenna and array design. This work addresses the issues of mm-wave antenna design and the problems that designer may face during the designing process. The proposed antenna has a resonating frequency of 26 GHz and 2 GHz bandwidth. Beam squint problem is also analysis in this work. The results showed that the gain of the mm-wave antenna array becomes a function of frequency which significantly reduces the performance of mm-wave communication system. Millimeter wave (mmWave) wireless technology has become a part of human life for high-speed and secure data transmission. A square microstrip patch antenna with a resonance frequency of 26GHz was proposed in this work for mmWave wireless communication. One square radiating element makes up the antenna. CST Microwave Studio, an electromagnetic simulation program, was used to construct and study the suggested antenna on a Rogers RO 3003 lossy substrate with a relative permittivity of 3. This work's outcome demonstrates a minimal return loss of -19.34 dB, a gain of 6.97 dBi, and a bandwidth of 2GHz at a resonant frequency of 26 GHz. The element is transformed into an 8-element uniform linear array; the suggested array boosts the gain to 16dBi while maintaining a high radiation efficiency. A high-gain wide-band planar antenna with a reconfigurable intelligent surface (RIS) is presented in this study for use in contemporary wireless communication applications. The antenna is composed of two primary components: a basic antenna component with two light-dependent resistor switches and cross-line slots, and a second component that uses the RIS layer for beam steering. The RIS is made up of two-sided, five-by-five-unit cells that form a square. The antenna substrate is a 1.6 mm thick dielectric layer of FR4 epoxy glass. In order to attain the appropriate electromagnetic properties at the frequency band of interest, the RIS inclusions are developed and numerically tested. The manufactured prototype achieves an antenna gain ranging from 10.5dBi to 16.8dBi and exhibits a wide band covering frequencies from 0.9GHz to 3.5GHz with S11 below -10dB. Effective aperture utilization is demonstrated by experimental measurements in all shapes, and beam steering from +22° to -22° is achieved without lowering side-lobe levels. To assess channel performance in terms of bit error rate (BER) and channel capacity (CC), the suggested antenna's performance is compared to actual measurements. In contrast to traditional RIS-assisted antennae that depend on PIN or varactor switches, the presented work conducted LDR-controlled design offers compact beam steering with little insertion loss.
Web performansının dinamik içerik optimizasyonu için çok silahlı hayvanlarla güçlendirilmiş öğrenim
The field of web performance optimization faces the challenge of dynamically optimizing web content to enhance user experience and maximize engagement. Traditional approaches to content optimization, such as static A/B testing or rule-based algorithms, often fall short in adapting to the dynamic nature of web content and fail to provide personalized experiences that align with user preferences. As a result, organizations struggle to achieve optimal web performance, leading to lower user satisfaction, decreased retention rates, and missed conversion opportunities. Additionally, the rapid growth of internet usage and the increasing demand for fast and efficient web experiences further intensify the need for effective content optimization strategies. Websites must continuously evaluate and adjust their content configuration to deliver engaging experiences that meet user expectations. However, manually identifying the most rewarding content variants in real-time becomes a daunting task as the number of potential combinations increases. Hence, there is a pressing need for an intelligent and data-driven approach that can dynamically optimize web content to maximize user engagement and improve web performance. this research aims to explore the application of reinforcement learning techniques with multi-armed bandits for dynamic content optimization of web performance.
Ekonomik analiz ile petrol rafinerisi iş akışı dönüşümü için modern web tabanlı karar destek sistemi
The refinery industry operates within a dynamic and competitive environment, requiring decision-makers to navigate complex challenges and uncertainties. This project focused on the development of a sophisticated Decision Support System (DSS) using advanced technologies such as NET 6 and C#. The primary objective was to empower decision-makers in the refinery industry with a robust tool to analyze extensive datasets, gain valuable insights, and make informed strategic choices. The project followed a systematic development process, involving the transformation of traditional Excel-based models into an automated application. By harnessing the capabilities of NET 6 and C#, the DSS facilitated seamless integration with backend systems, efficient data processing, and complex calculations. The frontend of the DSS was built using ReactJS, resulting in a user-friendly and visually appealing interface. Decision-makers can effortlessly navigate through the system, visualize data, and analyze key metrics, enabling them to make data-driven decisions with confidence. The implementation of the DSS offers significant advantages to the refinery industry. It reduces dependency on manual processes, automates data analysis, and provides real-time access to critical information. Decision-makers can explore various scenarios, evaluate financial implications, and optimize resource allocation. By streamlining the decision-making process, the DSS saves time, enhances operational efficiency, and empowers decision-makers to stay ahead in the competitive refinery industry. This project vii underscores the transformative potential of advanced technologies in leveraging data for informed decision-making, fostering a competitive edge in the refinery industry landscape. Keywords: Oil, Decision Support System, Web-Based Decision, Economic Analysis, Decision Support System (DSS).
Değiştirilmiş KNN algoritmasını ve özellik ağırlıklandırmayı kullanarak DNS trafiği tabanlı izinsiz giriş tespiti
All computer networks can be the target of attacks, from the smallest, such as home networks, to the largest and most complex, such as corporate networks. To make the situation even worse, programs can easily be obtained through the Internet (or by other means), and then used for malicious actions, making any inexperienced person have the same skills as a professional It is necessary to propose a new database that contains a portion of the traffic that is present on an existing actual network environment in order to carry out testing in a real environment. This is necessary in order to accomplish the goal of conducting testing in a real environment The main objective is to create a network user control mechanism in order to recognize the user's legitimacy through their actions.
Makine öğrenmesi yöntemleri ile çevrimiçi kredi kartı işlemlerinde Fraud analizi tahmini
Kredi kartları, dünya genelinde yaygın kullanımı ve sağlam altyapısı sayesinde hızla insanların günlük hayatlarına entegre olmuş ve güvenle kullanılan ödeme araçlarından biri haline gelmiştir. Ancak, kredi kartı sayılarının artması ve işlem hacminin hızla büyümesi, dolandırıcıları cezbetmiş ve haksız kazanç elde etme amacıyla çeşitli dolandırıcılık yöntemlerini ortaya çıkarmıştır. Günümüzde kredi kartı bilgilerine ulaşmanın kolaylaşması, kredi kartı dolandırıcılarının faaliyetlerini kolaylaştırmaktadır. Gelişen teknoloji ile hesap hareketleri zaman içinde analiz edilebilmekte ve kötü niyetli verilerin kullanımı izlenebilmektedir. Bu çalışmada, Kaggle veritabanından elde edilen Kredi Kartı Dolandırıcılık Teşhis veri seti kullanılarak bu çalışma, kredi kartı dolandırıcılığı tespiti için topluluk tabanlı XGBoost modeli ile diğer geleneksel makine öğrenmesi modellerini karşılaştırarak önemli bulgular ortaya koymaktadır. XGBoost, Random Forest ve CatBoost modellerinin, kesinlik, geri çağırma gibi performans ölçütleri üzerinde daha iyi performans sergilediği belirtilmektedir. Bu sonuçlar, finansal kurumların günlük operasyonlarında karşılaştıkları dolandırıcılık risklerini azaltma potansiyeline işaret etmektedir. Çalışmanın dikkate değer bir noktası, XGBoost, Random Forest ve CatBoost'un dengesiz veri setleriyle başa çıkma yeteneğinin vurgulanmasıdır. Geleneksel modellerin zayıf performans gösterdiği bu durumlarda, XGBoost, RandomForest ve CatBoost'un %99'a kadar tahmin doğruluğu sağladığı ve diğer modellere göre daha iyi bir performans sunduğu gözlemlenmektedir. Anahtar Kelimeler: Çok Katmanlı Sinir Ağları, Veri Madenciliği, Naive Bayes Yönetmi, Sahtekarlık tespiti.
Gerçek zamanlı hizmetler sunmak için sis-bulut ortamında ıot kaynak kullanımının artırılması
When resources and computers are made available on demand over the internet, this is called fog-cloud computing. This makes it easy to offer people a variety of integrated computing services without being limited by local resources. This means giving people more than just places to store and back up their data and ways to sync their own files. but it also has processing power and a simple software interface that lets the user control when it's connected to the network. This makes things easier by ignoring many details and internal processes. When it comes to the cloud, scheduling algorithms are necessary to provide services that meet goals like high performance, low prices, minimal energy use, and so on. It is an NP-hard problem to come up with scheduling methods that meet more than one of these goals. We introduce some new heuristic scheduling algorithms in this thesis that allow for multi-objective optimization. We then compare their effectiveness with some well-known scheduling algorithms to study how well they work. The first algorithm, the BDA, looks at the Make-span and the period it takes to complete jobs by the due date. Our algorithm takes into account a task's due date by giving the most weight for the job with the earlier due date and send it to the resource that can complete it in the shortest amount of time to achieve the shortest Makespan. Fog Max-Cloud Min is the name of the first part of our method. Ant Colony Optimization is the name of the second part. We looked at how well our suggested algorithm methods met deadlines and how long the system took to make. Compared to the tools we have now. The study results show that our algorithm is better at getting things done with the lowest Makespan and the best deadline satisfaction. In the second algorithm, We develop improvements to the performance and cost (PC) algorithm in order to give more weight Considering the significant expenses involved, reduce energy consumption, and reduce the duration of create a product. In this work, we describe an approach that is a combination of the PCA and GWO techniques. This algorithm is called the Performance and Cost-Gray Wolf Optimization (PC-GWO) algorithm. The results of the test indicate that the PC-GWO The algorithm decreases the average total energy usage by 12.17 percent, 11.5 percent, and 7.19 percent, as well as the Makespan by 16.72 percent, 16.38 percent, and 14.10 percent. When compared to the GWO algorithm, the PCA method, and the PSO algorithm, it also improves the best average resource consumption by 13.2 percent, 12.05%, and 10.9 percent respectively.
Deep kullanılarak el işaret dilinin sınıflandırılması öğrenme
People with hearing disabilities face many problems, which impede many of their social life issues in all areas of communication, so effective communication is crucial in the development of a nation. It promotes understanding and inclusivity among all members of the community, including those who are deaf. Good communication is key to building and maintaining a strong, cohesive society. I used 29,000 sign language images, each class contained 1,000 images. I built a model from scratch (ASL.model) and compared it with pre-existing models (Xception, Inception, ResNet, VGG16 and MobileNet). The study of intelligent computers that can carry out activities without direct human guidance is known as artificial intelligence (AI), a fast-growing topic within computer science. These tasks may include learning, decision making, and problem solving, and they are often accomplished through the use of algorithms, data, and machine learning techniques. To find the most appropriate classification features to be used for classification, deep learning techniques will be employed in this thesis to create a model for classifying sign language utilizing photographs obtained from the Kaggle depository as a training data set. Deep learning is now widely employed across a variety of industries due to its accuracy and efficiency, particularly for vast yet complicated data, such as photos, sounds, or text, where deep learning algorithms are taught using massive, labeled data sets. In this thesis, we used deep learning to classify a total of twenty-nine sign language-related classes. We put forth a fresh framework for categorizing sign language. The suggested model was put into practice, trained, verified, and tested. The model passed the test with a 99.97% success rate. To assess the effectiveness of our suggested model (ASL.model) with that of these methods, we also employed five pre-trained models (Xception, Inception, ResNet, VGG16, and MobileNet) of assisting the performance of deep learning algorithms that use Convolutional Neural Networks (CNNs). The five pre-trained models had F1-score levels of 100%, 99.51%, 99.87%, 100%, and 99.68%, respectively. The model Xception and VGG16 outperformed all others in terms of testing accuracy but when the testing time was smaller than 1.45 seconds, the suggested model outperformed all others in terms of time.
Derin öğrenme tekniklerinin hibrit yöntemini kullanarak farklı saldırıları tespit etmek için SDN ortamına yönelik siber güvenlik sistemi
As contemporary networks come under attack with sophisticated and mass-scale cyber-threats, traditional intrusion detection systems (IDS) are unable to cope with changing threats. Such limitations are the focus of this research, and two new deep learning-based frameworks are introduced, which are best suited for fast and effective intrusion detection in Software-Defined Network (SDN) environments: (1) the Adversarial Learning-based Multi-Branched Hybrid Architecture and (2) the Multi-Branched Hybrid Perceptron Network (MBHPN). Together, these models offer an intelligent and scalable IDS platform that can detect advanced threats, in particular Distributed Denial of Service (DDoS) attacks, through adaptive learning, real-time integration, and context analysis. The first architecture, as explained in Chapter 3, combines Convolutional Neural Networks (CNN), Capsule Networks, and Long Short-Term Memory (LSTM) layers in a multi-branch architecture. CNNs encode spatial dependencies, Capsule Networks preserve hierarchical relations among features, and LSTMs describe sequential behaviors of traffic flows. The framework also incorporates Dynamic Adversarial Learning, where adversarial samples are produced to mimic attacks, thus enhancing model robustness. Each branch produces a unique feature representation, which is combined using attention-weighted methods to create an integrated, discriminative feature space tailored for intrusion detection. In Chapter 4, this architecture is enriched by the addition of MBHPN, which is directly optimized for DDoS attack detection. This network combines three deep learning branches: MLP, DenseNet-like, and ResNet-like units. It adds Dynamic Feature Adaptation (DFA) to down-regulate noisy features and up-regulate pertinent traffic signatures. Multi-instance Learning (MIL) to process aggregated traffic flows instead of standalone instances, and thus greatly enhances contextual perception. These improvements make MBHPN robust against class imbalance, evasion attacks, and changing attack patterns prevalent in actual networks. The models are trained and tested on three intrusion detection benchmark datasets: UNSW-NB15, CICIDS2017, and CSE-CIC-IDS2018, which contain varied traffic patterns and attack conditions. Proven through extensive experimentation, the models outperform the current state of affairs. On the UNSW-NB15 dataset, MBHPN reports 99.31% accuracy, 98.02% precision, 98.87% recall, and an F1-score of 98.44% with a false positive ratio (FPR) decreased to 0.61% from the original 1.12% FPR of the base model. The time taken for inference reduced from 7.8 ms to 5.3 ms, which reflects a 32% boost in speed from SDNetc integration.
Coğrafi bilgi sistemleri (CBS) ve yapay zeka (Aİ) algoritmalarina dayali sürdürülebilir güç kaynaği yönetimi
Türkiye has been experiencing rapid economic development in recent decades, which has significantly increased its energy needs, and the need for sustainable energy sources such as wind energy has emerged. This thesis aims to provide an integrated framework for selecting optimal locations for wind farms in Türkiye by integrating geographic information systems (GIS) techniques with artificial intelligence (AI) algorithms, specifically supervised and unsupervised machine learning. The study involved the spatial analysis of natural data (wind speed, slope, and elevation), socio-economic factors (proximity to roads and urban areas), and environmental factors (protected areas and water bodies). The results of the unsupervised classification algorithms (K-Means and K-Medoids) demonstrated the ability to identify clear clusters of sites with varying degrees of relevance. In contrast, the supervised algorithms demonstrated high classification accuracy, with SVM outperforming with 95.1% accuracy, followed by K-NN and Random Forest, while Naïve Bayes was the least accurate owing to its assumption of feature independence. The thesis adopted Ensemble Learning to enhance the accuracy and diminish the variance between the algorithms' results, culminating in a final identification of the optimal locations where the findings of the four algorithms converge, which were depicted cartographically using GIS. The study affirms that integrating AI techniques with spatial analysis provides a potent tool for decision-makers, expediting the transition towards renewable energy and mitigating environmental impacts. The thesis recommends further development of the models through the use of real-time data and improved multi-criteria decision-making (MCDA) methods and suggests extending the application to other sustainable energy sources.
A new method based CNNcombined with genetic algorithm and support vector machine for COVID-19 detection by analyzing X-ray images
COVID-19 is an infectious disease caused by the newly discovered coronavirus. This new type of virus and disease was unknown before the outbreak that emerged in Wuhan, China in December 2019. COVID-19 poses a serious threat to public health. Older adults and people with pre-existing medical conditions such as diabetes, high blood pressure, heart disease, chronic lung disease and obesity are at increased risk for serious illness and complications. Last year, a computer scientist used various machine learning and deep learning techniques to detect Covid-19. In this study, efficient Covid-19 detection framework presented to detect Covid-19 by analyzing x-ray tests. The proposed framework based CNN combined with genetic algorithm and SVM classifier. The main contribution in this study is combining CNN with genetic algorithm and SVM to detect Covid-19 with accurate estimation and minimum execution time. Several scenarios are executed to validated the presented method . Finally, the obtained results compared with several studies presented to solve this problem
Drones swarm free space optical communication with pointing error
Drones are one of the Unmanned Arial Vehicles (UAV) used for short-length and low-altitude applications. Applying Free Space Optical (FSO) technology is modern in drone communication systems. The FSO technology has significant features such as high security due to narrow beamwidth, insusceptible to interferences, free license, landline connection is not appropriate, etc. The proposed system took the V-shape configuration and simulated it using MATLAB 2020. The V-shape configuration is one topology of networking topologies used in communication systems. The system has three subsystems: two Single-Input Single-Output (SISO) topologies and one Multiple-Input single-output (MISO) topology, connected by an optical beam and flying in weak atmospheric turbulence. The optical beam modulates using the PPM technique and faces many obstacles that affect the system's performance, such as the attenuation and turbulence of the atmospheric channel and misalignment. The obstacles represented by the channel gain H consist of three factors: the pointing error factor HP, the atmospheric attenuation Ha, and the atmospheric turbulence Hf. Hp is a significant factor that may be happening in the transmitter, receiver, or optical beam. The related parameters to Hp are the pointing error angle θr, the link distance Z, and the beamwidth wz. The pointing error angle θr must be < 〖10〗^(-4) rad for long distances (inter-satellite application). The goal of this dissertation is to measure the allowable range of the pointing error angles θr that is >〖10〗^(-4) rad at specific link distance Z where the system gets high performance and determine the link distance Z where the system failed at a specific pointing error angle θr. The results showed that high system performance happens at SNR ≥10 dB. For pointing error angle θr=〖10〗^(-2)rad, the values of the applicable link distances Z are nearly 10% of the link distances Z values for θr=〖10〗^(-3) rad. The same percent when increasing the pointing error angle θr to 〖10〗^(-1) rad. Also, the results showed that the best performance was at Z≤ 8500 m. Therefore, the system's highest altitude h system can reach ≤ 8500 m, depending on the drone's characteristics. Each subsystem can be considered an independent system and used in another configuration.
Employing machine learning techniques and fuzzy membership for detecting fraud transactions in credit card
Credit card fraud is becoming an increasingly prevalent problem in today's financial sector. An alarming rise in the number of fraud-related activities has been observed in the past few years, which has resulted in significant financial losses for numerous organizations, businesses, and government bodies. Because the numbers are projected to grow in the future, many researchers in this discipline have centered their efforts around identifying fraudulent behavior early on using advanced machine learning algorithms, which are becoming increasingly popular. However, the identification of credit card fraud is not easy for many reasons, including the fact that the fraudulent behaviors vary from one attempt to the next and the available datasets in this field are very few. In this thesis, a system is proposed to detect fraudulent transactions based on some methods of machine learning such like Logistic regression (LR), NaiveBayes (NB), as well as the Linear Discriminant Analysis(LDA), in addition to XGBoost algorithm. the techniques used to generate the models of the suggested system, they were trained and evaluated on the basis of two different datasets. Where the first dataset was the European Cardholders, and the second dataset was the Turkish dataset provided by the Yapi Kredi company. These datasets suffer from a big imbalance problem. Therefore, we used SMOTE to fix this issue. The system models' effectiveness was.measured employing a variety of metrics, specifically the confusion matrix, accuracy, F1score, recall, precision as well as AUC. Also, the fuzzy membership function was adopted to the dataset in order to raise the system's efficiency. The final results showed the high efficiency of XGBoost.
Designing a smart system to detect the intrusion in IoT
The Internet of Things (IoT) has been quickly growing during the previous few years, with the intention of having a wider impact on every aspect of life, from daily activities to vast industrial systems. Unluckily, A group of cybercriminals took notice of this, those responsible for turning the IoT into a vector for cybercrime, potentially exposing end nodes to attack. Since there are so many different kinds of IoT devices, There are difficulties to defend infrastructure for the IoT using a normal intrusion detection system (IDS). IoT devices need to be protected. We looked at data flow in the IoT. in this work taking two datasets (UNSW-NB15 and DoH20) and using three Machine Learning (ML) classifiers: Random Forest (RF), K-Nearest Neighbors (KNN) and Decision Tree (DT). For each method, we determined the Error Rate (ER), Accuracy(Acc), Precision, Recall, and F1 score. We received outstanding results (100%) when we combined these two classifiers. Detection rates are extremely high. A succinct summary of the results is presented.
Intelligent security system for mobile adhoc networks based on machine learning
Mobile Ad-hoc Networks (MANET) provide a service through network applications in one way or another. In many diverse fields, such as tactical networks used in military communications and environmental applications, the network is used as business networks such as Personal Area Network (PAN). They are used in educational environments such as schools and linked to Internet networks to expand coverage. Because of the multiple characteristics of MANET, including high- level mobility, node decentralization, physical insecurity, etc. MANET networks are susceptible to a variety of possible threats active or passive. MANET units (devices) are surveilled using detection technology to differentiate between ordinary and harmful actions. Therefore, Classification was accomplished utilize artificial intelligence withunsupervised Machine Learning(ML) approaches, which comprise the Random Forest (RF), Support Vector Machine (SVM), and Naïve-Bayes (NB) algorithms. Depending on a training samples in the feature space. The suggested Intrusion Detection System (IDS) is composed of the following basic stages: the traffic generation stage, the network simulation stage, the data collection. Five steps just for pre-processing a data set, the training stage, and the testing stage. We generated the dataset using Network Simulator-2 (NS2). The proposed system tested with dataset extracted from the trace file of network simulator. performance metrics are calculated: Confusing Matrix, Accuracy .rate, Error rate, Precision, Recall, F1. Besides determining each algorithm's training and testing duration. The findings revealed that the suggested system yielded promising outcomes, the accuracy rate of the R.F. algorithms reached 100%, SVM =99.18%, and NB =94.33%. Using trial and error feature selection improves the system and reduces training and testing time complexity. High levels of detection capacities and performance metrics have been evaluated by comparing and analyzing the results.
Securing critical information: An image cryptography digital based on multi level cryptographic
The importance of image encryption has considerably increased specially after the spectacular growth of internet of things (IoT) and due to the simplicity of capturing and transferring digital images. Although there are several encryption approaches, chaotic with image cryptography is considered the most appropriate approach for image applications due to its sensitivity to starting conditions and control parameter value. This research aims at generating an encrypted image free of statistical information to make cryptanalysis infeasible. Therefore, a new method was introduced in this thesis called Multi-layer Chaotic Maps (MLCM) based on confusion and diffusion. Basically, the confusion method uses the Sensitive Logistic Map (SLM), Hénon Map, and the additive white Gaussian noise to generate random numbers to be used in the pixel permutation method. However, the diffusion method uses Extended Bernoulli Map (EBM), Tinkerbell, Burgers, and Ricker maps to generate the random matrix. The correlation between adjacent pixels was minimized to have a very small value (x10-3). Besides, the keyspace was extended to be very large (2^450) considering the key sensitivity to hinder brute force attack. Finally, a histogram was idealized to be perfectly equal in all occurrences and the resulted information entropy was equal to the ideal value (8), which means that the resulted encrypted image is free of statistical properties in terms of the value of the histogram and the value of information entropy. Based on the findings, the high randomness of the generated random sequences of the proposed confusion and diffusion methods is capable of producing a robust image encryption framework against all types of cryptanalysis attacks.
Pedometer proposal for foot motion detection by arm motion detection
We developed and constructed a pedometer using low-cost components for this research. The proposed pedometer utilized a three-axis MEMS digital accelerometer type (ADXL345) in order to records moving activity, a 8-bit low- power microchip microcontroller pic18F4550 microchip microcontroller to that controls the overall process of pedometer and embedded the step algorithm. This algorithm can estimate the foot motion by detecting arm motion. The movement of the arm is detected by the accelerometer, it can evaluate the position and the dynamic change of the movement. Then, the algorithm calculates the steps taken by the operator and follows it by showing the total distance traveled by the person. The pedometer has also a Bluetooth module type (HC-05) to allow tracks the user activity with a smartphone or tablet communication pedometer. The system has been tested and compared with real reading and with Smartphone pedometer techniques available with Samsung smartphone. The results appeared that the proposed pedometer has less error in case of walking than jogging, for distance calculation, the error is less than 3% for walking and it less than 7% for jogging. For steps calculation, the error is (2%) for walking and (6%) for jogging. The test results also show that the proposed pedometer has
New smart grid framework based optimized ai method: Case studyattack detection
A smart grid is an electrical network based on digital technology that is used to deliver electricity to consumers through two-way digital communications. The system provides supply chain monitoring, analysis, control and reporting to increase efficiency, reduce energy consumption and costs, and ensure maximum supply chain transparency and energy reliability. The smart grid was introduced to fill the gaps in the existing power grids with smart electricity meters. In this paper, we propose the use of an automatic TSA-based deep encoder to solve the IDS problem in the Internet of Things. It is the first automatic encoder to be used to reduce the size of input data. In the second step, the output of the first autoencoder is connected to the second encoder and the size of the input data is also reduced. Then the output of the second autoencoder is connected to SoftMax. SoftMax categorizes the attributes into three attacks (normal and abnormal). The proposed system showed results with an accuracy of 98% compared to some studies.
Using artificial intelligence methods to improve the prediction rate and accuracy
The Extreme Learning Machine (ELM) is a single-hidden-layer feed forward neural network that starts the weights for the connections between the input and hidden layer, as well as the hidden layer bias, at random. There are several hidden layers and two secret layers. When applied to the same data set, ELMs have incorporated several improvements, such as preserving the randomly generated weights and biases and increasing the number of hidden layers. Within the ELM architecture, a greedy algorithm is also developed to penetrate sparse methods and provides an improvement over conventional. This research examines the testing accuracy of four ELM algorithms: three regular ELM algorithms and a greedy single hidden layer ELM, all of which are used to discover classification faults on three datasets. According to the findings of the studies, the greedy Extreme Learning Machine algorithm has a high prediction rate and a decent prediction accuracy.
Detection and classification of brain tumors in MRI images using deep convolutional neural network
Brain tumors is a chronic and inflammatory disease of the central nervous system (CNS), which causes white matter lesions in the brain and spinal cord, as a result of demyelination of axons The brain tumors. can be classified into four types: Remissive-recurrent (RRMS), Secondary Progressive (EMPS), Primary Progressive (PPME) and Recurrent Progressive (PRMS) Recent work in Machine Learning (Machine Learning) and Deep Learning (Deep Learning) has shown fruitful outcomes, as observed in clinical applications and medical image processing Manual segmentation of lesions on magnetic resonance images has become the standard despite being labour-intensive and subject to interobserver variability (e.g., interobserver variability) In this thesis, we aimed to test out these two novel CNN implementations in the hopes of noticing performance gains. Our approach differs from those of others in that we use a multi-layer CNN rather than the standard three-layer architecture (convolution + subsampling + classification). In order to automatically segment sclerotic lesions in a magnetic resonance image, it is necessary to: investigate the existing literature containing models for this task; detail the primary procedures involved in pre-processing the image. Evaluate MRI lesion segmentation strategies based on Convolutional neural networks; Incorporate a technique for automatic segmentation of lesions in MRIs of MS patients using magnetic resonance imaging analysis
Bulut için blok zincir tabanlı entegre ve kaynak sınıflandırması için yeni bir çerçeve sağlayın
Researchers in the manufacturing sector have recently shown an increased interest in the topic of cloud creation. Cloud generation is a customer-centric production strategy based on cloud computing, with the primary goal of providing access based on comprehensive service demand. The existing cloud architecture, on the other hand, has issues with a centralized network structure and operations. Basically, a centralized network offers insufficient flexibility, effectiveness, reliability, and security. In this study, the development of a decentralized network architecture for the cloud generation is made possible by blockchain technology. With Blockchain acting as a peer-to-peer network, a centralized peer-to-peer network might be created. The healthcare industry grew and became a significant actor in the provision of services. In recent years, patients have found remote observation services to be more beneficial. Healthcare practitioners create observation systems that diagnose and prescribe therapy in hospitals or clinics when patients are not actually present. Sensors are implanted in/on the bodies of patients. Wireless data interchange is possible across healthcare systems. Using characteristics from the blockchain, healthcare systems could be enhanced (e.g. distributed ledger, decentralized storage, and authentication). In this way, smart contracts automate the authentication and distribution of information, as well as participant criteria. When the number of users or other components increases, blockchains and smart contracts should maintain quality service. Scalability, increased security, and optimal performance are all technical challenges of smart contracts based on blockchains
Analysis of network security using cypher technique
Cryptographic procedures, such as encryption algorithms and cryptanalysis tools, necessitate a thorough understanding of mathematics. It's been used for a long time to teach these cryptographic concepts through hands-on experimentation supported by a theoretical framework. It's not enough to do theoretical hands-on trials if you're dealing with complex encryption approaches. There has been a huge improvement in educational programmes over the past few years, and cryptography is no exception. The goal of cryptographic programmes is to help students better comprehend the complex algorithms by presenting them visually in an understandable manner.
A new technique for peak-to-average power ratio reduction in 5G communication
Recently, due to the ongoing growth rising demand for a wide range of services and an increase in the number of devices with the highest level of reliability and the shortest possible delay, there is a rising need to achieve high data rates. As a result, the fifth generation (5G) is developed to deliver improved services at a high data rate, is then developed as a result of this. Instead of filtering the full band, it does it for sub-carriers. Out-of-band emissions (OOB) are reduced using sub-band filtering, which also minimizes carrier interference (ICI) between nearby users. The UFMC's fast latency, strong frequency offset, and low out-of-band (OOB) emission help to improve spectrum efficiency. However, the high peak-to-average power ratio (PAPR) problem at UFMC could impede the functionality of high-power amplifiers and lead to nonlinear distortion. The universal filtered multicarrier (UFMC) waveform has quasi-orthogonal properties among subcarriers and is cheap in complexity. Additionally, unlike the orthogonal frequency division multiplexing (OFDM) system, it can achieve significantly greater out-of-band emission performance. As a result, we suggest the Whale Optimization Algorithm in this study (WOA), which helped to greatly enhance the performance of the UFMC system by achieving a large reduction in PAPR. The UFMC system's BER and PAPR performance is assessed through simulation. MATLAB is used to run every simulation.
Hybrid deep learning model for automatic fake news detection
With the fast advancement in digital news, fake news has already caused grave threats to the public's actual judgment and credibility, in specific, with the wide use of social networking platforms, which provide a rich environment for the generation and dissemination of fake news. To cope with these challenges, several techniques were proposed to detect fake news, but still, there is an urgent need to propose an improved detection technique that provides a high level of detection performance in an automatic manner. Therefore, this article proposes a hybrid-improved deep learning model for automatic fake news detection. The proposed model adopts automatic data augmentation method, called Auxiliary Classifier Generative Adversarial Networks, to artificially synthesize new fake news samples, and then, hybridize the Convolutional Neural Network with the Recurrent Neural Networks to detect the fake news efficiently. The proposed model shows superior results against the state-of-the-art models as it provides 93.87% accuracy, 10.39% recall, 93.12%% precision in detecting the fake news using Buzzfeed, FakeNewsNet and FakeNewsChallenges datasets.