Welcome to SMART lab
Analyzing Aging Effects on SRAM PUFs: Implications for Security and Reliability
2023 – Niraj Prasad Bhatta, Harshdeep Singh, Fathi Amsaad
Impact of aging effects on Static Random-Access Memory Physical Unclonable Functions (SRAM PUFs) presents critical implications for security and reliability in modern hardware. Emerging as promising hardware-based security primitives, SRAM PUFs harness process variations in integrated circuits for secure key generation and device authentication. However, aging phenomena like Bias Temperature Instability (BTI) and Hot Carrier Injection (HCI) can potentially alter SRAM cell characteristics, compromising PUF responses. This study delves into the multifaceted challenges of aging-induced variations, exposing underlying security vulnerabilities and offering innovative strategies to mitigate risks. Examining reliability implications, it introduces mitigation techniques such as adaptive reconfiguration, error correction codes, and multi-modal PUFs to enhance the resilience of SRAM PUFs. The investigation concludes by mapping future research directions and prospects for improving SRAM PUF-based security solutions, in light of the complexities associated with semiconductor device aging.
Accelerated Aging Effects on SRAM PUF reliability at various Temperature and Voltage conditions
2023 – Harshdeep Singh, Niraj Prasad Bhatta, Md Tauhidur Rahman, Fathi Amsaad
This paper thoroughly investigates the trustworthiness of Static Random-Access Memory (SRAM) Physical Unclonable Functions (PUFs) within hardware security. It presents an overview of SRAM PUFs, their significance in hardware security, and their essential features. The paper explores the realms of evaluation metrics and practical examination of SRAM cell start-up values concerning an 8-bit SRAM dataset. Concurrently, it navigates the encountered challenges. A comprehensive pattern analysis of SRAM cell stability is discussed with accelerated aging. Compared the outputs at multiple trials to authenticate the result. The complex interplay of factors that affect the dependability of SRAM PUFs, encompassing environmental conditions, the accumulation of aging effects, and the complex landscape of process variations, is carefully explained and clarified. The paper concludes by summarizing key findings, emphasizing the importance of reliable hardware security, and suggesting future research directions for enhancing SRAM PUF reliability.
A Golden-Free Unsupervised ML-Assisted Security Approach for Detection of IC Hardware Trojans
2023 – Ashutosh Ghimire, Mohammed Alkurdi, Fathi Amsaad, Md Tauhidur Rehman
Hardware Trojans are deliberate malicious hardware modifications inserted in semiconductor Integrated circuits (ICs) for the purpose of stealing or leaking sensitive information, as well as disrupting critical systems upon activation, underscoring the importance of robust detection mechanisms. Emerging hardware security research highlight the criticality of employing AI for effective detection within the semiconductor IC supply chain. The efficient detection of these malicious Trojan circuits is of utmost significance, as it holds paramount importance in cultivating trust within the semiconductor IC supply chain. However, prevailing detection methodologies, predominantly reliant on side-channel analysis, often necessitate the utilization of golden chips for validation. This paper heralds a new era in Hardware Trojan detection, harnessing the prowess of unsupervised machine learning in conjunction with side-channel analysis to eliminate the need for golden data. Through FPGA-based experimentation involving Trojans of varying dimensions, the efficacy of this innovative approach was evaluated. Employing unsupervised clustering, the methodology effectively uncovered anomalies. The application of unsupervised learning techniques not only showcased a superior false positive rate but also demonstrated a comparable accuracy level when compared to supervised counterparts such as the K-Nearest Neighbors (KNN) classifier, Support Vector Machine (SVM), and Gaussian classifier—methods reliant on the availability of golden data for training. Notably, the proposed model exhibited an impressive accuracy rate of 93%, particularly excelling in …
Malicious Vehicle Detection Using Layer-Based Paradigm and the Internet of Things
2023 – Abdul Razaque, Gulnara Bektemyssova, Joon Yoo, Aziz Alotaibi, Mohsin Ali, Fathi Amsaad, Saule Amanzholova, Majid Alshammari
Deep learning algorithms have a wide range of applications, including cancer diagnosis, face and speech recognition, object recognition, etc. It is critical to protect these models since any changes to them can result in serious losses in a variety of ways. This article proposes the consortium blockchain-enabled conventional neural network (CBCNN), a four-layered paradigm for detecting malicious vehicles. Layer-1 is a convolutional neural network-enabled Internet-of-Things (IoT) model for the vehicle; Layer-2 is a spatial pyramid polling layer for the vehicle; Layer-3 is a fully connected layer for the vehicle; and Layer-4 is a consortium blockchain for the vehicle. The first three layers accurately identify the vehicles, while the final layer prevents any malicious attempts. The primary goal of the four-layered paradigm is to successfully identify malicious vehicles and mitigate the potential risks they pose using multi-label classification. Furthermore, the proposed CBCNN approach is employed to ensure tamper-proof protection against a parameter manipulation attack. The consortium blockchain employs a proof-of-luck mechanism, allowing vehicles to save energy while delivering accurate information about the vehicle’s nature to the “vehicle management system.” C++ coding is employed to implement the approach, and the ns-3.34 platform is used for simulation. The ns3-ai module is specifically utilized to detect anomalies in the Internet of Vehicles (IoVs). Finally, a comparative analysis is conducted between the proposed CBCNN approach and state-of-the-art methods. The results confirm that the proposed CBCNN approach outperforms competing …
2023 – Vishnu Vardhan Baligodugula, Fathi Amsaad, Varshitha Vunnam Tadepalli, Vojjala Radhika, Yetukuri Sanjana, Sai Shiva, Supriya Meduri, Majdi Maabreh, Naser Alsaadi, Omar Darwish, Abdul Razaque, Yahya Tashtoush
Cloud-based processes store enormous amounts of data to support several daily applications, especially in intelligent systems and the Industrial Internet of Things (IIoT) applications. The advancement of hosted cloud storage services endorses the productivity of IIoT applications. It is critical to enabling businesses to increase data security, privacy, efficiency, accessibility, and flexibility. There are many concerns for cloud providers, including storage overhead, upload/download bandwidth, and data security and privacy. Utilizing secure data deduplication technology, cloud service providers may maximize their available storage capacity, resulting in effective and secure disk space use. This review focuses on a comparative study of secure data deduplication mechanisms for cloud-based applications, including MD5 and various versions of SHA. These techniques remove duplicate data in different storage levels and …
Toward Secure and Efficient CNN Recognition with Different Activation and Optimization Functions
2023 – Fathi Amsaad, PL Prasanna, T Pravallika, G Mamatha, B Raviteja, M Lakshmi, Nasser Alsaadi, Abdul Razaque, Yahya Tashtoush
Precise data input classification and identification are essential for developing high-quality intelligent systems, including defending systems against cyberattacks and guaranteeing data protection. The Internet of Things (IoT), in particular, offers specific problems for cybersecurity not because of the volumes of data generated in a short time but also for the diversity of data sources. Thus, creating an accurate identification system is crucial to preserve data integrity and security. It has been demonstrated that convolutional neural networks (CNNs) perform well in developing data-driven applications, including cybersecurity. However, hyperparameters play a significant role in their prediction quality which may increase/decrease the false alarm rates. CNNs’ performance may be improved by the activation and optimization functions utilized using nonlinear expression and reducing loss functions, enabling them to be more …
A comprehensive review on blockchains for Internet of Vehicles: Challenges and directions
2023 – Brian Hildebrand, Simra Tabassum, Bharath Konatham, Fathi Amsaad, Mohamed Baza, Tara Salman, Abdul Razaque
Internet of Vehicles (IoVs) consists of smart vehicles, Autonomous Vehicles (AVs) as well as roadside units (RSUs) that communicate wirelessly to provide enhanced transportation services such as improved traffic efficiency and reduced traffic congestion and accidents. Unfortunately, current IoV networks suffer from security, privacy, and trust issues. Blockchain technology emerged as a decentralized approach for enhanced security without depending on trusted third parties to run services. Blockchain offers the benefits of trustworthiness and immutability and mitigates the problem of a single point of failure and other attacks. In this work, we present the state-of-the-art Blockchain-enabled IoVs (BIoVs) with a particular focus on their applications, such as crowdsourcing-based applications, energy trading, traffic congestion reduction, collision, accident avoidance, infotainment, and content caching. We also present in …
Unsupervised-based Distributed Machine Learning for Efficient Data Clustering and Prediction
2023 – Vishnu Vardhan and Fathi Amsaad
Machine learning techniques utilize training data samples to help understand, predict, classify, and make valuable decisions for different applications such as medicine, email filtering, speech recognition, agriculture, and computer vision, where it is challenging or unfeasible to produce traditional algorithms to accomplish the needed tasks. Unsupervised ML-based approaches have emerged for building groups of data samples known as data clusters for driving necessary decisions about these data samples and helping solve challenges in critical applications. Data clustering is used in multiple fields, including health, finance, social networks, education, and science. Sequential processing of clustering algorithms, like the K-Means, Minibatch K-Means, and Fuzzy C-Means algorithms, takes a long time, especially with many data samples, regardless of whether the results obtained may be accurate or not. This thesis proposes parallel and distributed computing unsupervised ML techniques to improve the execution time of different ML algorithms. The application of different ML techniques on each system and their specific variations is outlined. Various parallelized unsupervised ML models are developed, implemented, and tested to demonstrate the efficiency, in terms of execution time and accuracy, of the serial methods as compared to the parallelized ones. For that, parallel K-Means, parallel Minibatch K-Means, and Fuzzy parallel C-Means using an MPI model are developed. A distributed time estimation approach is created that utilizes the AWS could computing architecture. The Sequential, Parallel, and distributed approaches of K-Means …
Implementation of Secure and Privacy-aware AI Hardware using Distributed Federated Learning
2023 – Ashutosh Ghimire, Ahmad Nasser Asiri, Brian Hildebrand, Fathi Amsaad
In modern devices, such as smartphones and IoT, AI hardware implements different ML models to train massive amounts of data for various applications. However, based on the sensitivity of this data, privacy and security concerns, or both, may restrict users from accessing the data storage to conduct the ML training using conventional methods. Federated learning (FL) is consequently emerged to maintain training data distribution among smart mobile devices while aggregating locally processed updates. In addition to improving the model training performance as the number of clients rises, FL also creates a privacy-preserved shared data model. FL execution, however, can be time-consuming. This study proposes a parallelized approach to enhance the performance and privacy of the FL algorithm (FedAvg). In this regard, the FedAvg algorithm is expanded to our proposed model, distributed FedAvg (D-FedAvg …
2023 – Abdul Razaque, Mohamed Ben Haj Frej, Gulnara Bektemyssova, Muder Almi’ani, Fathi Amsaad, Aziz Alotaibi, Noor Z Jhanjhi, Mohsin Ali, Saule Amanzholova, Majid Alshammari
The Quality-of-Service (QoS) provision in machine learning is affected by lesser accuracy, noise, random error, and weak generalization (ML). The Parallel Turing Integration Paradigm (PTIP) is introduced as a solution to lower accuracy and weak generalization. A logical table (LT) is part of the PTIP and is used to store datasets. The PTIP has elements that enhance classifier learning, enhance 3-D cube logic for security provision, and balance the engineering process of paradigms. The probability weightage function for adding and removing algorithms during the training phase is included in the PTIP. Additionally, it uses local and global error functions to limit overconfidence and underconfidence in learning processes. By utilizing the local gain (LG) and global gain (GG), the optimization of the model’s constituent parts is validated. By blending the sub-algorithms with a new dataset in a foretelling and realistic setting, the PTIP validation is further ensured. A mathematical modeling technique is used to ascertain the efficacy of the proposed PTIP. The results of the testing show that the proposed PTIP obtains lower relative accuracy of 38.76% with error bounds reflection. The lower relative accuracy with low GG is considered good. The PTIP also obtains 70.5% relative accuracy with high GG, which is considered an acceptable accuracy. Moreover, the PTIP gets better accuracy of 99.91% with a 100% fitness factor. Finally, the proposed PTIP is compared with cutting-edge, well-established models and algorithms based on different state-of-the-art parameters (e.g., relative accuracy, accuracy with fitness factor, fitness process, error reduction, and …
Information-Centric IoT-Based Smart Farming with Dynamic Data Optimization
2023 – Souvik Pal, Hannah VijayKumar, D Akila, NZ Jhanjhi, Omar A Darwish, Fathi Amsaad
Smart farming has become a strategic approach of sustainable agriculture management and monitoring with the infrastructure to exploit modern technologies, including big data, the cloud, and the Internet of Things (IoT). Many researchers try to integrate IoT-based smart farming on cloud platforms effectively. They define various frameworks on smart farming and monitoring system and still lacks to define effective data management schemes. Since IoT-cloud systems involve massive structured and unstructured data, data optimization comes into the picture. Hence, this research designs an Information-Centric IoT-based Smart Farming with Dynamic Data Optimization (ICISF-DDO), which enhances the performance of the smart farming infrastructure with minimal energy consumption and improved lifetime. Here, a conceptual framework of the proposed scheme and statistical design model has been well defined. The information storage and management with DDO has been expanded individually to show the effective use of membership parameters in data optimization. The simulation outcomes state that the proposed ICISF-DDO can surpass existing smart farming systems with a data optimization ratio of 97.71%, reliability ratio of 98.63%, a coverage ratio of 99.67%, least sensor error rate of 8.96%, and efficient energy consumption ratio of 4.84%.
E-Learning Course Recommender System Using Collaborative Filtering Models
2022 – Kalyan Kumar Jena, Sourav Kumar Bhoi, Tushar Kanta Malik, Kshira Sagar Sahoo, NZ Jhanjhi, Sajal Bhatia, Fathi Amsaad
e-Learning is a sought-after option for learners during pandemic situations. In e-Learning platforms, there are many courses available, and the user needs to select the best option for them. Thus, recommender systems play an important role to provide better automation services to users in making course choices. It makes recommendations for users in selecting the desired option based on their preferences. This system can use machine intelligence (MI)-based techniques to carry out the recommendation mechanism. Based on the preferences and history, this system is able to know what the users like most. In this work, a recommender system is proposed using the collaborative filtering mechanism for e-Learning course recommendation. This work is focused on MI-based models such as K-nearest neighbor (KNN), Singular Value Decomposition (SVD) and neural network–based collaborative filtering (NCF) models. Here, one lakh of Coursera’s course review dataset is taken from Kaggle for analysis. The proposed work can help learners to select the e-Learning courses as per their preferences. This work is implemented using Python language. The performance of these models is evaluated using performance metrics such as hit rate (HR), average reciprocal hit ranking (ARHR) and mean absolute error (MAE). From the results, it is observed that KNN is able to perform better in terms of higher HR and ARHR and lower MAE values as compared to other models.
Credit Card-Not-Present Fraud Detection and Prevention Using Big Data Analytics Algorithms
2022 – Abdul Razaque, Yaser Jararweh, Aziz Alotaibi, Fathi Amsaad, Bandar Alotaibi, Munif Alotaibi
The rapid development of the emerging Internet of things (IoT) regime and of smart applications has given rise to many new wireless security vulnerabilities in consumer electronic environment networks, which compromises the integrity of the whole environment. Attacks often encompass the abuse of vulnerable wireless IoT consumer electronics devices, compromise information security, and leak sensitive and private data. This paper proposes a decentralized blockchain-enabled framework network infrastructure to improve wireless security and mitigate new wireless attacks on connected consumer electronics as a proof of concept. The system model of the proposed framework was verified mathematically and was then implemented and tested in case studies. The experimental results show that the proposed decentralized blockchain technology-featured air-cracking tool (BTFAT) provides a robust detection of …
The Role of Cutting-Edge Technologies in Industry 4.0
2022 – Imdad Ali Shah, Noor Zaman Jhanjhi, Fathi Amsaad, Abdul Razaque
The Fourth Industrial Revolution (Industry 4.0) represents a significant change in how we live and interact. A new era of human development has begun due to massive technological developments similar to the Industrial Revolution. These advances are integrating the physical and digital worlds in a way that also carries significant risks. Revolutions are driving us to rethink how nations evolve. Industry 4.0 offers opportunities for all citizens from all socioeconomic levels and countries. There is real hope in going outside of technology and creating a method to enable it. It takes a large number of people to have a positive influence on an individual, organization, and community. Achieving resource-efficient and user-friendly production solutions require intelligent and precise equipment. Industry 4.0 is expected to benefit from new technologies and applications by increasing production and providing customized …