A5065588879- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- Parallel Computing and Optimization Techniques
- EEG and Brain-Computer Interfaces
- Brain Tumor Detection and Classification
- IoT and Edge/Fog Computing
- Cryptography and Data Security
- Distributed systems and fault tolerance
- Real-Time Systems Scheduling
- Advanced Computing and Algorithms
- Artificial Intelligence in Healthcare
- Blind Source Separation Techniques
- Embedded Systems Design Techniques
- Machine Learning and ELM
- Coding theory and cryptography
- Electric Vehicles and Infrastructure
- ECG Monitoring and Analysis
- Machine Learning in Healthcare
- Security in Wireless Sensor Networks
- Functional Brain Connectivity Studies
- Network Security and Intrusion Detection
- Handwritten Text Recognition Techniques
- Peer-to-Peer Network Technologies
- Autism Spectrum Disorder Research
- Vehicle License Plate Recognition
University of Management and Technology
2021-2025
COMSATS University Islamabad
2011-2023
Shaheed Zulfiqar Ali Bhutto Institute of Science and Technology
2018-2021
Government of Khyber Pakhtunkhwa
2021
Autism spectrum disorder is a severe, life‐prolonged neurodevelopmental disease typified by disabilities that are chronic or limited in the development of socio‐communication skills, thinking abilities, activities, and behavior. In children aged two to three years, symptoms autism more evident easier recognize. The major part existing literature on covered prediction system based traditional machine learning algorithms such as support vector machine, random forest, multiple layer perceptron,...
In this paper, a model based on discrete wavelet transform and convolutional neural network for brain MR image classification has been proposed. The proposed is comprised of three main stages, namely preprocessing, feature extraction, classification. the median filter applied to remove salt-and-pepper noise from MRI images. transform, Harr used. model, 3-level decomposition images low-level detail reduce size Next, used classifying into normal abnormal. also prevalent method widely in...
With technological advancement, cloud computing paradigms are gaining massive popularity in the ever-changing advancement. The main objective of system is to provide on-demand storage and resources users on pay-per-use policy. It allows small businesses use top-notch infrastructure at low expense. However, due resource sharing property, data privacy security significant concerns barriers for smart systems constantly transfer generated resources, which a third-party provider manages. Many...
Conventional RSA algorithm, being a basis for several proposed cryptosystems, has remarkable security laps with respect to confidentiality and integrity over the internet which can be compromised by state-of-the-art attacks, especially, different types of data generation, transmission, analysis IoT applications. This threat hindrance is considered hard problem solve on classical computers. However, bringing quantum mechanics into account, concept no longer holds true. So, this calls out...
Vaccine acceptance is a crucial component of viable immunization program in healthcare system, yet the disparities new and existing vaccination adoption rates prevail across regions. Disparities rate vaccine result low coverage slow uptake newly introduced vaccines. This research presents an innovative AI-driven predictive model, designed to accurately forecast within programs, while providing high interpretability. Primarily, contribution this study classify acceptability into Low, Medium,...
An efficient resource allocation scheme plays a vital role in scheduling applications on high-performance computing resources order to achieve desired level of service. The major part the existing literature is covered by real-time services having timing constraints as primary parameter. Resource schemes for have been designed with various architectures (static, dynamic, centralized, or distributed) and quality service criteria (cost efficiency, completion time minimization, energy memory...
Detection of epileptic seizures on the basis Electroencephalogram (EEG) recordings is a challenging task due to complex, non-stationary and non-linear nature these biomedical signals. In existing literature, number automatic seizure detection methods have been proposed that extract useful features from EEG segments classify them using machine learning algorithms. Some characterizing non-epileptic signals overlap; therefore, it requires analysis must be performed diverse perspectives. Few...
Grid resource allocation mechanism maps tasks to the available grid resources according some predefined criterion, such as minimizing makespan or execution cost, load balancing, energy efficiency, maintaining user-defined task deadlines, and efficiently using memory. The minimization of is a dominant criterion more challenging when computationally intensive have realtime deadlines data requirements. Such require files for processing that are transferred from storage computing resources,...
Datacentres provide the foundations for cloud computing, but require large amounts of electricity their operation. Approaches that promise to reduce power use by minimizing execution time, example using different scheduling and resource management techniques, are discussed in literature. This paper summarizes some most important techniques clouds focusing on consumption, covering VM-level, host-level task-level where promising approach is task level scheduling, with energy savings means load...
Brain MRI classification is one of the key areas research. The brain can help radiologists in different disease diagnostics without invasive measures. a difficult task due to variance and complexity diseases. We have proposed novel efficient binary model for images. includes discrete wavelet transform (DWT) used features extraction, statistical diminishing number features, blended artificial neural network classification. with less challenging task. In this paper, we technique statical...
In this paper, we have proposed a novel methodology based on statistical features and different machine learning algorithms. The model can be divided into three main stages, namely, preprocessing, feature extraction, classification. the preprocessing stage, median filter has been used in order to remove salt-and-pepper noise because MRI images are normally affected by type of noise, grayscale also converted RGB stage. histogram equalization enhance quality each channel. extraction channels,...
Abstract When there is a mismatch between the cardinality of periodic task set and priority levels supported by underlying hardware systems, multiple tasks are grouped into one class so as to maintain specific level confidence in their accuracy. However, such transformation achieved at expense loss schedulability original set. We further investigate aforementioned problem report following contributions: (i) novel technique for mapping unlimited reduced number classes that do not violate (ii)...
The mobility of user equipment (UE) in cellular network is a challenging issue terms its management. Current traditional handover Long-Term Evolution (LTE) managed by evolved Node B or eNodeB (eNB) which decentralized solution. In contrast to existing technology, software defined (SDN) has the capability serving packets switching without involving SDN controller except for first one. We have proposed an SDN-based centralized solution management LTE network. Based on our solution, being keeps...
Among several approaches to privacy-preserving cryptographic schemes, we have concentrated on noise-free homomorphic encryption. It is a symmetric key encryption that supports operations encrypted data. We present fully (FHE) scheme based sedenion algebra over finite ℤ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</sub> rings. The innovation of the compression 16-dimensional vector for application Frobenius automorphism. For sedenion, p <sup...
Cloud computing is the de facto platform for deploying resource- and data-intensive real-time applications due to collaboration of large scale resources operating in cross-administrative domains. For example, systems are generated by smart devices (e.g., sensors homes that monitor surroundings real-time, security cameras produce video streams cloud gaming, social media streams, etc.). Such low-end form a microgrid which has low computational storage capacity hence offload data unto...
Wireless Sensor Networks (WSNs) provide noteworthy advantages over conventional methods for various real-time applications, i.e., healthcare, temperature sensing, smart homes, homeland security, and environmental monitoring. However, limited resources, short life-time network constraints, security vulnerabilities are the challenging issues in era of WSNs. Besides, WSNs performance is susceptible to anomalies, particularly misdirection attacks. The above-mentioned pose our attentions produce...
Load balancing among cores or systems plays a vital role in overall performance of High Performance Computing (HPC). Efficient results may not be obtained unless specific load is properly balanced HPC. The main focus this research multi-core environment by using task shifting (migration) and splitting mechanism. In mechanism having minimum all the tasks on high utilized core fully shifted migrated to low for balancing. mechanism, from shared core, loads cores. It concludes given that balance...