SASadeq Al-Suhemat

Sadeq Al-Suhemat

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Sadeq Al-Suhemat has a total of 7 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Offline Signature Verification Using Multi Classifiers: A Deep Feature Extraction Framework

    by Fadi Alsuhimat, Khalid Alemerien, Sadeq Alsuhemat, Lamis Alqoran, Enshirah Altarawneh

    Recently, there has been a lot of interest in offline signature verification utilising handwritten signature images. Despite the large number of studies on the topic, the performance of the methods used to verify signatures is insufficient. Additionally, there are a variety of issues that are frequently associated with signature verification systems during the phase of feature extraction, which researchers find to be extremely challenging to solve. This study proposes an enhanced model that can extract signature features and verify signatures. The signature features are extracted ‎from the image using Features from Accelerated Segment Test (FAST) and ‎Histogram Orientation Gradient (HOG), where (FAST) is used for selecting the ‎strength points in signature image, and then (HOG) is used to extract features ‎from the selected strength points. For our experiment, the databases UTSig and CEDAR were utilised. This study's fundamental improvement is in its ability to detect offline signature images, verify them, and determine the forged signature based on selected features points from the system database. We utilised three classifiers to assess the efficiency of the proposed method: ‎ Support Vector Machine, Long Short-Term Memory, and K-Nearest Neighbor. ‎Our results ensure the effectiveness of our proposed model in verifying offline signature and detecting forged signature with different types of datasets. The testing results showed that our proposed model performs fairly well regarding of performance and predictive capacity and had an accuracy of (91.6%, 92.4%, and 91.7%) with the UTSig dataset and (92.7%, 92.1%, and 91.7%) with the CEDAR dataset. When compared to several state-of-the-art techniques, our results are promising and have the potential to increase the reliability of signature verification.


  • Open Access

    A Hybrid CNN-RNN Model Changes the Game for Side-Channel Attack Prediction

    by Khalid Alemerien, Sadeq Al-Suhemat, Enshirah Altarawneh, Malek Alksasbeh, Tawfiq Alrawashdeh, Fadi Alsuhimat, Zaid T. Alhalhouli

    A common type of cyberattacks is the Side Channel Attack (SCA), which affects many devices and equipment connected to a network. These attacks have different types, such as power attacks like Differential Power Analysis (DPA), electromagnetic attacks, storage attacks, and others. Researchers and cyber security experts are concerned about SCA targeting devices, which can lead to losing and stealing important information. Many researchers have tried to reduce the impact of this attack. Deep Learning (DL) based solutions have shown significant promise in detecting SCAs due to their ability to identify complex patterns and anomalies in large datasets. Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN), including Long Short-Term Memory (LSTM) networks, are effective for analyzing time-series data and identifying temporal dependencies in side channel data. Despite the DL models having been successfully applied to detect power analysis, electromagnetic, and timing attacks, they still face data collection, model generalization, adversarial, and robustness challenges. This study aims to enhance side-channel attack prediction using deep-learning algorithms by investigating the use of the DL models in predicted SCA that focus on DPA, the DPA contest v4.2 dataset was used to predict power attacks, it contains 3253 features and 80,000 traces. Several data preprocessing techniques, including Hamming and Hanning windows, and feature selection using SelectKBest, were used to enhance model performance. The efficacy of CNN, RNN, and a hybrid CNN-RNN model was explored. The study results indicated that the hybrid CNN-RNN model achieved accuracy (99.6\%), loss (0.0270), and executed time (15,240.42 seconds). This study underscores the importance of optimal hyperparameter tuning, selecting appropriate activation functions (SeLU) and activation (Adam optimizer), in improving DL model performance.