International Journal of Electrical and Computer Engineering Systems

Optimized Triple-Slot Patch Antenna with Electromagnetic Band Gap Structures for Enhanced Performance

2025-04-08T10:32:30+02:00

This paper presents an antenna integrated with an Electromagnetic Bandgap (EBG) structure to enhance its radiation performance compared to a conventional antenna. MEBG (Mushroom EBG) and EEBG (Edge via EBG) structures are analyzed, integrating MEBG with the triple-slot patch antenna, which demonstrates superior performance. Using the same conventional dimensions, the proposed antenna achieves a gain of 6.15 dB, a directivity of 7.51 dB, and a return loss of 37 dB at 5.2 GHz, providing a 1.92 dB gain improvement over the conventional design. This design is simulated using the HFSS software. The measurement results are validated with simulation results. The fabricated, compact antenna can be used for IoT applications at 5.2 GHz.

Lightweight Block Cipher for Security in Resource-Constrained Network

2025-04-14T13:50:03+02:00

As the proliferation of resource-constrained devices continues in various application domains, the need for energy-efficient cryptographic algorithms becomes paramount for ensuring their security. Lightweight block ciphers play a crucial role in securing communication and data integrity in resource-poor environments. This paper presents the design, simulation, and evaluation of a novel symmetric Energy Efficient Lightweight Block Cipher (EE-LBC), tailored for such environments, which employs a balanced combination of substitution-permutation network (SPN) structure with larger diffusion and substitution box activation properties to achieve high security with minimal energy consumption and implementation cost. Through rigorous cryptanalysis and performance evaluations, EE-LBC demonstrates superior throughput and efficiency compared to prevailing lightweight block ciphers, making it an ideal choice for securing resource-constrained network.

Optimized t-Test Feature Selection for Real-Time Detection of Low and High-Rate DDoS Attacks

2025-03-02T18:30:02+01:00

Distributed Denial of Service (DDoS) attacks stand out as a serious threat, capable of disrupting online services and businesses. The main aim of Distributed Denial of Service (DDoS) attacks is to make system services unavailable to the legitimate users. To detect these attacks, intrusion detection systems (IDS) continually monitor the network traffic. During this process, the IDS system generates high false positive rates while distinguishing low-rate DDoS (LRDDoS) and high-rate DDoS (HRDDoS) attack traffic from legitimate traffic. The idea behind feature selection is that picking the right network features is a key part of interpreting the difference between normal traffic and LRDDoS or HRDDoS attack traffic. This means the IDS performance will automatically get better. In this paper, we propose a scalable feature selection method that utilizes the statistical t-test to identify an optimal feature subset from original feature set at a low computational cost. We strongly hypothesize that the proposed feature selection method yields an optimal feature subset and the machine learning classifiers trained on this feature set can effectively distinguish benign, LRDDoS, and HRDDoS network traffic. We evaluated the proposed method on the publicly available benchmark datasets CICIDS2017, CICIDS2018, and CICDDoS2019, utilizing twelve supervised machine learning classifiers. Among the twelve classifiers, the Extra Tree Classifier (EXT) demonstrated superior performance, achieving an average accuracy of 96.50%, precision of 96.58%, and an F-Score of 96.50% across the four sample test datasets (D1, D2, D3, and D4). The proposed method showed consistent and superior performance in distinguishing the LRDDoS, HRDDoS, and benign traffic to the state-of-the-art existing works over the four test datasets.

Echocardiographic Left Ventricular Segmentation Using Double-layer Constraints on Spatial Prior Information

2025-04-23T12:44:56+02:00

Real-time segmentation of echocardiograms is of great practical significance for doctors' clinical diagnosis. This paper addresses the existing echocardiogram segmentation models' pursuit of high segmentation accuracy in insufficient training data, which leads to high model complexity and low learning efficiency. This paper fully exploits the spatial prior characteristics of the image itself. It proposes an echocardiographic left ventricular segmentation algorithm that utilizes double-layer constraints of prior information on spatial anatomical structures. The algorithm is based on the following two principles. Firstly, the segmentation model is initialized using a self-supervised sorting model based on the spatial anatomy to fully learn the orderly image features of the left ventricular spatial anatomy and achieve same-domain transfer of images, allowing the segmentation network to learn segmentation information more effectively; Secondly, the segmentation network is subjected to mask shape constraints, and the output space is limited by imposing anatomical shape priors to expand the global training goals of the CNN model. Finally, the algorithm proposed in this paper was verified using three classic segmentation models. The experimental results showed that on the public echocardiography dataset CETUS (Challenge on Endocardial Three-dimensional Ultrasound Segmentation), compared with the classic Resnet, Unet, and VGG segmentation models, the double-layer constrained segmentation model that introduces prior features has increased the segmentation accuracy (Dice index) by 5.6%, 4.9%, and 4.8%, respectively. The MIOU (Mean Intersection over Union) index increased by 7%, 5.5%, and 6.8%, respectively, demonstrating robustness to slice misalignment.

Computational intelligence in chromosomal primitive extraction and speaker recognition

2025-04-04T13:20:10+02:00

This research presents an innovative approach leveraging computational intelligence for chromosomal primitive extraction and speaker recognition. The study emphasizes real-time digital signal processing (DSP) embedded systems integrating chromosomal-inspired techniques to enhance auditory feature extraction and speaker identification accuracy. By applying Gamma chromosomal factors, Mel-Frequency Cepstral Coefficients (MFCC) are refined through convolution, emulating human cochlear functionality. This integration aligns well with the perceptual auditory mechanisms and computational intelligence paradigms. The proposed methodology incorporates feature extraction techniques like Linear Predictive Coding (LPC), Linear Predictive Cepstral Coefficients (LPCC), and MFCC, followed by robust classifiers such as Support Vector Machines (SVM), Artificial Neural Networks (ANN), and Recurrent Self-Organizing Maps (RSOM). Experimental results demonstrate superior performance of RSOM, achieving a recognition rate of up to 99.7% with Gamma-enhanced MFCCs, compared to 98.6% for SVM and 91% for SOM. The RSOM model effectively identifies speakers across diverse conditions, albeit with slightly increased response times due to its dynamic recurrence loop. This work addresses challenges like environmental noise and variability in speech styles by introducing the Gamma chromosomal factor, a logarithmic nonlinear enhancement model. The experimental setup, executed on DSP boards using Python, highlights the advantages of computationally intelligent systems in real-world applications such as biometric authentication and decision-making systems. These findings underscore the potential of chromosomal-inspired computational techniques to advance speaker recognition technology, offering high accuracy and reliability in adverse conditions. Future research will focus on optimizing architectural and software frameworks to improve response times and further integrate this approach into constrained real-time systems.

A Scalable Distributed Approach for Exploration Global Frequent Patterns

2025-04-24T11:59:48+02:00

Finding patterns in transactional databases regularly is an essential part of data mining since it makes it simpler to identify significant connections and reoccurring patterns in datasets. Scalable, high-performance computing solutions that employ parallel computing systems to optimize resource efficiency and data analysis as data volumes continue to grow are necessary for efficiently processing large databases. To solve these issues, this paper presents Exploration Global Frequent Patterns (EGFP), a new parallel algorithm designed to generate global frequent patterns in different distributed datasets. By facilitating the distribution of workloads and data partitioning, the approach reduces communication costs and ensures efficient parallel execution. Our approach uses two prefix-tree structures to generate a significantly compacted and structured representation of frequent patterns. The first structure local-tree serves to store local support values to effectively collect and arrange transaction data. Global prefix counts are then aggregated and ranked to improve frequency-based analysis and provide a more organized and useful representation of frequent patterns. To find the globally prevalent patterns, a Master site develops a second structure global-tree for each prefix based on this arranged data. Experimental results on large-scale benchmark datasets show that EGFP outperforms other existing methods including CD and PFP-tree in terms of execution time and scalability, while incurring considerably less communication cost.