Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Path planning and obstacle avoidance for UAVs using Theta* and modulated velocity obstacle avoidance with 2D LiDAR
Full text linkThis paper proposes a novel framework for autonomous unmanned aerial vehicle (UAV) navigation in complex environments, seamlessly integrating Theta* for global path planning with a simplified modulated velocity obstacle avoidance (MVOA) algorithm for local obstacle avoidance. Theta* generates optimal, smooth paths, while MVOA processes 2D LiDAR data as a single obstacle block to compute modulated velocities, enabling efficient avoidance of static and dynamic obstacles with minimal computational overhead. Compared to MVOA-only navigation, the integration of Theta* and MVOA produced shorter trajectories and faster mission completion with smoother velocity adjustments, demonstrating clear improvements in efficiency and stability. Simulation results show the framework maintains a 0.6 m safety distance and operates at 10 Hz, underscoring its robustness and reliability. The resulting control velocity is transmitted to an ArduPilot-based flight controller via MAVLink, ensuring precise, real-time execution. The current implementation focuses on 2D navigation in a planar environment as a foundation for future 3D expansion, with all results obtained through high-fidelity simulation. Building on these findings, the framework shows strong potential for real-time applications such as swarm UAV coordination, terrain surveying, and indoor navigation, offering a scalable solution for autonomous systems in dynamic settings
Fast lightweight convolutional neural network for Turkish sentiment analysis
Full text linkThis study presents a fast, lightweight, and high-performing fast convolutional neural network (Fast CNN) model tailored for Turkish sentiment analysis (SA). The agglutinative morphology of Turkish, combined with the limited availability of high-quality linguistic resources, introduces significant challenges for conventional approaches. To address these issues, we propose a streamlined Fast CNN architecture consisting of an embedding layer, global max-pooling, dropout, and fully connected layers. Despite its simplicity, the model outperforms seven state-of-the-art convolutional neural network (CNN)-based systems across four benchmark Turkish sentiment datasets. It achieves an average area under the curve (AUC) of 0.94, representing a 6.8% improvement over the strongest baseline and a gain of over 80% relative to several deeper architectures. In addition to its superior accuracy, the model demonstrates reduced computational complexity, making it well-suited for real-world deployment in resourceconstrained environments. Potential applications include customer feedback mining and digital marketing analytics in Turkish-language domains
Evaluating the effectiveness of Havij for structured query language injection exploitation in web applications
Full text linkStructured query language injection (SQLi) is still one of the most critical risks to web application security, as it allows attackers to interfere with sensitive data and even a complete database infrastructure. Although many automated tools are available, previous studies usually achieve only descriptive briefs, which do not offer empirical assessments that measure the performance and the usability. This research fills this void by a systematic five-stage experimental analysis of the Havij automated SQLi tool under a controlled and ethical test setup. Confirmation of vulnerability, automated exploitation, data extraction and benchmarking of performance were performed as the methodology, and the results were compared against the industry standard SQLmap tool. It was found that in less than a minute Havij was able to locate the target database, scan its structure, and steal authentication credentials, which is quite efficient and user-friendly. In contrast to the literature, our work presents not only quantitative measures (time-to-exploit, request volume, and success rate) but also a qualitative evaluation (user accessibility and limitations), which gives a comprehensive evaluation. The results highlight trade-offs between the depth and accessibility, the continued dangers of SQLi in practice, and provide recommendations that developers and security experts can implement
Optimizing gaussian filter implementation for canny edge detection using graph-based MCM algorithms
Full text linkThis study presents an optimized implementation of the gaussian filter in the Canny edge detection algorithm, focusing on reducing computational complexity while balancing power, timing, and resource utilization. Traditional implementations rely on the common subexpression elimination (CSE) algorithm for multiplierless constant multiplication, which results in high logic operations and resource consumption. To address this, we explore the constant array vector multiplication (CAVM) technique with two graph-based algorithms (exact GB and approximate GB). These algorithms offer a novel graph-structured approach to constant multiplication, differing from existing methods by modeling multiple paths to achieve optimal adder reuse. The architectures were implemented using Xilinx system generator (XSG) and evaluated in Vivado 2018.1. Experimental results reveal that both exact GB and approximate GB reduce logic operations and improve timing performance compared to CSE_csd. Among them, approximate GB achieves the fastest computation and lowest LUT utilization, making it the most hardware-efficient design. However, it exhibits the highest power consumption, whereas exact GB offers the best trade-off between speed and power efficiency. This optimization framework shows potential not only in image processing but also in embedded vision systems and low-power digital signal processing (DSP) applications. These findings demonstrate that GB Algorithms can effectively optimize gaussian filter design for real-time image processing applications
MIMO-enhanced distributed spectrum sensing with diffusion based algorithms for cognitive radio systems
Full text linkSpectrum sensing (SS) is a fundamental function in cognitive radio (CR) networks, enabling efficient spectrum utilization by identifying available channels. However, existing SS methods face challenges such as low accuracy in dynamic and low signal-to-noise ratio (SNR) environments, as well as high computational complexity. To address these issues, this paper presents a distributed SS technique that combines multiple-input multiple-output (MIMO) technology with a diffusion-based (DB) cooperative algorithm. MIMO enhances spatial diversity to improve detection performance, while the DB algorithm enables efficient collaboration among secondary users, reducing both sensing time (ST) and computational time (CT). Simulations over Rayleigh (RL) and Rician (RC) fading channels evaluated metrics such as probability of detection and false alarm. Results demonstrate that the proposed MIMO-DB method outperforms existing approaches, including honey badger remora optimization (HBRO)-AlexNet, by reducing ST by 18 seconds and CT by 45 seconds at 5 dB SNR, while achieving higher detection accuracy across varying SNR levels. These findings highlight the method’s robustness and efficiency, making it a promising solution for dynamic spectrum management in 5G, internet of thing (IoT) and other next-generation wireless systems
Evaluating maintainability metrics in microservices-based student registration systems
Full text linkAs governments redefine educational policy and schools evolve their priorities, more schools must have software that recalibrates with minimal friction. To provide objective guidelines, this study rigorously measures maintainability attributes in a microservices-styled student registration platform, framing the assessment with the ISO/IEC 25010 maintainability specification. We steered each of the standard's maintainability sub-characteristics into defined quantitative constructs, executed in the context of a production microservices topology. Architectural and behavioural views were analysed using Structure101 in static tool runs, and unified modeling language (UML) model inspection anchored the derivation of key metrics, ensuring that stakeholder-defined structures and live microservices concurrency both shaped the evaluation. Results indicate moderate system modularity with average component dependency (ACD) of 2.14, propagation cost (PC) of 10.2%, and identification of one non-trivial cycle group involving three classes. Cohesion analysis revealed structural improvement opportunities in core classes such as admin and candidate lack of cohesion in methods 4 (LCOM4)≥2). The inheritance structure shows optimal characteristics with shallow depth (depth of inheritance tree (DIT)≤1), and controlled breadth (number of children (NOC)=2), supporting both analyzability and modifiability. These findings provide actionable insights for enhancing system maintainability in microservices architectures, particularly for educational domain applications requiring frequent policy adaptations
Continual learning on audio scene classification using representative data and memory replay GANs
Full text linkThis paper proposes a methodology aimed at resolving catastropic forgetting problem by choosing a limited portion of the historical dataset to act as a representative memory. This method harness the capabilities of generative adversarial networks (GANs) to create samples that expand upon the representative memory. The main advantage of this method is that it not only prevents catastrophic forgetting but also improves backward transfer and has a relatively stable and small size. The experimental results show that combining real representative data with artificially generated data from GANs, yielded better outcomes and helped counteract the negative effects of catastrophic forgetting more effectively than solely relying on GAN-generated data. This mixed approach creates a richer training environment, aiding in the retention of previous knowledge. Additionally, when comparing different methods for selecting data as the proportion of GAN-generated data increases, the low probability and mean cluster methods performed the best. These methods exhibit resilience and consistency by selecting more informative samples, thus improving overall performance
Identification and validation of factors affecting the success of smart village services
Full text linkThis study aims to explore and validate the factors that influence the performance and effectiveness of smart village services. Smart villages have become a focus for improving the quality of life of rural communities in the era of digital technology. However, there is a lack of methods to measure and evaluate the effectiveness of smart villages. We propose a holistic framework to measure and evaluate the effectiveness of smart services in smart villages. In this study, factors that influence the success of smart village effectiveness are identified. How effective the smart village services are can be understood using the information system success model approach by DeLone and McLean. This framework is expected to provide a better understanding of the effectiveness of smart village services so that people are willing to adopt the smart village service concept. In addition, this model can also be used as decision-making support for stakeholders and is expected to improve the quality of life of rural communities in a sustainable manner
Driving training-based optimization technique for estimating synchronous motor excitation current
Full text linkThis paper introduces an innovative application of the driving training-based optimization (DTBO) technique to optimize a multiple linear regression (MLR) model for estimating synchronous motor (SM) excitation current. Inspired by structured learning in driving training, DTBO is utilized to accurately determine regression coefficients with fast convergence. The DTBO-based MLR model is compared with other optimization techniques, such as gravitational search algorithm (GSA), artificial bee colony (ABC), genetic algorithm (GA), symbiotic organisms search (SOS), and various machine learning algorithms. Using a dataset of 557 samples (390 for training, 167 for testing), the DTBO-based model achieves the lowest objective function value, demonstrating superior performance in minimizing estimation errors. Key metrics like maximum error, error percentage, standard deviation, and root mean square error (RMSE) validate the results. The DTBO-based approach not only outperforms other methods but also provides a clear mathematical relationship between excitation current and input features, enabling easier hardware implementation and faster computation. This study establishes the DTBO-based MLR model as a robust and efficient alternative to complex machine learning algorithms for estimating SM excitation current, offering significant contributions to power systems engineering and smart grid applications
Hybrid algorithm for optimized clustering and load balancing using deep Q reccurent neural networks in cloud computing
Full text linkCloud services are among the technologies that are developing the fastest. Additionally, it is acknowledged that load balancing poses a major obstacle to reaching energy efficiency. Distributing the load among several resources in order to provide the best possible services is the main purpose of load balancing. The network's accessibility and dependability are increased through the usage of fault tolerance. An approach for hybrid deep learning (DL)-based load balancing is proposed in this paper. Tasks are first distributed in a round-robin fashion to every virtual machine. When assessing whether a virtual machine (VM) is overloaded or underloaded, the deep embedding cluster (DEC) also considers the central processing unit (CPU), bandwidth, memory, processing elements, and frequency scaling factors. For cloud load balancing, the tasks completed on the overloaded VM are assigned to the underloaded VM based on their value. To balance the load depending on many aspects like supply, demand, capacity, load, resource utilization, and fault tolerance, the deep Q recurrent neural network (DQRNN) is also suggested. Additionally, load, capacity, resource consumption, and success rate are used to evaluate the efficacy of this approach; optimum values of 0.147, 0.726, 0.527, and 0.895 are attained