Articles

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Advanced Search
Filter Icon

Filters article

Years

FAQ Arrow
Single Year
Range Year
0
0

Source Title

FAQ Arrow

Authors

FAQ Arrow

24,371 Article Results

Enhancing internet of things security against structured query language injection and brute force attacks through federated learning

10.11591/ijece.v15i1.pp1187-1199
Aigul Adamova , Tamara Zhukabayeva , Zhanna Mukanova , Zhanar Oralbekova
The internet of things (IoT) encompasses various devices for monitoring, data collection, tracking people and assets, and interacting with other gadgets without human intervention. Implementing a system for predicting the development and assessing the criticality of detected attacks is essential for ensuring security in IoT interactions. This work analyses existing methods for detecting attacks, including machine learning, deep learning, and ensemble methods, and explores the federated learning (FL) method. The aim is to study FL to enhance security, develop a methodology for predicting the development of attacks, and assess their criticality in real-time. FL enables devices and the aggregation server to jointly train a common global model while keeping the original data locally on each client. We demonstrate the performance of the proposed methodology against structured query language (SQL) injection and brute force attacks using the CICIOT2023 dataset. We used accuracy and F1 score metrics to evaluate the effectiveness of our proposed methodology. As a result, the accuracy in predicting SQL injection reached 100%, and for brute force attacks, it reached 98.25%. The high rates of experimental results clearly show that the proposed FL-based attack prediction methodology can be used to ensure security in IoT interactions.
Volume: 15
Issue: 1
Page: 1187-1199
Publish at: 2025-02-01

Hierarchical Bayesian optimization based convolutional neural network for chest X-ray disease classification

10.11591/ijece.v15i1.pp569-579
Bharath Kumar Gowru , Appa Rao Giduturi , Anuradha Sesetti
Pneumonia is an infection that affects the lungs, caused by bacteria or viruses inhaled through the air, leading to respiratory problems. The previous researches on this subject have limitations of high dimensional feature subspace and overfitting which minimize the classifier performance. In this research, hierarchical Bayesian optimization based convolutional neural network (HBO-CNN) method is proposed to effectively classify chest X-ray diseases. The proposed HBO algorithm optimizes hyperparameters of CNN which minimizes the overfitting issue and enhances the performance of classification. The hybrid Mexican axolotl optimization (MAO) and tuna swarm optimization (TSO) based feature selection method is used for selecting relevant features for classification that minimizes the high dimensional features. The ResNet 50 method is used for feature extraction to extract hierarchical features from the pre-processed images to differentiate the classes. The proposed HBO-CNN technique is estimated with performance metrics of accuracy, precision, recall, and F1-score. The proposed method attains the highest accuracy 97.95%, precision 92.00%, recall 89.00% and F1-score 92.00%, as opposed to the conventional methods, deep convolutional neural network (DCNN).
Volume: 15
Issue: 1
Page: 569-579
Publish at: 2025-02-01

ReRNet: recursive neural network for enhanced image correction in print-cam watermarking

10.11591/ijece.v15i1.pp356-364
Said Boujerfaoui , Hassan Douzi , Rachid Harba , Khadija Gourrame
Robust image watermarking that can resist camera shooting has gained considerable attention in recent years due to the need to protect sensitive printed information from being captured and reproduced without authorization. Indeed, the evolution of smartphones has made identity watermarking a feasible and convenient process. However, this process also introduces challenges like perspective distortions, which can significantly impair the effectiveness of watermark detection on freehandedly digitized images. To meet this challenge, ResNet50-based ensemble of randomized neural networks (ReRNet), a recursive convolutional neural network-based correction method, is presented for the print-cam process, specifically applied to identity images. Therefore, this paper proposes an improved Fourier watermarking method based on ReRNet to rectify perspective distortions. Experimental results validate the robustness of the enhanced scheme and demonstrate its superiority over existing methods, especially in handling perspective distortions encountered in the print-cam process.
Volume: 15
Issue: 1
Page: 356-364
Publish at: 2025-02-01

Energy analysis of active photovoltaic cooling system using water flow

10.11591/ijece.v15i1.pp1-14
Ant. Ardath Kristi , Erwin Susanto , Agus Risdiyanto , Agus Junaedi , Rudi Darussalam , Noviadi Arief Rachman , Ahmad Fudholi
An active water-cooling system is one of several technologies that has been proven to be able to reduce heat losses and increase electrical energy in photovoltaic (PV) module. This research discusses a comparative experimental study of three pump activation controls in cooling of PV module with the aim of evaluating specifically the PV output power, net energy gain, water flow rate, and module temperature reduction. The three pump activation controls being compared are continuously active during the test, active based on setpoint temperature, and active by controlling the pump voltage using pulse width modulation (PWM) control in adjusting water flow rate smoothly. The results show that controlling the pump voltage using PWM in the PV cooling process produces energy of 437.95 Wh, slightly lower than the others and the average module cooling temperature is 35.24 °C, higher of 1-3 °C than the others. Nevertheless, PWM control of cooling pump has resulted the percentage of net energy gain of 9.94%, greater than other controls, and with an average flow rate of 2.17 L/min, more efficient than the others. Thus, this control is quite effective as it can produce higher net PV energy yield and lower water consumption.
Volume: 15
Issue: 1
Page: 1-14
Publish at: 2025-02-01

Sailfish-cat algorithm-enhanced generative adversarial network for attack detection in internet of things-Fog network authentication

10.11591/ijece.v15i1.pp1109-1122
Pallavi Kanthamangala Niranjan , Ravikumar Venkatesh
The internet of things (IoT) has emerged as a prominent and influential concept within the realm of computing. Various attack detection methods are devised for detecting attacks in IoT-Fog environment. Despite all these efforts, attack detection still remained as a challenging task due to factors such as low latency, resource constraints of IoT devices, scalability issues, and distribution complexities. All these challenges are addressed in this paper by designing an efficient attack detection technique named as sailfish- cat optimization-based generative adversarial network (SaCO-based GAN) tailored for the IoT-Fog framework. This proposed approach introduces the SaCO-based GAN for IoT-Fog attack detection utilizing deep learning and feature-based classification, validated through experiments showing superior performance metrics. Notably, the SaCO optimization technique is utilized to train the GAN. Experimental results demonstrate the efficacy of the SaCO-based GAN with a maximum recall of 92.15%, a maximum precision of 91.21%, and a maximum F-Measure of 92.16%, outperforming existing techniques in IoT-Fog attack detection. The paper recommends enhancing scalability, implementing real-time detection strategies, rigorously testing robustness against diverse attack scenarios, and integrating with existing IoT security frameworks for practical deployment.
Volume: 15
Issue: 1
Page: 1109-1122
Publish at: 2025-02-01

Timed concurrent system modeling and verification of home care plan

10.11591/ijece.v15i1.pp870-882
Acep Taryana , Dieky Adzkiya , Muhammad Syifa'ul Mufid , Imam Mukhlash
A home care plan (HCP) can be integrated with an electronic medical records (EMR) system, serving as an example of a real-time system with concurrent processes. To ensure effective operation, HCPs must be free of software bugs. In this paper, we explore the modeling and verification of HCPs from the perspective of scheduling data operationalization. Specifically, we investigate how patients can obtain home services while preventing scheduling conflicts in the context of limited resources. Our goal is to develop and verify robust models for this purpose. We employ formalism to construct and validate the model, following these steps: i) develop requirements and specifications; ii) create a model with concurrent processes using timed automata; and iii) verify the model using UPPAAL tools. Our study focuses on HCP implementation at a regional general hospital in Banyumas District, Central Java, Indonesia. The results include models and specifications based on timed automata and timed computation tree logic (TCTL). We successfully verified a concurrent model that utilizes synchronized counter variables and a sender-receiver approach to analyze collision constraints arising from the synchronization of patient and resource plans.
Volume: 15
Issue: 1
Page: 870-882
Publish at: 2025-02-01

Deep learning for infectious disease surveillance integrating internet of things for rapid response

10.11591/ijece.v15i1.pp1175-1186
Subramanian Sumithra , Moorthy Radhika , Gandavadi Venkatesh , Babu Seetha Lakshmi , Balraj Victoria Jancee , Nagarajan Mohankumar , Subbiah Murugan
Particularly in the case of emerging infectious diseases and worldwide pandemics, infectious disease monitoring is essential for quick identification and efficient response to epidemics. Improving surveillance systems for quick reaction might be possible with the help of new deep learning and internet of things (IoT) technologies. This paper introduces an infectious disease monitoring architecture based on deep learning coupled with IoT devices to facilitate early diagnosis and proactive intervention measures. This approach uses recurrent neural networks (RNNs) to identify temporal patterns suggestive of infectious disease outbreaks by analyzing sequential data retrieved from IoT devices like smart thermometers and wearable sensors. To identify small changes in health markers and forecast the development of diseases, RNN architectures with long short-term memory (LSTM) networks are used to capture long-range relationships in the data. Spatial analysis permits the integration of geographic data from IoT devices, allowing for the identification of infection hotspots and the tracking of afflicted persons' movements. Quick action steps like focused testing, contact tracing, and medical resource deployment are prompted by abnormalities detected early by real-time monitoring and analysis. Preventing or lessening the severity of infectious disease outbreaks is the goal of the planned monitoring system, which would enhance public health readiness and response capacities.
Volume: 15
Issue: 1
Page: 1175-1186
Publish at: 2025-02-01

Secured and cloud-based electronic health records by homomorphic encryption algorithm

10.11591/ijece.v15i1.pp1152-1161
Bala Annapurna , Gaddam Geetha , Priyanka Madhiraju , Subbarayan Kalaiselvi , Mishmala Sushith , Rathinasabapathy Ramadevi , Pramod Pandey
This uses homomorphic encryption in cloud-based platforms to improve electronic health records (EHR) security and accessibility. Protecting sensitive medical data while enabling data processing and analysis is the main goal. The study examines how homomorphic encryption protects EHR data privacy and integrity. Its main purpose is to reduce risks of unauthorized access and data breaches to build trust between healthcare professionals and patients in digital healthcare. The research uses homomorphic encryption to safeguard cloud EHR storage and transmission. Results will highlight the algorithm's influence on data security and computing efficiency, revealing its potential use in healthcare to protect patient privacy and meet regulatory requirements. Results from dataset of patient health metrics show in the 1st instance sample data for 5 instances with ages between 57 to 88, blood pressure (BP) values from 33 to 85, glucose values from 5 to 99, and heart rate values from 24 to 88. In another study of 5 patients, cholesterol levels ranged from 10 to 80 mg/dL, body mass index (BMI) from 10 to 96 kg/m², smoking status from 14 to 79, and medication adherence from 6 to 78%.
Volume: 15
Issue: 1
Page: 1152-1161
Publish at: 2025-02-01

Measuring anxiety level on phobia using electrodermal activity, electrocardiogram and respiratory signals

10.11591/ijece.v15i1.pp337-348
Khusnul Ain , Osmalina Nur Rahma , Endah Purwanti , Richa Varyan , Sayyidul Istighfar Ittaqilah , Danny Sanjaya Arfensia , Tiara Dyah Sosialita , Fitriyatul Qulub , Rifai Chai
People with spider phobia experience excessive anxiety reactions when exposed to spiders that will interfere with daily life. Diagnosing and measuring anxiety levels in patients with spider phobia is a complex challenge. Conventional diagnosis requires psychological evaluations and clinical interviews that take time and often result in a high degree of subjectivity. Therefore, there is a need for a more objective and efficient approach to measuring anxiety levels in patients. This study performs anxiety level classification based on electrodermal activity, electrocardiogram (ECG) and respiratory signals using the dataset of Arachnophobia subjects. Each raw data is preprocessed using 24 types of features. Feature performance is processed using the recursive feature elimination method. Data processing was performed in 3 anxiety levels (high, medium, low) and two anxiety levels (high, low) with the support vector machine method and hold-out validation method (7:3). The performance of the model is evaluated by showing the accuracy, precision, recall and F1 score values. The polynomial kernel can perform optimal classification and obtain 100% accuracy in 2 classes and three classes with 100% precision, recall, and F1 score values. This result shows excellent potential in measuring anxiety levels that correlate with mental health issues.
Volume: 15
Issue: 1
Page: 337-348
Publish at: 2025-02-01

Method of undetermined coefficients for circuits and filters using Legendre functions

10.11591/ijece.v15i1.pp846-854
Zhanat Manbetova , Pavel Dunayev , Assel Yerzhan , Manat Imankul , Zhazira Zhazykbayeva , Zhadra Seitova , Raushan Dzhanuzakova , Gayni Karnakova
This article presents a new way to implement matching networks and filters using the method of undetermined coefficients. A method is proposed for approximating the transmission coefficient of the synthesized filter, taking into account the required amplitude-frequency characteristics. To synthesize the filter, an approximating function (AF) was used using orthogonal Legendre polynomials, which is a mathematical description using a system of equations. Filter properties whose implementation is based on modified Legendre approximating functions usually depend on the interval on which they are defined and have the property that they are orthogonal on this interval. An example of seventh order filter synthesis using modified Legendre approximating functions is given. The filter circuit is implemented, the elements of the filter circuit are calculated based on the selected approximating modified function. The criteria used were minimization of the unevenness of the group delay time (GDT) and minimization of the complex approximation error for given values of the AF parameters. As a result, the number of filter elements, the group delay value and the complex approximation error are significantly reduced.
Volume: 15
Issue: 1
Page: 846-854
Publish at: 2025-02-01

Determination of biomass energy potential based on regional characteristics using adaptive clustering method

10.11591/ijece.v15i1.pp46-55
Ginas Alvianingsih , Haslenda Hashim , Jasrul Jamani Jamian , Adri Senen
Determining the energy potential of biomass is the first step in selecting the most suitable and efficient energy conversion technology based on regional characteristics. The approach to estimating and determining biomass potential generally uses geospatial technology related to collecting and processing data about mapping an area. Unfortunately, this method is inadequate for simulating the interaction between variables, nor can it provide accurate predictions for the biomass supply chain. As a result, the results obtained from this method tend to be biased and macro, particularly in regions experiencing rapid land-use development. In this paper, the author has developed a clustering methodology with a fuzzy c-means (FCM) algorithm to determine biomass energy potential based on regional characteristics to produce data clusters with high accuracy. Grouping the characteristics of clustering-based areas involves grouping physical or abstract objects into classes or similar objects.
Volume: 15
Issue: 1
Page: 46-55
Publish at: 2025-02-01

Intrusion detection and prevention using Bayesian decision with fuzzy logic system

10.11591/ijece.v15i1.pp1200-1208
Satheeshkumar Sekar , Palaniraj Rajidurai Parvathy , Gopal Kumar Gupta , Thiruvenkadachari Rajagopalan , Chethan Chandra Subhash Chandra Basappa Basavaraddi , Kuppan Padmanaban , Subbiah Murugan
Nowadays, intrusion detection and prevention method has comprehended the notice to decrease the effect of intruders. denial of service (DoS) is an attack that formulates malicious traffic is distributed into an exacting network device. These attackers absorb with a valid network device, the valid device will be compromised to insert malicious traffic. To solve these problems, the Bayesian decision model with a fuzzy logic system based on intrusion detection and prevention (BDFL) is introduced. This mechanism separates the DoS packets based on the type of validation, such as packet and flow validation. The BDFL mechanism uses a fuzzy logic system (FLS) for validating the data packets. Also, the key features of the algorithm are excerpted from data packets and categorized into normal, doubtful, and malicious. Furthermore, the Bayesian decision (BD) decide two queues as malicious and normal. The BDFL mechanism is experimental in a network simulator environment, and the operations are measures regarding DoS attacker detection ratio, delay, traffic load, and throughput.
Volume: 15
Issue: 1
Page: 1200-1208
Publish at: 2025-02-01

Influence of metal particles shape on direct current voltage electric properties of nanofluids

10.11591/ijece.v15i1.pp56-66
Daniar Fahmi , Muhammad Fadlan Akbar , I Made Yulistya Negara , I Gusti Ngurah Satriyadi Hernanda , Dimas Anton Asfani , Risyad Alauddin Zaidan , Arkan Fadhilah
It is widely recognized that the application of nanoparticles has the potential to improve the dielectric properties of transformer oil. Nevertheless, there is a scarcity of studies that have utilized pure nanofluids, and in practical applications, it is inevitable for transformer oil to become contaminated. Therefore, this study conducted tests to investigate how the shape and size of metal contaminants impact the dielectric performance of Fe3O4 nanofluids. The findings from the levitation voltage test indicate that as the size and diameter of the particle increase, the levitation voltage value measured also increases, and conversely. Moreover, a higher concentration of nanoparticles leads to a higher measured levitation voltage value. On the other hand, the breakdown voltage test results demonstrate that larger and sharper particles result in lower measured breakdown voltage values, and vice versa. The simulation outcomes regarding electric field distribution reveal that larger and sharper particles correspond to higher measured electric field values, while the opposite is true for smaller and less sharp particles.
Volume: 15
Issue: 1
Page: 56-66
Publish at: 2025-02-01

Utilization meta-analysis to identify the convenience of eBooks (visual and audio) for learning

10.11591/ijece.v15i1.pp529-539
Jefri Mailool , Janu Arlinwibowo , Yulia Linguistika
This research aims to conclude the influence of eBooks in the learning process throughout the world. The meta-analysis design taken was a group contrast between control and experimental groups with a random effect size model. The criteria used are time “data published 2018–2023,” published in English, type of publication is a quantitative research article, the research design is a difference between control and experimental groups, containing complete data “mean, sample size, and standard deviation,” and recorded in the Scopus database. Data collection was guided by the PRISMA method. The results of the analysis showed that the data were heterogeneous and free from publication bias. The results of the analysis showed that there was a large “positive” effect as indicated by a p-value <0.001<5% “95% confidence interval” and a total effect size=0.86 [0.61; 1.11]. It can be concluded based on the latest findings that eBooks have an equally good effect on all conditions which are influenced by the type of competency developed, the eBook information base, the type of eBook, and class size.
Volume: 15
Issue: 1
Page: 529-539
Publish at: 2025-02-01

Q-learning based forecasting early landslide detection in internet of thing wireless sensor network

10.11591/ijece.v15i1.pp425-434
Devasahayam Joseph Jeyakumar , Boominathan Shanmathi , Parappurathu Bahulayan Smitha , Shalini Chowdary , Thamizharasan Panneerselvam , Rajagopalan Srinath , Muthuraj Mariselvam , Mohanan Murali
The issue of climate modification and human actions terminates in a chain of hazardous developments, comprehensive of landslides. The traditional approaches of observing the environmental attributes that is actually obtaining rainfall data from places can be cruel and suppressing supervising necessitated for careful infliction. Thus, landslide forecasting and early notice is a significant application via wireless sensor networks (WSN) to reduce loss of life and property. Because of the heavy preparation of sensors in landslide prostrate regions, clustering is a resourceful method to minimize unnecessary transmission. In this article we introduce Q-learning based forecasting early landslide detection (Q-LFD) in internet of things (IoT) WSN. The Q-LFD mechanism utilizes a dingo optimization algorithm (DOA) to choose the best cluster head (CH). Furthermore, the Q-learning algorithm forecast the landslide by soil water capacity, soil layer, soil temperature, Seismic vibrations, and rainfall. Experimental results illustrate the Q-LFD mechanism raises the landslide detection accuracy. In addition, it minimizes the false positive, false negative ratio.
Volume: 15
Issue: 1
Page: 425-434
Publish at: 2025-02-01
1 2 3
...
1624 1625
Show 3 of 1625

Discover Our Library

Embark on a journey through our expansive collection of articles and let curiosity lead your path to innovation.

Explore Now
Library 3D Ilustration