Internet of Things and Artificial Intelligence Journal

Smart Room Based on ESP32 and Google Assistant: Automation Solution in Electrical Engineering Laboratory Environment

Wahyu Syamsuri — 2025-05-28

The development of home automation technology and smart rooms is growing rapidly, especially through the utilization of voice commands to control electronic devices. This research develops a voice command-based smart room system using an ESP32 module connected with Google Assistant and the IFTTT platform for device control in an electrical engineering laboratory. The system allows users to turn on or off devices such as lights and TVs with just a voice, improving the efficiency and convenience of laboratory operations. The implementation using ESP32 as a Wi-Fi microcontroller, which activates relays to control devices, was tested in real laboratory conditions that have challenges in the form of electromagnetic interference and noise. The test results showed a fast and accurate system response of up to 90%, although there were some failures in high noise conditions. The system has the potential to be an effective automation solution to support learning and practical processes in electrical engineering laboratories, with further development needed to improve the system's robustness against noise interference.

IoT-Based Real-Time Train Position Monitoring System Using GPS

Andriana Andriana — 2025-06-08

In this modern era, the need for fast, efficient, and quality transportation is important, one of which is in the railway sector. Although trains have special tracks, accidents can still occur due to various factors, such as infrastructure damage and officer negligence. One of the accidents that occurred on January 5, 2024, between the Turangga train and the Bandung Commuter Line caused fatalities and injuries. To overcome this problem, a real-time train position monitoring system is needed. This research aims to design and develop a train position monitoring system using a GPS module. GPS (Global Positioning System) is widely used for tracking systems. This system monitors the position of trains and other information that can be accessed by station officers and the general public. By using the ESP32 microcontroller and NEO-6M GPS module, this system is expected to improve the safety of train travel, minimize the potential for accidents, and provide more accurate information about train arrival times at the station. This research also aims to test the performance of the monitoring system in real conditions. It is hoped that with this system, train accidents due to negligence can be minimized, and provide significant benefits for officers and the public who use train transportation.

Artificial-Based Pulse Learning Web Application Intelligence Using Convolutional Neural Networks

Chayadi Oktomy Susanto — 2025-05-11

Studying pulse waveforms in healthcare is crucial as they aid in diagnosing and treating chronic diseases. However, the limited data on pulse waveforms makes it challenging for health education to teach this topic effectively. Practitioners of Traditional Chinese Medicine (TCM) require a significant amount of time to obtain pulse wave data accurately. Additionally, the pulse wave data collected by TCM practitioners exhibit various forms and characteristics. This study aims to integrate web-based pulse waveform learning with Artificial Intelligence (AI) using Convolutional Neural Network (CNN) to enhance effectiveness and efficiency. Pulse waveform data were obtained from Traditional Chinese Pulse Diagnosis and were redrawn to achieve diverse and accurate results. A total of 400 images were generated for each of the five types of pulse waveforms to improve data quality. The redrawn data were then tested to ensure accuracy. Once validated, a comparison of deep learning models using three CNN architecturesâ€”VGG16, VGG19, and ResNet50â€”was conducted, with VGG19 achieving the highest accuracy among the models. Consequently, the VGG19 model was implemented into a web-based pulse waveform learning application using JavaScript, HTML5, and TensorFlow. The results demonstrate that the VGG19 model outperformed other architectures in terms of accuracy. The successful integration of the VGG19 model into the web-based application shows that AI can be used to create an interactive learning platform for pulse waveform education. This study proves that the collaboration between web-based pulse waveform learning and AI can serve as an interactive educational tool for the future.

Design and Implementation of a Real-Time Monitoring System for a 150 kV Substation with Multi-Platform Notification and Visualization

Eka Anggara Yuda Kartika — 2025-05-27

This paper presents the development and implementation of an innovative real-time monitoring and notification system for a 150 kV electrical substation, leveraging Raspberry Pi 3, Node-RED, MySQL, and Firebase. The system measures key electrical parameters such as voltage, current, power, and frequency using sensors connected to a Programmable Logic Controller (PLC). The data is processed and displayed through a single-line diagram on both a web-based dashboard and an Android application. Color-coded indicators, controlled by JavaScript, reflect real-time equipment status, with normal conditions marked in red and fault conditions indicated in black. The novelty of this system lies in its integration of real-time data processing, dynamic visualization, and multi-channel notification mechanisms, combining web, mobile app, and messaging services like WhatsApp and email for operator alerts. This multi-layered approach improves operator response time and enhances monitoring accuracy, especially in remote or field environments. Experimental tests, including high-voltage and low-voltage fault simulations, demonstrated the systemâ€™s ability to accurately detect faults and communicate them through the notifications in real-time, with an average measurement error of just 1.56%. The system not only provides enhanced situational awareness but also offers an efficient, cost-effective solution for remote substation monitoring, ensuring continuous supervision and immediate response to power system anomalies.

The Fabrication of Electronics Wearable Bracelets and approach to telecommunications: A Review

Tjahjo Adiprabowo — 2025-06-01

This research aims to develop an innovative and easy-to-use fetal heart rate monitoring bracelet for pregnant women. The design of this bracelet involves various electronic aspects, including signal acquisition using optical sensors (Ultrasound Doppler, ECG, or PPG), signal processing with a microcontroller and efficient algorithms to eliminate noise and detect heart rate, as well as data display and communication via LCD screen, LED indicator, or Bluetooth connection. The power aspect is an important consideration in the selection of the right battery and power management. The design of the bracelet considers ergonomics and safety for pregnant women and fetuses. Testing and validation are carried out thoroughly, including functionality testing and clinical trials. This research also considers ethical aspects, such as data privacy and information reliability. By integrating all these aspects, it is hoped that this research can produce a prototype fetal heart rate monitoring bracelet that is accurate, safe, and comfortable to use, thus contributing to improving maternal and child health. Wearable Bracelet must-have comfort factors components consisting of Size & Dimensions, Device Weight, Strap Material, Thickness, Flexibility, Component Distribution, Edges & Corners, and Skin Ventilation. From the intructables' sample results the comfort level is 52.5%, while the tool created by Puput Dani Prasetyo has a comfort level of only 50%. And this needs to be increased to >90%.

Controlling a Delivery Robot Based on Color Using Fuzzy Logic Method

Hidayat Sapriyanto — 2025-06-04

This research presents the design and implementation of an IoT-based delivery robot equipped with a line navigation system and color classification using Mamdani fuzzy logic. The robot follows a black line using infrared sensors and identifies delivery zones via color markers (red, green, blue, yellow) detected by a TCS3200 sensor. Fuzzy logic processes RGB input to classify colors and trigger item placement using a servo mechanism. The system integrates hardware components such as Arduino Mega, ESP8266, and L298N motor drivers, and utilizes the Blynk application for remote control. Navigation decisions are enhanced through fuzzy inference, enabling adaptation to sensor uncertainty and lighting variations. Experimental results show a 94% success rate in path following, 92% color classification accuracy, and sub-2-second response time for remote commands. The combination of fuzzy logic and IoT enables flexible, real-time control, making the system suitable for dynamic indoor environments like offices or labs.

Decision Support System for Selection of High-Aching Students Using Simple Additive Weighting Method

Karim Thamrin — 2025-05-14

Determination of outstanding students is done manually and tends to be subjective. This has shortcomings that cause several problems, namely, it takes a long time and requires energy for manual data processing, then requires extra care and thought in processing student value data. This study aims to produce a decision support system for selecting outstanding students with a matrix normalization process to a scale that can be compared with all existing alternative ratings using ISO 25010 with 8 testing aspects. This development uses the Waterfall model with research steps, namely: 1) Requirement, 2) Design, 3) Implementation, 4) Verification, and 5) Maintenance. In the early stages, researchers seek information related to user needs for the system to be created. Furthermore, the media design is designed based on the results of the needs analysis, after which the media is developed. The final stage is testing. Data collection instruments are conducted through Observation, interviews, and questionnaires. The data analysis technique used is based on the 8 testing characteristics in ISO 25010. The results of the functional suitability aspect test with a result of 98.6% of the criteria "Very Eligible". Performance efficiency testing with a result of 85.2% of the category grade B which is categorized as "Good". Usability testing of the percentage of user responses with a result of 84% of the criteria "Very Good". The results of the compatibility test show that the system has been successfully run on several versions of the browser well and does not cause problems when accessing the system. The results of the security test show that the system has good security with a grade of B. Based on these data, it can be concluded that the system that has been developed has successfully met 8 aspects of testing using ISO 25010.

Geographic Information System for Drought Potential Areas in Kampar District

Nindy Raisa Hanum — 2025-06-10

Drought is a natural disaster that has an impact on various sectors. Drought occurs during the dry season, which is from May to October. So that there is a need for spatial information in the Kampar Regency area for drought potential so that it can help overcome the problem of drought. The data consists of Landsat 8 OLI / TRIS images and rainfall data. The transformation method used is the Tasseled Cap Transformation to get the wetness and revision index and NDVI to get the vegetation index. The results of each index will be weighted and overlaid so that a map of the area with drought potential is obtained. The results of this study are Kampar District which has an area of 7% located in Tapung Hilir District, while 33% of Kampar District has a very low drought potential. In general, the Kampar Regency is classified as having moderate drought potential.

Image quality enhancement by applying a combination of filtering between the median filter and CLAHE

Adinda Tarisyah Hsb — 2025-05-31

Digital image quality is often degraded due to disturbances such as salt-and-pepper noise and uneven lighting. This can hinder further image analysis and processing. This research aims to improve image quality by applying a combination of two filtering methods, namely Median Filter and CLAHE (Contrast Limited Adaptive Histogram Equalization), using MATLAB. Median Filter is used to remove impulsive noise without obscuring important details, while CLAHE is applied to improve image contrast adaptively and locally. Tests were conducted on grayscale images with artificial noise added. The experimental results show that the combination of both methods provides a significant improvement in image quality compared to the use of either method alone. Thus, this approach is effectively used for image pre-processing that requires detail recovery and contrast enhancement.

Analysis of Teaching Material Delivery Techniques by Lecturers Based on E-Learning at Universitas Pembangunan Panca Budi Medan

Fauziah Fauziah — 2025-05-25

The rapid development of technology has also affected the world of education, both in terms of facilities, infrastructure, and learning methods. Schools and universities now utilize technology to support the teaching and learning process, making it more efficient, interesting, and accessible. Technology acts as a means, media, and learning resources that help facilitate learning between educators and students. One form of technology application in education is the use of videos and E-Learning-based learning portals. At Universitas Pembangunan Panca Budi, the E-Learning system continues to be developed in terms of features, security, and interface to support effective distance learning between students and lecturers.

Implementation of One-Time Password and SHA-3 Algorithm on the Lab Inventory Website of the Department of Informatics and Computer Engineering

Muhammad Hakim Sadiq — 2025-05-11

This study evaluates the effectiveness of the One-Time Password (OTP) system on the Inventory Lab website of the Department of Informatics and Computer Engineering, focusing on OTP and user password security against Brute Force attacks. The objectives include testing OTP validation, analyzing OTP vulnerabilities to Brute Force attacks, and examining the resilience of user passwords under similar attacks. The study contributes to cyber security research by offering insights into implementing OTP and SHA-3 encryption algorithms on websites. Its findings aim to enhance the security measures of the Inventory Lab website. Results indicate that OTP delivery on the website is both successful and secure, with codes encrypted using SHA-3, rendering them unreadable in the database. OTP validation effectively distinguished correct and incorrect codes, including those that expired due to time limits. However, Brute Force trials on OTPs succeeded in some cases due to extended expiration times. Reducing the expiration period to one minute significantly minimized this risk. Similarly, trials on user passwords showed that passwords with complex character combinations resisted attacks more effectively than simpler ones. In summary, the OTP system and SHA-3 encrypted passwords demonstrate robust security but require adjustments to OTP expiration settings and stronger password policies to mitigate the risks of brute-force attacks. These improvements will further safeguard the websiteâ€™s security infrastructure.

Implementation of the MDLC Method in the Development of Android-Based Augmented Reality for Traditional House Recognition

Andri Nofiar Am — 2025-05-07

Indonesia has a rich cultural diversity, one of which is traditional houses (rumah adat). However, learning about traditional houses is still largely based on textbooks. With the advancement of information and communication technology, teaching and learning processes can now be supported by interactive learning media. One of the emerging technologies that is widely used in education is Augmented Reality (AR). AR can display 3D animations from all angles, helping to increase public interest in learning more about Indonesiaâ€™s diverse traditional houses. The purpose of this research is to develop an Android-based Augmented Reality application for introducing traditional houses in Indonesia as a learning medium. This application aims to provide users convenience in learning, and seeing various types of houses there. The method used is the Multimedia Development Life Cycle (MDLC) which consists of a concept, design, collection of materials, assembly process, testing process, and distribution process. The application uses Unity, Blender, and Vuforia SDK. The results of this study will show that the Android-based Augmented Reality (AR) marker application, specifically the traditional houses in Indonesia. 3D assets of traditional houses can be adjusted flexibly, for example, rotated by 360 degrees, accompanied by descriptive information on each traditional house, so that information can be received flexibly and real-like.

Implementation of Synchronous Messaging with Visual Basic as a Support for Learning the Parallel and Distributed Systems Course

Muhammad Romi Nasution — 2025-06-11

There are two types of message delivery, namely synchronous and asynchronous messages. Synchronous communication refers to communication that occurs directly and simultaneously, where both parties interact in real-time. Common examples of synchronous communication in online learning include video conferencing, web conferencing, and live chat. Asynchronous communication, which is indirect, allows for more flexible learning through tools such as discussion forums, emails, and blogs. In this article, a chatting application will be designed and built as part of the implementation of the parallel and distributed systems course to support student learning. The application was built using Visual Basic software with important supporting components such as Winsock, TCP/IP protocol, and Local Port, to communicate directly (synchronously) between the client and server. From the test results of conducting chatting between the server and client, it can be concluded that the application runs well as expected without any errors, with an error rate of 0% and a success accuracy rate of 100%. For future researchers, this application can be used as teaching material in the parallel and distributed systems course, and can be developed with other models and the latest software, for example, not only sending messages but also sending files such as images and the like. This research approach can look at multiple sides, for example, from AI-Enhanced Communication, Hybrid Legacy-Modern Integration, Adaptive Protocol Switching, Dynamic Key Exchange, Auto-Discovery Port Management, ML-Optimized UI Components, Context-Aware UX, Predictive Load Balancing, Behavioral Pattern Analysis, AI-Assisted Bug Detection, QoE Assessment Model, dan Open Source Framework.

Digitization of Student Activity Assessment and Reporting at Kampar Polytechnic

Fenty Kurnia Oktorina — 2025-05-07

Students in higher education institutions are expected to actively participate in various academic and non-academic activities to develop their potential. At Politeknik Kampar, student activity records are still managed manually by the Student Affairs Administration Division (BAK), which leads to difficulties in obtaining valid and organized data. This research aims to build a Digital Assessment system of student activities at Kampar Polytechnic whose purpose is to facilitate the process of managing student participation data. Especially evaluation and reporting. The method used is the development of the System Development Life Cycle (SDLC). The programming language used in this research is PHP with the Laravel Framework, this framework is known for its good data management and accuracy and has structured data management. The result of this research is the development of an application that will focus on the process of reporting activities carried out by students directly, in real-time, and can be monitored and validated directly. With this system, it is ensured that the management of student activities can be more organized, transparent, and accessible to students, BAK, and all related parties.

Architectural Design of Public Service Application Mojorejo 01 Elementary School, Madiun City

Hani Atun Mumtahana — 2025-05-31

The delivery of information to the general public is one of the business processes that need to be developed at Mojorejo 01 Elementary School, Madiun City. The school profile website can be used as a medium for distributing school information, so that the general public, guardians, students, and teachers can obtain information on time. In addition, information about guidance and counseling from student achievements needs to be published to increase the school's excellence. In this study, an architectural design was designed for the development of a school profile website and e-counseling application to support the business processes at Mojorejo 01 Elementary School, Madiun City. The development of architectural design includes several stages of business process architecture design, data and information architecture, application architecture, and technology architecture. The results of the study are in the form of architectural design and recommendations for developing school web applications and e-counseling that can be used by schools and developers in developing applications. To assess the suitability of the architectural design to the needs of the business process and business processes of Mojorejo 01 Elementary School, an evaluation has been carried out with a result of 83.8%, which shows that the architectural design can be applied in application development.

Intelconn Smart Connection System to Improve Battery Energy Storage Reliability

Sutisna Abdul Rahman — 2025-05-31

Degradation in battery cells is an unavoidable phenomenon, especially in battery modules connected in parallel. Inconsistencies in current distribution and temperature rise can accelerate degradation and pose potential safety issues. Therefore, evaluation and mitigation strategies are needed to reduce the impact of degradation on each battery cell. This research develops an extended single-particle model with Pade's approach and Taylor expansion to simplify the conventional electrochemical mechanism. Based on this approach, a multidomain electrochemical mechanism simulation model for battery modules connected in parallel is developed. The model was used to analyze the effect of cell degradation on battery module voltage, internal current distribution, and temperature rise under various charging and discharging conditions. The evaluation was conducted using the parameter sensitivity analysis method to assess the contribution of each parameter to the degradation of the parallel-connected battery cells. The results of this study are expected to help in identifying battery cells that degrade faster and developing optimal strategies in battery management for longer life and safe use.

Content Blocking Method To Reduce False Positives Based On Machine Learning

Andi Iksan Arkam — 2025-05-28

This study presents an experimental approach to enhance content-blocking systems by integrating machine learning with domain classification and Pi-hole DNS server technology. While traditional blocking mechanisms often result in false positivesâ€”legitimate domains mistakenly blockedâ€”this research aims to mitigate such issues. By implementing various testing scenarios, including TF-IDF and N-gram feature extraction with and without preprocessing, the study evaluates the classification performance using the Naive Bayes algorithm. The results reveal the highest accuracy of 84% achieved with the N-gram method without preprocessing. This integrated approach shows promise in improving the precision of ad and website blocking mechanisms.

Design and Development of Information System Security with Authentication Using One-Time Password Identification Based on SMS with MD5 Hash

Irzon Meiditra — 2025-05-29

Login security to access WEB-based applications, in the form of security using OTP (One Time Password) which is generated using Hash MD5 and generates a code sent via SMS. The system will take the email field, password, and phone number. The result of the hash function will produce a 32-digit hexadecimal number. Furthermore, four digits of the hexadecimal number are taken. The four numbers are sent as OTP with Zenziva's Cloud SMS Gateway service and the OTP code will be temporarily stored in the database. The OTP sent to the user will be matched with the one stored in the database table to check its validity. If the OTP sent with the one stored in the table matches, then the user can access the WEB-based application. The OTP generated is for security authentication of the WEB user account after logging in by entering the username and password. Users who enter the wrong OTP 3 times will be blocked, the restriction is to narrow the hackers to intercept and infiltrate.

Statistical Data-Driven Decision-Making Considering Bias, Fairness, and Transparency in AI

T. G. Vasista — 2025-05-11

Bias, fairness, and transparency are critical issues in Artificial Intelligence (AI). These problems can arise from sources such as biased training data, algorithmic bias, and reinforcement learning bias. Bias may lead to unintended consequences while attempting to correct bias. The use of the black-box model, along with proprietary and confidentiality constraints, can further obscure decision-making processes. Regulatory challenges complicate the governance of AI systems. Unfairness can arise when the algorithm uses inappropriate features in AI-based decision-making. Lack of transparency in AI-based computation leads to reduced trust, accountability issues, and difficulty in understanding or challenging automated decisions. Addressing bias, fairness, and transparency in AI is crucial to ensure ethical, responsible, and inclusive technology. Governments, organizations, and researchers must work together to create AI systems that serve humanity without reinforcing discrimination. Without addressing these problems, AI will have to risk inequalities and lose public trust. For example, â€œif you tell an AI image tool to create a man writing with his LEFT hand, the AI will create a man writing with his right handâ€ Indiaâ€™s PM Modiji pointed it out in a Paris speech. Unfairness can arise when the algorithm uses inappropriate features or a biased training data set to make a decision.

The Fabrique: A Pathfinding Algorithm in a Mobile Game Developed Using Construct 3*

Ruby Kusumawardhani — 2025-05-12

The rapid growth of the digital game industry, particularly on mobile platforms, has driven the development of algorithms to enhance gameplay quality and player experience. The A* algorithm is a widely used pathfinding method for controlling the movement of non-playable characters (NPCs) in games. This study aims to evaluate the implementation of the A* algorithm in The Fabrique, a mobile game developed using Construct 3, a 2D game development engine. Testing was conducted across various path and obstacle scenarios. The results indicate that the A* algorithm delivers fast computation time and optimal pathfinding for short-distance navigation. In the medium to high obstacle scenarios, the algorithm maintained good performance with only minimal increases in processing time. The implementation of the A* algorithm in The Fabrique proved effective, contributing to a more dynamic and interactive gameplay experience. With an average user satisfaction rate of 81.94%, the algorithm demonstrates not only technical efficiency but also strong user acceptance.

System for Determining Plant Types Based on Weather Characteristics and Soil pH Using Artificial Intelligence

Trisakti Akbar — 2025-05-09

This research implements the Long Short-Term Memory (LSTM) algorithm for weather forecasting using minimum temperature, maximum temperature, average temperature, air humidity, rainfall, and solar radiation values over the past 30 days. The output consists of forecasts for average temperature, air humidity, rainfall, and solar radiation for the next 30 days. The LSTM model output and soil pH are used to determine plant types using the K-Nearest Neighbor (K-NN) algorithm. Based on the LSTM model testing results, the minimum temperature feature achieved a Mean Absolute Error (MAE) of 0.0078, a maximum temperature of 0.0054, an average temperature of 0.009, air humidity of 0.0099, rainfall of 0.0042, and solar radiation of 0.0208. For the K-NN model, an accuracy of 98% was obtained.

Sentiment analysis of Faculty of Science and Technology students' satisfaction with the 2024 graduation using the Naïve Bayes method

Kalfida Eka Wati Siregar — 2025-05-30

Sentiment analysis of UINSU student graduation based on academic data is one of the efforts to understand the factors that affect the success of student studies. This research aims to analyze the sentiment of UINSU student graduation by utilizing academic data such as cumulative grade point average (GPA), number of credits taken, and other relevant attributes, using the Naive Bayes method. Naive Bayes was chosen because of its ability to classify data efficiently and accurately, even though the data used has noise or inconsistency. The research process begins with collecting student data from the university database, and then data cleaning is carried out to ensure the quality of the data used. Next, the data is processed and classified using the Naive Bayes algorithm in Weka software to predict graduation status based on academic parameters. The results show that the Naive Bayes method is able to produce quite high accuracy in predicting student graduation, with accuracy values ranging from 75% to more than 85% depending on parameter selection and data cleaning. GPA is the most influential attribute on the prediction results, while other attributes such as class activity and organizational experience also contribute, although not as much as GPA. These findings provide important insights for the campus in designing more effective academic coaching and planning programs and can be a reference in the development of data mining-based decision support systems to improve the quality of computer science graduates.