Jurnal Teknik dan Sistem Komputer

Pengembangan Alat Penyiraman Otomatis Melalui Aplikasi Hydropro untuk Budidaya Tanaman Tomat

Tuada Rahadatul Aisy — 2026-06-26

Tomato plants are known as hotricultural plants which are often cultivated, because they contain rich nutrients such as vitamins, minerals, carbohydrates, proteins and fats, which makes tomato plants have high value and benefits for meeting daily needs. In cultivation, various conditions such as climate, temperature and humidity affect growth. Not only that, tomatoes also require special treatment such as watering, soil conditions and vitamins. Manual watering is often inefficient because it requires manual measurement of the air before it is applied to each plant. This becomes increasingly difficult if the tomato cultivation location is far from the user's reach. Therefore, an IoT was designed from automatic watering using an air reservoir that can be used to water plants in close or outdoor conditions. This system is a watering tool that is integrated with a mobile application using the Kotlin programming language with Firebase as data storage. Through the analysis carried out, a tool was created using a pH sensor, TDS sensor, and DHT-22 sensor and NodeMCU ESP32, which was then designed and tested before finally being implemented.

Implementasi Smart Watering Berbasis IoT untuk Penyiraman Anggrek Menggunakan NodeMCU ESP8266 dan Telegram

Ernando Rizki Dalimunthe — 2026-06-26

Orchid cultivation requires proper watering management because insufficient or excessive water can reduce plant quality and damage the growing medium. Manual watering is also less effective when plant owners cannot monitor soil moisture regularly. This study evaluates an Internet of Things-based automatic watering system using NodeMCU ESP8266, a capacitive soil moisture sensor, a relay module, a water pump, a 16x2 LCD, and Telegram notification. The research was conducted using a prototype method consisting of communication, quick planning, quick design, prototype construction, and user feedback. System testing was performed on the sensor, pump, LCD, relay, Telegram notification, and the integrated device. The results show that the system can automatically activate the pump when soil moisture is in the dry range of 0%-30% and deactivate the pump when the moisture is above 31%. Telegram notification testing produced a 100% successful delivery rate with an average delivery time of 30 seconds. The pump operated at an average voltage of 11.95 V and an average current of 0.95 A. These results indicate that the proposed system can support orchid watering automation and remote monitoring for small-scale ornamental plant cultivation.

Rekayasa Priority Notification System Menggunakan Model Berbasis Aturan untuk Mendukung Manajemen Waktu Mahasiswa Aktif

Widyashari Nur Arafah — 2026-06-26

Active university students frequently struggle with time management and academic procrastination, worsened by notification fatigue from undifferentiated task reminder systems. This study engineers a Priority Notification System using a rule-based model, implemented in an Android task scheduling application called SchedU. The theoretical framework draws on Self-Regulated Learning, the Eisenhower Matrix, and rule-based systems, with a Research and Development (R&D) approach using the Agile methodology. Data were collected through needs analysis questionnaires, Black Box Testing, and application log recording. The rule-based model classifies tasks into four priority quadrants based on importance and urgency parameters adapted from the Eisenhower Matrix, with the Priority Notification System applying differentiated notification behaviors including persistent notifications for high-priority tasks. Black Box Testing confirmed all core features functioned as designed. Log analysis of 50 users showed an average On-Time Completion Rate of 84.80%, demonstrating SchedU's effectiveness in supporting structured academic time management.

Tantangan dan Tren Masa Depan Keamanan Federated Learning pada Ekosistem IoT Tinjau Pustaka Sistematis

Ardiansyah Hasan — 2026-06-29

The rapid growth of the Internet of Things (IoT) ecosystem has generated a massive surge in data, driving the utilization of Federated Learning (FL) as a decentralized model training solution that guarantees user privacy. Despite its promise, the implementation of FL in IoT infrastructures is frequently hindered by cybersecurity vulnerabilities and hardware computational constraints. Therefore, this paper presents a Systematic Literature Review (SLR) guided by the PRISMA 2020 standards to analyze the architectures, security protocols, and future projections of FL in IoT environments. Through the extraction and synthesis of 25 selected primary literatures indexed in the Scopus database, the findings indicate that Deep Learning models-particularly CNN and LSTM-are the most dominant approaches. Furthermore, the combination of Homomorphic Encryption and Differential Privacy is established as the primary benchmark in mitigating model inversion attacks. In terms of constraints, the high communication overhead during gradient exchange on edge devices remains the largest operational hurdle. Ultimately, this study concludes that the integration of FL with Serverless infrastructure and Blockchain technology represents a highly potential future innovation to eliminate centralized failure risks while realizing a secure and autonomous IoT network.

Analisis Kelayakan Penerapan Machine Learning pada Industri Musik

Farid Hasfindra — 2026-07-05

The application of machine learning (ML) has transformed the music industry, offering efficiencies in music generation, genre classification, vocal training, and music education. However, this vast potential is countered by fundamental challenges, including dataset bias, legal issues surrounding Artificial Intelligence (AI) copyright, ethical challenges related to algorithmic transparency, and accessibility barriers for independent musicians. This literature review aims to assess the feasibility of ML solutions in the music industry through a comprehensive analysis of technical, legal, ethical, and economic aspects. The method used is a Systematic Literature Review (SLR) guided by the PRISMA framework, involving a thematic analysis of 29 scientific articles that passed the selection and quality assessment process. The results demonstrate that ML is proven efficient in improving the accuracy of music generation (83-90%), genre classification performance, and personalized vocal learning experiences. Nevertheless, full feasibility is hindered by data bias issues (the dominance of Western music datasets), ambiguity in copyright regulations for AI generated music, computational costs, and a lack of model transparency. In conclusion, current ML implementation is feasible within specific contexts, such as educational institutions and large platforms with adequate infrastructure. Future prospects require the development of more representative datasets, transparent models, and clear regulatory frameworks to address industry inequities.