<![CDATA[The radio frequency spectrum is a finite resource that plays a crucial role in supporting communication systems, particularly in the aviation sector. Unauthorized use and signal interference within this spectrum can pose serious threats to flight safety. This study aims to develop a classification system for aviation radio signals using machine learning algorithms, namely Random Forest and Support Vector Machine. The dataset consists of tabular monitoring data with two main features: frequency and signal power level. The research process includes data preprocessing, model training, validation through K-Fold Cross Validation, and performance evaluation based on accuracy, precision, recall, and F1-score metrics. Testing was carried out under both simulated and real-world conditions using Software Defined Radio (SDR) equipment. Results indicate that the Random Forest model achieved more consistent performance, with an accuracy of 79.77% and balanced F1-scores across both classes. Additionally, the system was found to be user-friendly and responsive, especially for non-technical operators, thanks to its intuitive visual interface. This research contributes to enhancing the effectiveness of aviation spectrum monitoring and offers potential for future integration with real-time detection systems]]>
Copyrights © 2026
