The electromagnetic spectrum has emerged as a critical operational domain in modern military and civilian communication environments. The rapid proliferation of wireless devices, software-defined radios (SDRs), unmanned systems, satellite communications, and Internet of Things (IoT) technologies has significantly increased spectrum congestion and operational complexity. As a result, the ability to monitor, classify, authenticate, and protect radio frequency (RF) transmissions has become an essential requirement for communication security, electronic warfare (EW), and signal intelligence (SIGINT) [16], [32], [33].

Signal Intelligence (SIGINT) refers to the collection and analysis of electromagnetic emissions to extract actionable information regarding communication systems, emitters, and operational activities. Traditionally, SIGINT systems relied on expert-driven signal processing techniques and manually engineered features to identify communication signals and characterize emitters. However, the growing diversity of RF emitters, increasing spectrum density, and the emergence of sophisticated electronic attack techniques such as jamming and spoofing have exposed limitations in conventional approaches [16], [17]. Modern tactical environments demand intelligent systems capable of rapidly analysing large volumes of RF data while maintaining high classification accuracy and operational adaptability.