DEEP LEARNING-BASED SIGNAL CLASSIFICATION FOR INTELLIGENT COMMUNICATION SYSTEMS

Abstract

Intelligent communication systems require accurate and real-time classification of wireless signals to support applications such as spectrum sensing, adaptive modulation, cognitive radio, and interference management. Traditional signal classification techniques rely on handcrafted features and statistical models, which often fail to achieve robust performance in complex and noisy environments. This paper presents a deep learning-based signal classification framework that automatically learns discriminative features from raw or preprocessed signal data. Convolutional and recurrent neural network models are employed to classify different modulation and signal types under varying channel conditions. MATLAB/Simulink-based simulations demonstrate improved classification accuracy and robustness compared to conventional machine learning approaches, validating the effectiveness of deep learning for intelligent communication systems.