Hardware Trojans pose a significant threat to the securitySecurity and integrity of integrated circuits, particularly in the context of complex System-on-Chip designs that incorporate third-party intellectual property. This paper presents a novel approach to hardware Trojan detectionHardware trojan detection using deep learningDeep learning techniques applied to side-channel analysis. We employ a one-dimensional convolutional neural network to classify electromagnetic side-channel signals from AES encryption cores, aiming to detect the presence of hardware Trojans. Our method leverages the Trust-Hub dataset, focusing on AES-T1300 and AES-T1400 benchmarks with varying trigger conditions and payloads. The proposed CNN architecture demonstrates high accuracy in distinguishing between Trojan-free and Trojan-infected circuits. This research contributes to the field of hardware securitySecurity by offering a non-invasive, machine learning-based approach to Trojan detection that can be applied in real-world scenarios where golden chip references are unavailable. The success of our method underscores the potential of deep learningDeep learning in enhancing the securitySecurity of integrated circuits against sophisticated hardware-level threats.