The identification of heart arrhythmias is critical for timely and accurate diagnosis and treatment of cardiac diseases. Electrocardiogram (ECG) signals provide valuable information for detecting and diagnosing heart arrhythmias. However, the interpretation of ECG signals requires specialized medical knowledge and experience. In present years, deep learning techniques have shown promising results in ECG signal analysis for heart arrhythmia detection. This paper proposes a deep learning suggest to the identification of heart arrhythmias by utilizing ECG signals on edge devices. The proposed approach employs a convolutional neural network (CNN) framework to analyze the ECG signals and identify different types of heart arrhythmias. The CNN architecture is optimized for edge devices, ensuring that the proposed approach can be implemented in resource-constrained environments. To assess the effectiveness of the suggested strategy, a dataset containing ECG signals from patients with different types of heart arrhythmias is used. The experimental findings show that the suggested method successfully detects heart arrhythmias with high accuracy, up to 99.83%.Moreover, the proposed approach shows significant advantages in terms of execution time and energy consumption when compared to traditional ECG signal analysis techniques