The advancements in wireless technologies, artificial intelligence (AI), the internet of bio-nano things (IoBNT), and integrated electronic circuits have revolutionized the healthcare sector. Traditional device-based healthcare monitoring practices demand wearable or implantable sensing devices, with associated risks and drawbacks. An alternative safer approach involves leveraging device-free technologies. Acknowledging the significance of sleep monitoring in assessing a subject’s health, in the proposed work, we introduce a Convolutional Neural Network (CNN)-based architecture for the classification of sleep-posture transition. Our approach integrates signal processing techniques with deep learning (DL) to enhance performance, achieving a promising sleep-posture recognition accuracy of 77.56% without data augmentation on publicly available datasets demonstrating superior performance over existing state-of-the-art methods. This reduces the computational overhead of the proposed method. Moreover, after employing data augmentation techniques, the classification performance increases to 83.03% outperforming other methods.