Using efficient deep learning techniques for mobile crowd sensing detection in an IOTA-based framework

This paper introduces a novel approach for securing mobile crowd sensing (MCS) systems, with a focus on improving the safety and efficiency of crowd management during the Hajj pilgrimage through the integration of deep learning techniques within an IOTA-based framework. The proposed method employs a logit-boosted convolutional neural network (Logit-CNN) model to address significant security threats, such as jamming, spoofing, and faked sensing attacks, which are prevalent in large-scale, dynamic, and heterogeneous networks. Through comprehensive performance evaluations, the Logit-CNN model demonstrated superior accuracy and reliability, achieving a 99.5% accuracy, 99% precision, and 98% recall, outperforming traditional security methods by significant margins. These results highlight the model's ability to provide real-time anomaly detection, ensuring enhanced security and resource allocation. Furthermore, the study underscores the practical implications of deploying deep learning models in MCS systems, offering valuable insights into the challenges of real-world implementation and suggesting pathways for future research to further refine these security measures. The integration of deep learning with MCS systems not only elevates the overall security and management of large-scale events like the Hajj but also paves the way for its application in other similar environments.