Iterative Method For Blockchain-Based Qos-Aware Iov Network Using Ai-Driven Anomaly Detection And Dynamic Network Slicing
Year: 2024
Type: article
Abstract: The rapid deployment of 5G networks necessitates the development of secure, scalable, and efficient Internet of Vehicles (IoV) systems. Existing IoV solutions often struggle with real-time threat detection, scalability, efficient resource allocation, and privacy preservation. This work proposes an integrated framework leveraging blockchain technology, AI-driven anomaly detection, dynamic network slicing, and secure multi-party computations. We introduce AI-Driven Anomaly Detection and Mitigation (ADAM) to identify and respond to security threats in real-time. Utilizing Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), ADAM analyzes network traffic data to detect anomalies with a detection accuracy of 95%, a false positive rate of 2%, and an average response timestamp of 50 ms. To tackle scalability and latency issues inherent in traditional blockchain systems, we propose Edge-Based Blockchain Sharding (EBBS).The innovative use of a modified Proof-of-Stake (PoS) mechanism tailored for edge environments further enhances the scalability of the IoV system. AI-Enabled Dynamic Network Slicing (ADNS) is implemented to optimize resource allocation based on real-time traffic demands and QoS requirements. Finally, we incorporate Secure Multi-Party Computation for Collaborative Data Processing (SMPC-CDP) to enable secure, privacy-preserving data analysis among IoV entities ensuring privacy with a computation overhead of 20%, and data utility preservation of 95%.
Authors: Pranjali Ulhe -, Suresh Asole -
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4405622744