A critical review on the convergence of blockchain and machine learning in Deep Packet Inspection systems to enhance network traffic security, performance and management

Abstract

Abstract The growing volume and complexity of network data necessitate advance solutions for network traffic analysis and security. The Deep Packet Inspection (DPI) offers a granular approach to monitoring, filtering and classifying network traffic to enforce security policies, optimize QoS and detect malicious activities. The proposed study addresses these issues by exploring the emerging but promising integration of blockchain and machine learning techniques to improve DPI. It contributes by providing a comprehensive details on the application domain of DPI with a focus on network security, performance and management. Also, the study proposes a research roadmap to guide the future development on the development of blockchain-enabled intelligent solutions for DPI. Using PRISMA methodology, several existing studies were evaluated addressing the potential application of blockchain and machine learning in DPI. The survey has identified significant challenges towards the integration including real-time IP packet inspection efficiency, QoS performance and the impact of high traffic volume on DPI. It concludes that DPI has wider applications to be integrated with emerging technologies particularly in machine learning and blockchain. The future research should focus on advance machine learning paradigms such as continual and federated learning while blockchain technology should be resolved with scalability challenges to be utilized effectively for next-generation DPI solutions.