Securing differentiated services code points using blockchain-enabled intelligent deep packet inspection

Abstract The differentiated services code points (DSCPs) are widely used to enforce Quality of Service (QoS) policies in IP networks. However, their lack of intrinsic integrity protection exposes networks to priority manipulation and unfair resource allocation. Existing deep packet inspection systems and intrusion detection methods provide limited protection primarily focused on payload and header-based anomaly detection. This study proposes an intelligent blockchain-enabled deep packet inspection (IBDPI) architecture which integrates deep learning-based traffic classification with a tamper-resistant blockchain ledger to secure DSCP integrity during network transmission. Multiple deep learning architectures including convolutional neural networks, recurrent neural networks, long short-term memory networks, autoencoders and an ensemble model are employed to classify IP packets based on DSCP characteristics extracted from packet headers. To address blockchain scalability constraints, a hybrid architecture integrating Apache Kafka and off-chain NoSQL storage is introduced to enable selective on-chain recording of priority-critical packet metadata using Ethereum smart contracts. The experimental evaluation using a simulated multi-application network testbed demonstrates high DSCP classification accuracy up to 99.72% while maintaining stable performance under varying traffic loads. The blockchain integration provides immutable auditability of priority-critical packets thus mitigates the DSCP manipulation risks and ensures reliable forensic analysis. The proposed IBDPI framework offers a practical and scalable approach for enhancing QoS security in modern IP networks and establishes a foundation for blockchain-assisted intelligent traffic management and auditing systems.