An Intrusion Detection System for 5G SDN Network Utilizing Binarized Deep Spiking Capsule Fire Hawk Neural Networks and Blockchain Technology

• Published in: Future internet 2024-10, Vol.16 (10), p.359
• Main Authors: Nayak, Nanavath Kiran Singh, Bhattacharyya, Budhaditya
• Format: Article
• Language: English
• Subjects: 5G SDN; Accuracy; Algorithms; Artificial intelligence; binarized deep spiking capsule fire hawk neural network; Blockchain; Controllers; Cybersecurity; Data acquisition; Data encryption; Data integrity; Data transmission; Denial of service attacks; Encryption; Energy efficiency; Innovations; Internet of Things; intrusion detection system; Intrusion detection systems; Machine learning; Network latency; Neural networks; Packet transmission; Real time; Security systems; Software-defined networking; Spiking; Trusted third parties
• ISSN: 1999-5903; 1999-5903
• DOI: 10.3390/fi16100359

The advent of 5G heralds unprecedented connectivity with high throughput and low latency for network users. Software-defined networking (SDN) plays a significant role in fulfilling these requirements. However, it poses substantial security challenges due to its inherent centralized management strategy. Moreover, SDN confronts limitations in handling malicious traffic under 5G’s extensive data flow. To deal with these issues, this paper presents a novel intrusion detection system (IDS) designed for 5G SDN networks, leveraging the advanced capabilities of binarized deep spiking capsule fire hawk neural networks (BSHNN) and blockchain technology, which operates across multiple layers. Initially, the lightweight encryption algorithm (LEA) is used at the data acquisition layer to authenticate mobile users via trusted third parties. Followed by optimal switch selection using the mud-ring algorithm in the switch layer, and the data flow rules are secured by employing blockchain technology incorporating searchable encryption algorithms within the blockchain plane. The domain controller layer utilizes binarized deep spiking capsule fire hawk neural network (BSHNN) for real-time data packet classification, while the smart controller layer uses enhanced adapting hidden attribute-weighted naive bayes (EAWNB) to identify suspicious packets during data transmission. The experimental results show that the proposed technique outperforms the state-of-the-art approaches in terms of accuracy (98.02%), precision (96.40%), detection rate (96.41%), authentication time (16.2 s), throughput, delay, and packet loss ratio.