Agentic-SecPBFT: Agentic AI-Driven Proactive Security Framework for Wireless PBFT Consensus in Mobile Ad-Hoc Networks

2026-07-03

AI Summary

This study addresses the vulnerability of standard PBFT protocols in mobile ad hoc networks to sophisticated coordinated attacks, including Sybil attacks, Byzantine collusion, and message tampering. To this end, it pioneers the integration of the Agentic AI paradigm into wireless PBFT consensus security by constructing a distributed multi-agent system wherein each node is equipped with an intelligent agent. These agents leverage local behavioral observations, message consistency checks, and dynamic reputation scoring to collaboratively learn and execute proactive defense strategies in real time under PBFT quorum rules, using a hierarchical multi-agent deep Q-network (MADQN). Experimental results demonstrate that the proposed approach achieves a 95.0% attack detection rate with only a 1.8% false positive rate; under a 33% malicious node scenario, it improves throughput by 3.1× and reduces latency by 56%, substantially outperforming existing PBFT variants and effecting a paradigm shift from passive defense to active learning and collaborative response.

Abstract

The standard Practical Byzantine Fault Tolerance (PBFT) protocol, designed for stable, wired environments, exhibits critical vulnerabilities when deployed in settings like mobile ad-hoc networks, thus making it susceptible to sophisticated threats such as Sybil attacks, Byzantine collusion, and message manipulation. Existing static defense mechanisms are ill-equipped to handle the intelligent and coordinated nature of these attacks. To address this challenge, this paper leverages the Agentic AI paradigm to build a distributed multi-agent system in which each consensus node is equipped with an intelligent agent. These agents employ a hierarchical Multi-Agent Deep Q-Network (MADQN) algorithm to learn and execute proactive security policies in real-time. By observing local network behavior, message consistency, and dynamically maintained reputation scores, the agents collaboratively identify suspicious behavior and recommend defensive actions under standard PBFT quorum and membership rules, thereby improving the integrity of the consensus process. We refer to the resulting framework as Agentic-SecPBFT. Then, we formally model key attack vectors and conduct extensive simulations. The results demonstrate that Agentic-SecPBFT reaches a 95.0% attack detection rate with a 1.8% false positive rate. Compared with mainstream PBFT variants, it achieves 3.1* higher throughput with 56% lower latency on average under 33% malicious nodes, offering a robust and adaptive security solution for decentralized wireless systems.

Problem

• PBFT
• mobile ad-hoc networks
• Byzantine attacks
• Sybil attacks
• message manipulation

Innovation

• Agentic AI
• Multi-Agent Deep Q-Network
• Proactive Security
• Wireless PBFT
• Mobile Ad-Hoc Networks