Enhancing intrusion detection in MANETs with blockchain-based trust management and enhanced GRU model

This paper proposes a Blockchain-based Intrusion Detection System (IDS) for Secure Mobile Ad-Hoc Networks (MANET) to overcome limitations in existing intrusion detection techniques. The scheme starts with noise elimination through data preprocessing using the Mantis Search Algorithm (MSA) for featur...

Full description

Saved in:
Bibliographic Details
Published in:Peer-to-peer networking and applications 2025-02, Vol.18 (1), p.1-22
Main Authors: Krishna, ES Phalguna, Sandeep, Daria, Kocherla, Raviteja, Lella, Kranthi Kumar, Molugu, Srikanth, Ibrahim, Shaik Hussain Shaik, Vatambeti, Ramesh
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Blockchain
Collaboration
Communications Engineering
Computer Communication Networks
Computer science
Confidentiality
Cybersecurity
Energy consumption
Energy efficiency
Engineering
Feature selection
Information Systems and Communication Service
Infrastructure
Intrusion detection systems
Mobile ad hoc networks
Networks
Optimization
Peer to peer computing
Routing (telecommunications)
Search algorithms
Signal,Image and Speech Processing
Trustworthiness
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper proposes a Blockchain-based Intrusion Detection System (IDS) for Secure Mobile Ad-Hoc Networks (MANET) to overcome limitations in existing intrusion detection techniques. The scheme starts with noise elimination through data preprocessing using the Mantis Search Algorithm (MSA) for feature selection. A lightweight consensus algorithm is introduced for blockchain-based trust management in MANETs, ensuring tamper-proof distributed trust framework for node routing. Integration of blockchain with the Optimized Link State Routing Protocol (OLSR) addresses security concerns effectively. Each node independently executes security operations within the proposed framework, fostering cooperation to combat network adversaries. Gated Recurrent Units (GRU) are employed for classification, with hyperparameter optimization using Giant Armadillo Optimization (GAO). Comparative analysis demonstrates the enhanced GRU model's superiority in intrusion detection, achieving 99.6% sensitivity, 99.4% specificity, 99.4% F-measure, and 99.5% accuracy. Additionally, the proposed GRU model enhances packet delivery ratio, throughput, and reduces routing overhead in MANET.
ISSN:1936-6442
1936-6450
DOI:10.1007/s12083-024-01877-7