Optimal detection of border gateway protocol anomalies with extensive feature set

For effective and secure access of Internet, the Border Gateway Protocol (BGP) has to be capable to identify and stop odd concurrences in realistic time. Despite the fact that more studies were done over the precedent 10 years to find out anomalies in BGP, the issue is still demanding since attacker...

Full description

Saved in:
Bibliographic Details
Published in:Multimedia tools and applications 2024-05, Vol.83 (17), p.50893-50919
Main Authors: Sunita, M., Mallapur, Sujata V.
Format: Article
Language:English
Subjects:
Accuracy
Anomalies
Computer Communication Networks
Computer Science
Data Structures and Information Theory
Feature extraction
Multimedia Information Systems
Special Purpose and Application-Based Systems
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:For effective and secure access of Internet, the Border Gateway Protocol (BGP) has to be capable to identify and stop odd concurrences in realistic time. Despite the fact that more studies were done over the precedent 10 years to find out anomalies in BGP, the issue is still demanding since attackers and network misconfigurations frequently exhibit new, peculiar behavior. The following two main parts establishes a novel BGP anomaly detection model: It reads, "(i) Feature Extraction; (ii) Anomaly Detection."Extensive features, such as "statistical features," "higher-order statistical features," "improved holo-entropy features," and "correntropy features" are retrieved to improve the detection's accuracy and dependability. Next, the proposed DBN is deployed to identify the existence or absence of an anomaly. Furthermore, a hybrid RHMFO Optimization is used to fine-tune the weight of DBN in order to improve classification accuracy. The DBN result lets us know whether there are network anomalies or not. Finally, analysis is done, where, accuracy of the DBN + RHMFO is ( ~) 97%, which is 12.3%, 27.83%, 48.4%, 69.07%, and 51.5% improved than MLP-NN, SVM-BGPAD, DBN + ROA, DBN + EHO, and DBN + MFO, respectively.
ISSN:1573-7721
1380-7501
1573-7721
DOI:10.1007/s11042-023-17135-6