NEXT-GENERATION INTRUSION DETECTION AND PREVENTION SYSTEMS FOR IT AND NETWORK SECURITY
Abstract
In cybersecurity, the constant evolution of threats demands the development of next-generation Intrusion Detection and Prevention Systems (IDPS) to safeguard IT infrastructure and networks effectively. This research embarks on the journey of designing an innovative IDPS using a Dense VGG classifier, fueled by IoT data as its primary input source. Our approach combines the robustness of the Dense VGG architecture with the rich information generated by Internet of Things (IoT) devices, enhancing the system ability to detect and prevent intrusions. We gather diverse IoT data from sensors and devices within the IT infrastructure, ensuring the availability of labeled data that signifies known intrusion events. After meticulous preprocessing and feature engineering, we adapt the Dense VGG model, originally designed for image classification, to work with tabular IoT data. Transfer learning techniques are applied, leveraging pre-trained VGG models to expedite convergence and enhance performance. Real-time data streaming mechanisms are established to seamlessly integrate IoT data, making the system proactive in identifying threats. Upon detection, the system can respond by isolating affected devices, blocking suspicious network traffic, or initiating incident response protocols. Continuous monitoring and evaluation ensure the system reliability, with key metrics serving as indicators of its efficacy. Deployment considerations, such as scalability and redundancy, guarantee the system readiness to handle the influx of IoT data. Furthermore, integration with other security tools and compliance with regulatory standards strengthen the system overall cybersecurity posture. The core of our system lies in its intrusion detection logic, a set of rules and thresholds that trigger alerts or preventive measures based on model predictions. In testing, our system demonstrated an impressive intrusion detection accuracy of over 95%, significantly reducing false positives.

Authors
S. Bhaggiaraj1, S. Shanthini2, S.S. Sugantha Mallika3, R. Muthuram4
Sri Ramakrishna Engineering College, India1,2,3, Government College of Technology, Coimbatore, India4

Keywords
Prevention Systems, Intrusion Detection, IoT Data, Dense VGG Classifier, Intrusion Detection Accuracy, Cybersecurity
Yearly Full Views
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
231015000000
Published By :
ICTACT
Published In :
ICTACT Journal on Communication Technology
( Volume: 14 , Issue: 3 , Pages: 2992 - 2997 )
Date of Publication :
September 2023
Page Views :
506
Full Text Views :
29

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.