Application of Machine Learning Techniques for Secure Traffic in NoC-based Manycores
- URL: http://arxiv.org/abs/2501.12034v1
- Date: Tue, 21 Jan 2025 10:58:09 GMT
- Title: Application of Machine Learning Techniques for Secure Traffic in NoC-based Manycores
- Authors: Geaninne Lopes, César Marcon, Fernando Moraes,
- Abstract summary: This document explores an IDS technique using machine learning and temporal series for detecting DoS attacks in NoC-based manycore systems.<n>It is necessary to extract traffic data from a manycore NoC and execute the learning techniques in the extracted data.<n>The developed platform will have its data validated with a low-level platform.
- Score: 44.99833362998488
- License: http://creativecommons.org/licenses/by-nc-nd/4.0/
- Abstract: Like most computer systems, a manycore can also be the target of security attacks. It is essential to ensure the security of the NoC since all information travels through its channels, and any interference in the traffic of messages can reflect on the entire chip, causing communication problems. Among the possible attacks on NoC, Denial of Service (DoS) attacks are the most cited in the literature. The state of the art shows a lack of work that can detect such attacks through learning techniques. On the other hand, these techniques are widely explored in computer network security via an Intrusion Detection System (IDS). In this context, the main goal of this document is to present the progress of a work that explores an IDS technique using machine learning and temporal series for detecting DoS attacks in NoC-based manycore systems. To fulfill this goal, it is necessary to extract traffic data from a manycore NoC and execute the learning techniques in the extracted data. However, while low-level platforms offer precision and slow execution, high-level platforms offer higher speed and data incompatible with reality. Therefore, a platform is being developed using the OVP tool, which has a higher level of abstraction. To solve the low precision problem, the developed platform will have its data validated with a low-level platform.
Related papers
- Deep Learning-Based Intrusion Detection for Automotive Ethernet: Evaluating & Optimizing Fast Inference Techniques for Deployment on Low-Cost Platform [2.4554107118749555]
Deep Learning-based Intrusion Detection Systems (IDS) are often designed to combat this problem, but they require expensive hardware to run in real time.<n>We propose to evaluate and apply fast neural network inference techniques like Distilling and Prunning for deploying IDS models on low-cost platforms in real time.<n>The results show that these techniques can achieve intrusion detection times of up to 727 mus using a Raspberry Pi 4, with AUCROC values of 0.9890.
arXiv Detail & Related papers (2025-07-01T22:05:02Z) - A Review of Various Datasets for Machine Learning Algorithm-Based Intrusion Detection System: Advances and Challenges [0.40964539027092917]
IDS aims to protect computer networks from security threats by detecting, notifying, and taking appropriate action to prevent illegal access and protect confidential information.<n>Researchers are enhancing the effectiveness of IDS by incorporating popular datasets into machine learning algorithms.<n>This paper explores the methods of capturing and reviewing intrusion detection systems (IDS) and evaluates the challenges existing datasets face.
arXiv Detail & Related papers (2025-06-03T04:47:21Z) - Topology-aware Detection and Localization of Distributed Denial-of-Service Attacks in Network-on-Chips [2.6490401904186758]
This paper presents a framework to conduct topology-aware detection and localization of DDoS attacks using Graph Neural Networks (GNNs)<n>By modeling the NoC as a graph, our method utilizes traffic features to effectively identify and localize DDoS attacks.<n> Experimental results demonstrate that our approach can detect and localize DDoS attacks with high accuracy (up to 99%) while maintaining consistent performance under diverse attack strategies.
arXiv Detail & Related papers (2025-05-20T20:49:34Z) - CANTXSec: A Deterministic Intrusion Detection and Prevention System for CAN Bus Monitoring ECU Activations [53.036288487863786]
We propose CANTXSec, the first deterministic Intrusion Detection and Prevention system based on physical ECU activations.<n>It detects and prevents classical attacks in the CAN bus, while detecting advanced attacks that have been less investigated in the literature.<n>We prove the effectiveness of our solution on a physical testbed, where we achieve 100% detection accuracy in both classes of attacks while preventing 100% of FIAs.
arXiv Detail & Related papers (2025-05-14T13:37:07Z) - Intelligent IoT Attack Detection Design via ODLLM with Feature Ranking-based Knowledge Base [0.964942474860411]
Internet of Things (IoT) devices have introduced significant cybersecurity challenges.
Traditional machine learning (ML) techniques often fall short in detecting such attacks due to the complexity of blended and evolving patterns.
We propose a novel framework leveraging On-Device Large Language Models (ODLLMs) augmented with fine-tuning and knowledge base (KB) integration for intelligent IoT network attack detection.
arXiv Detail & Related papers (2025-03-27T16:41:57Z) - Preliminary study on artificial intelligence methods for cybersecurity threat detection in computer networks based on raw data packets [34.82692226532414]
In this paper, we investigate deep learning methodologies capable of detecting attacks in real-time directly from raw packet data within network traffic.
We propose a novel approach where packets are stacked into windows and separately recognised, with a 2D image representation suitable for processing with computer vision models.
arXiv Detail & Related papers (2024-07-24T15:04:00Z) - Federated Learning for Zero-Day Attack Detection in 5G and Beyond V2X Networks [9.86830550255822]
Connected and Automated Vehicles (CAVs) on top of 5G and Beyond networks (5GB) make them vulnerable to increasing vectors of security and privacy attacks.
We propose in this paper a novel detection mechanism that leverages the ability of the deep auto-encoder method to detect attacks relying only on the benign network traffic pattern.
Using federated learning, the proposed intrusion detection system can be trained with large and diverse benign network traffic, while preserving the CAVs privacy, and minimizing the communication overhead.
arXiv Detail & Related papers (2024-07-03T12:42:31Z) - TSTEM: A Cognitive Platform for Collecting Cyber Threat Intelligence in the Wild [0.06597195879147556]
The extraction of cyber threat intelligence (CTI) from open sources is a rapidly expanding defensive strategy.
Previous research has focused on improving individual components of the extraction process.
The community lacks open-source platforms for deploying streaming CTI data pipelines in the wild.
arXiv Detail & Related papers (2024-02-15T14:29:21Z) - A high throughput Intrusion Detection System (IDS) to enhance the security of data transmission among research centers [39.65647745132031]
This paper presents a packet sniffer that was designed using a commercial FPGA development board.
The system can support a data throughput of 10 Gbit/s with preliminary results showing that the speed of data transmission can be reliably extended to 100 Gbit/s.
It is particularly suited for the security of universities and research centers, where point-to-point network connections are dominant.
arXiv Detail & Related papers (2023-11-10T14:30:00Z) - DCDetector: An IoT terminal vulnerability mining system based on
distributed deep ensemble learning under source code representation [2.561778620560749]
The goal of the research is to intelligently detect vulnerabilities in source codes of high-level languages such as C/C++.
This enables us to propose a code representation of sensitive sentence-related slices of source code, and to detect vulnerabilities by designing a distributed deep ensemble learning model.
Experiments show that this method can reduce the false positive rate of traditional static analysis and improve the performance and accuracy of machine learning.
arXiv Detail & Related papers (2022-11-29T14:19:14Z) - Towards a Privacy-preserving Deep Learning-based Network Intrusion
Detection in Data Distribution Services [0.0]
Data Distribution Service (DDS) is an innovative approach towards communication in ICS/IoT infrastructure and robotics.
Traditional intrusion detection systems (IDS) do not detect any anomalies in the publish/subscribe method.
This report presents an experimental work on simulation and application of Deep Learning for their detection.
arXiv Detail & Related papers (2021-06-12T12:53:38Z) - Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks,
and Defenses [150.64470864162556]
This work systematically categorizes and discusses a wide range of dataset vulnerabilities and exploits.
In addition to describing various poisoning and backdoor threat models and the relationships among them, we develop their unified taxonomy.
arXiv Detail & Related papers (2020-12-18T22:38:47Z) - Data Poisoning Attacks on Federated Machine Learning [34.48190607495785]
Federated machine learning enables resource constrained node devices to learn a shared model while keeping the training data local.
The communication protocol amongst different nodes could be exploited by attackers to launch data poisoning attacks.
We propose a novel systems-aware optimization method, ATTack on Federated Learning (AT2FL)
arXiv Detail & Related papers (2020-04-19T03:45:05Z) - Security of Distributed Machine Learning: A Game-Theoretic Approach to
Design Secure DSVM [31.480769801354413]
This work aims to develop secure distributed algorithms to protect the learning from data poisoning and network attacks.
We establish a game-theoretic framework to capture the conflicting goals of a learner who uses distributed support vector machines (SVMs) and an attacker who is capable of modifying training data and labels.
The numerical results show that distributed SVM is prone to fail in different types of attacks, and their impact has a strong dependence on the network structure and attack capabilities.
arXiv Detail & Related papers (2020-03-08T18:54:17Z) - A Content-Based Deep Intrusion Detection System [12.590415345079995]
We propose a framework, called deep intrusion detection (DID) system, that uses the pure content of traffic flows in addition to traffic metadata.
To this end, we deploy and evaluate an offline IDS following the framework using LSTM as a deep learning technique.
arXiv Detail & Related papers (2020-01-14T19:08:57Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.