• Corpus ID: 17955048

Investigating Cross-Platform Robustness for Machine Learning Based IDSs on 802.11 Networks

  title={Investigating Cross-Platform Robustness for Machine Learning Based IDSs on 802.11 Networks},
  author={Adetokunbo Makanju and Ayse Nur Zincir-Heywood},
Security and Intrusion detection in 802.11 networks is currently an active area of research where WiFi specific Data Link layer attacks are an area of focus. While these attacks are very simple in implementation, their effect on WiFi networks can be devastating. Recent research has focused on producing machine learning based IDSs for these attacks. Such IDSs have shown promise. Our work investigates the CrossPlatform robustness of such machine learning based solutions. By cross-platform… 

Figures and Tables from this paper

Adaptabilty of a GP Based IDS on Wireless Networks
The results show that the GP based detection system is much more robust against modified attacks compared to Snort-Wireless, and by focusing on the method(s) used in feature preprocessing for presentation to learning algorithms, GP based IDS can achieve infrastructure independent detection and can adapt to similar unknown attacks too.
Robust learning intrusion detection for attacks on wireless networks
This work employs Genetic Programming for evolving classifiers and Artificial Neural Networks as machine learning paradigms under three different Denial-of-Service attacks at the Data Link layer De-authentication, Authentication and Association attacks to address the problem of evaluating the robustness of machine learning based detectors for deployment in real life networks.
Knowledge Discovery from Trouble Ticketing Reports in a Large Telecommunication Company
This paper will present the techniques used for its achievement, as well as the results obtained so far, showing how these techniques may help important companies to save plenty of time and resources in fault management, improving the service quality.


A Comparison Between Signature and GP-Based IDSs for Link Layer Attacks on WiFi Networks
This paper compares the ability of Snort-Wireless and a genetic programming (GP) based intrusion detector, in the detection of a particular data link layer attack, namely the deauthentication attack.
A Wireless Intrusion Detection Method Based on Dynamic Growing Neural Network
  • Yanheng Liu, D. Tian, B. Li
  • Computer Science
    First International Multi-Symposiums on Computer and Computational Sciences (IMSCCS'06)
  • 2006
The results show that the intrusion detection method based on dynamic growing neural network can detect new intrusion behavior and some improving methods are presented in the conclusions.
Genetic programming based WiFi data link layer attack detection
A genetic programming based detection system for data link layer attacks on a WiFi network that can achieve a detection rate above 90% and a false positive rate below 1% is presented.
Training genetic programming on half a million patterns: an example from anomaly detection
The hierarchical RSS-DSS algorithm is introduced for dynamically filtering large datasets based on the concepts of training pattern age and difficulty, while utilizing a data structure to facilitate
Applying Genetic Programming to Intrusion Detection
This paper shows how an intrusion detection system can be implemented using autonomous agents, and how these agents can be built using Genetic Programming and Automatically Defined Functions (ADFs) can be used to evolve genetic programs that contain multiple data types and yet retain type-safety.
Adaptation in natural and artificial systems
Names of founding work in the area of Adaptation and modiication, which aims to mimic biological optimization, and some (Non-GA) branches of AI.
Page-based linear genetic programming
  • Malcolm J. Heywood, A. N. Zincir-Heywood
  • Computer Science
    Smc 2000 conference proceedings. 2000 ieee international conference on systems, man and cybernetics. 'cybernetics evolving to systems, humans, organizations, and their complex interactions' (cat. no.0
  • 2000
A page-based definition of individuals is used to restrict crossover to equal length code fragments and by using a register-machine context, the significance of a priori internal register external output definitions is emphasized.
Dynamic Training Subset Selection for Supervised Learning in Genetic Programming
This paper describes how to reduce the number of function-tree evaluations by selecting a small subset of the training data set on which to actually carry out the GP algorithm.
Snort wireless, http://www.snort-wireless.org
  • Retrieved from the web., 2005.
  • 2005
Void11 main page, www.wirelessdefence.org/contents/void11main Retrieved from the Web
  • Void11 main page, www.wirelessdefence.org/contents/void11main Retrieved from the Web
  • 2006