This paper presents a genetic programming based detection system for Data Link layer attacks on a WiFi network. We explore the use of two different fitness functions in order to achieve both a high detection rate and a low false positive rate. Results show that the detection system developed can achieve a detection rate above 90% and a false positive rate below 1%.

Data Link layer attacks on WiFi networks are known to be one of the weakest points of WiFi networks. While these attacks are very simple in implementation, their effect on WiFi networks can be devastating. To this end, several Intrusion Detection Systems (IDS) have been employed to detect these attacks. In this paper, we compare 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. We focus particularly on a scenario where the attacker stealthily injects the attack frames into the target network. Results show that the GP based detection system is much more robust against the different versions of the attack compared to Snort-Wireless and can achieve a detection rate in average 100% and a false positive rate in average 0.1%.

Abstract. After the Fukushima accident, initiatives emerged from the public to carry out themselves measurements of the radioactivity in the environment with various devices, among which smartphones, and to share data and experiences through collaborative tools and social networks. Such measurements have two major interests, on the one hand, to enable each individual of the public to assess his own risk regarding the radioactivity and, on the other hand, to provide "real time" data from the field at various locations, especially in the early phase of an emergency situation, which could be very useful for the emergency management.The objective of the OPENRADIATION project is to offer to the public the opportunity to be an actor for measurements of the radioactivity in the environment using connected dosimetric applications on smartphones. The challenge is to operate such a system on a sustainable basis in peaceful time and be useful in case of emergency. In "peaceful situation", this project is based on a collaborative approach with the aim to get complementary data to the existing ones, to consolidate the radiation background, to generate alerts in case of problem and to provide education & training and enhanced pedagogical approaches for a clear understanding of measures for the public. In case of emergency situation, data will be available "spontaneously" from the field in "real time" providing an opportunity for the emergency management and the communication with the public. … The practical objective is i) to develop a website centralising data from various systems/dosimeters, providing dose maps with raw and filtered data and creating dedicated areas for specific initiatives and exchanges of data and ii) to develop a data acquisition protocol and a dosimetric application using a connected dosimeter with a bluetooth connection.This project is conducted within a partnership between organisms' representative of the scientific community and associations to create links with the public.

Genetic Programming (GP) based Intrusion Detection Systems (IDS) use connection state network data during their training phase. These connection states are recorded as a set of features that the GP uses to train and test solutions which allow for the efficient and accurate detection of given attack patterns. However, when applied to a 802.11 network that is faced with attacks specific to the 802.11 protocol, the GP's detection rate reduces dramatically. In this work we discuss what causes this effect, and what can be done to improve the GP's performance on 802.11 networks.

No abstract