No abstract

No abstract

Multi-Agent Systems (MAS) have been widely used in many areas like modeling and simulation of complex phenomena, and distributed problem solving. Likewise, MAS have been used in cyber-security, to build more efficient Intrusion Detection Systems (IDS), namely Collaborative Intrusion Detection Systems (CIDS). This work presents a taxonomy for classifying the methods used to design intrusion detection systems, and how such methods were used alongside with MAS in order to build IDS that are deployed in distributed environments, resulting in the emergence of CIDS. The proposed taxonomy, consists of three parts: 1) general architecture of CIDS, 2) the used agent technology, and 3) decision techniques, in which used technologies are presented. The proposed taxonomy reviews and classifies the most relevant works in this topic and highlights open research issues in view of recent and emerging threats. Thus, this work provides a good insight regarding past, current, and future solutions for CIDS, and helps both researchers and professionals design more effective solutions.

Anti-malware tools remain the primary line of defense against malicious software. There is a wide variety of commercial anti-malware tools in the IT security market. However, no single tool is able to provide a full protection against the overwhelming number of daily released malware. Hence, collaboration among malware detection tools is of paramount importance. In this paper, we propose MACoMal, a multiagent based decision mechanism, which assists heterogeneous anti-malware tools to collaborate with each other in order to reach a consensual decision about the maliciousness of a suspicious file. MACoMal consists of two main elements: (1) an executable file identification model, and (2) a collaborative decision-making scheme. MACoMal is analyzed with respect to network connectivity and global decision correctness. By leveraging a multi-agent simulation tool and a set of real malware samples, we present a simulation methodology to assess its effectiveness and efficiency. Experimental results show that MACoMal is able to immunize a network against a malware threat within a time that ranges from a few seconds to a few minutes after the threat detection.

Abstract. This paper presents and evaluates a multi-agent approach for range image segmentation. A set of reactive and autonomous agents perform a collective segmentation by partitioning a range image in its different planar regions. The agents move over the image and perform cooperative and competitive actions on the pixels, allowing a robust region extraction, and an accurate edge detection. An artificial potential field, created around the pixels of interest, ensures the agent coordination. It allows the agents to concentrate their actions around the edges and the noise regions. The experimental results show the potential of the proposed approach for scene understanding in range images, regarding both segmentation efficiency, and detection accuracy.