@article{RoschkeChengMeinel2012,
  author    = {Roschke, Sebastian and Cheng, Feng and Meinel, Christoph},
  title     = {An alert correlation platform for memory-supported techniques},
  series = {Concurrency and computation : practice \& experience},
  volume    = {24},
  journal   = {Concurrency and computation : practice \& experience},
  number    = {10},
  publisher = {Wiley-Blackwell},
  address   = {Hoboken},
  issn      = {1532-0626},
  doi       = {10.1002/cpe.1750},
  pages     = {1123 -- 1136},
  year      = {2012},
  abstract  = {Intrusion Detection Systems (IDS) have been widely deployed in practice for detecting malicious behavior on network communication and hosts. False-positive alerts are a popular problem for most IDS approaches. The solution to address this problem is to enhance the detection process by correlation and clustering of alerts. To meet the practical requirements, this process needs to be finished fast, which is a challenging task as the amount of alerts in large-scale IDS deployments is significantly high. We identifytextitdata storage and processing algorithms to be the most important factors influencing the performance of clustering and correlation. We propose and implement a highly efficient alert correlation platform. For storage, a column-based database, an In-Memory alert storage, and memory-based index tables lead to significant improvements of the performance. For processing, algorithms are designed and implemented which are optimized for In-Memory databases, e.g. an attack graph-based correlation algorithm. The platform can be distributed over multiple processing units to share memory and processing power. A standardized interface is designed to provide a unified view of result reports for end users. The efficiency of the platform is tested by practical experiments with several alert storage approaches, multiple algorithms, as well as a local and a distributed deployment.},
  language  = {en}
}
@article{RoschkeChengMeinel2013,
  author    = {Roschke, Sebastian and Cheng, Feng and Meinel, Christoph},
  title     = {High-quality attack graph-based IDS correlation},
  series = {Logic journal of the IGPL},
  volume    = {21},
  journal   = {Logic journal of the IGPL},
  number    = {4},
  publisher = {Oxford Univ. Press},
  address   = {Oxford},
  issn      = {1367-0751},
  doi       = {10.1093/jigpal/jzs034},
  pages     = {571 -- 591},
  year      = {2013},
  abstract  = {Intrusion Detection Systems are widely deployed in computer networks. As modern attacks are getting more sophisticated and the number of sensors and network nodes grow, the problem of false positives and alert analysis becomes more difficult to solve. Alert correlation was proposed to analyse alerts and to decrease false positives. Knowledge about the target system or environment is usually necessary for efficient alert correlation. For representing the environment information as well as potential exploits, the existing vulnerabilities and their Attack Graph (AG) is used. It is useful for networks to generate an AG and to organize certain vulnerabilities in a reasonable way. In this article, a correlation algorithm based on AGs is designed that is capable of detecting multiple attack scenarios for forensic analysis. It can be parameterized to adjust the robustness and accuracy. A formal model of the algorithm is presented and an implementation is tested to analyse the different parameters on a real set of alerts from a local network. To improve the speed of the algorithm, a multi-core version is proposed and a HMM-supported version can be used to further improve the quality. The parallel implementation is tested on a multi-core correlation platform, using CPUs and GPUs.},
  language  = {en}
}