As clusters of Linux workstations have gained in popularity, security in this environment has become increasingly important. While prevention methods such as access control can enhance the security level of a cluster system, intrusions are still possible and therefore intrusion detection and recovery methods are necessary. In this thesis, a system architecture for an intrusion detection system in a cluster environment is presented. A prototype system called pShield based on this architecture for a Linux cluster environment is described and its capability to detect unique attacks on MPI programs is demonstrated. The pShield system was implemented as a loadable kernel module that uses a neural network classifier to model normal behavior of processes. A new method for generating artificial anomalous data is described that uses a limited amount of attack data in training the neural network. Experimental results demonstrate that using this method rather than randomly generated anomalies reduces the false positive rate without compromising the ability to detect novel attacks. A neural network with a simple activation function is used in order to facilitate fast classification of new instances after training and to ease implementation in kernel space. Our goal is to classify the entire trace of a program¡¯s execution based on neural network classification of short sequences in the trace. Therefore, the effect of anomalous sequences in a trace must be accumulated. Several trace classification methods were compared. The results demonstrate that methods that use information about locality of anomalies are more effective than those that only look at the number of anomalies. The impact of pShield on system performance was evaluated on an 8-node cluster. Although pShield adds some overhead for each API for MPI communication, the experimental results show that a real world parallel computing benchmark was slowed only slightly by the intrusion detection system. The results demonstrate the effectiveness of pShield as a light-weight intrusion detection system in a cluster environment. This work is part of the Intelligent Intrusion Detection project of the Center for Computer Security Research at Mississippi State University.

Master of Science / Department of Computing and Information Sciences / Xinming (Simon) Ou / A well-known problem in current intrusion detection tools is that they create too many low-level alerts and system administrators find it hard to cope up with the huge volume. Also, when they have to combine multiple sources of information to confirm an attack, there is a dramatic increase in the complexity. Attackers use sophisticated techniques to evade the detection and current system monitoring tools can only observe the symptoms or effects of malicious activities. When mingled with similar effects from normal or non-malicious behavior they lead intrusion analysis to conclusions of varying confidence and high false positive/negative rates. In this thesis work we present an empirical approach to the problem of modeling uncertainty where inferred security implications of low-level observations are captured in a simple logical language augmented with uncertainty tags. We have designed an automated reasoning process that enables us to combine multiple sources of system monitoring data and extract highly-confident attack traces from the numerous possible interpretations of low-level observations. We have developed our model empirically: the starting point was a true intrusion that happened on a campus network we studied to capture the essence of the human reasoning process that led to conclusions about the attack. We then used a Datalog-like language to encode the model and a Prolog system to carry out the reasoning process. Our model and reasoning system reached the same conclusions as the human administrator on the question of which machines were certainly compromised. We then automatically generated the reasoning model needed for handling Snort alerts from the natural-language descriptions in the Snort rule repository, and developed a Snort add-on to analyze Snort alerts. Keeping the reasoning model unchanged, we applied our reasoning system to two third-party data sets and one production network. Our results showed that the reasoning model is effective on these data sets as well. We believe such an empirical approach has the potential of codifying the seemingly ad-hoc human reasoning of uncertain events, and can yield useful tools for automated intrusion analysis.

ITC/USA 2015 Conference Proceedings / The Fifty-First Annual International Telemetering Conference and Technical Exhibition / October 26-29, 2015 / Bally's Hotel & Convention Center, Las Vegas, NV / The implementation of Integrated Network Enhanced Telemetry (iNET) in telemetry applications provides significant enhancements to telemetry operations. Unfortunately such networking brings the potential for devastating cyber-attacks and networked telemetry is also susceptible to these attacks. This paper demonstrates a worked example of a social engineering attack carried out on a test bed network, analyzing the attack process from launch to detection. For this demonstration, a penetration-testing tool is used to launch the attack. This attack will be monitored to detect its signature using a network monitoring tool, and this signature will then be used to create a rule which will trigger an alert in an Intrusion Detection System. This work highlights the importance of network security in telemetry applications and is critical to current and future telemetry networks as cyber threats are widespread and potentially devastating.

The general problem we consider in this thesis is the following: we have to analyze a stream of data (records, packets, events ...) by successively applying to each piece of data a set of ``rules'. Rules are best viewed as lightweight parallel processes synchronizing on each arrival of a new piece of data. In many applications, such as signature-based intrusion detection, only a few rules are concerned with each new piece of data. But all other rules have to be executed anyway just to conclude that they can ignore it. Our goal is to make it possible to avoid this useless work completely. To do so, we perform a static analysis of the code of each rule and we build a decision tree that we apply to each piece of data before executing the rule. The decision tree tells us whether executing the rule or not will change anything to the global analysis results. The decision trees are built at compile time, but their evaluation at each cycle (i.e., for each piece of data) entails an overhead. Thus we organize the set of all computed decision trees in a way that makes their evaluation as fast as possible. The two main original contributions of this thesis are the following. Firstly, we propose a method to organize the set of decision trees and the set of active rules in such a way that deciding which rules to execute can be made optimally in O(r_u), where r_u is the number of useful rules. This time complexity is thus independent of the actual (total) number of active rules. This method is based on the use of a global decision tree that integrates all individual decision trees built from the code of the rules. Secondly, as such a global tree may quickly become much too large if usual data structures are used, we introduce a novel kind of data structure called sequential tree that allows us to keep global decision trees much smaller in many situations where the individual trees share few common conditions. (When many conditions are shared by individual trees the global tree remains small.) To assess our contribution, we first modify the implementation of ASAX, a generic system for data stream analysis based on the rule paradigm presented above. Then we compare the efficiency of the optimized system with respect to its original implementation, using the MIT Lincoln Laboratory Evaluation Dataset and a classical set of intrusion detection rules. Impressive speed-ups are obtained. Finally, our optimized implementation has been used by Nicolas Vanderavero, in his PhD thesis, for the design of stateful honeytanks (i.e., low-interaction honeypots). It makes it possible to simulate tens of thousands hosts on a single computer, with a high level of realism.

The standardization and interconnection of supervisory control and data acquisition (SCADA) systems has exposed the systems to cyber attacks. To improve the security of the SCADA systems, intrusion detection system (IDS) design is an effective method. However, traditional IDS design in the industrial networks mainly exploits the prede fined rules, which needs to be complemented and developed to adapt to the big data scenario. Therefore, this thesis aims to design an anomaly-based novel hierarchical online intrusion detection system (HOIDS) for SCADA networks based on machine learning algorithms theoretically and implement the theoretical idea of the anomaly-based intrusion detection on a testbed. The theoretical design of HOIDS by utilizing the server-client topology while keeping clients distributed for global protection, high detection rate is achieved with minimum network impact. We implement accurate models of normal-abnormal binary detection and multi-attack identification based on logistic regression and quasi-Newton optimization algorithm using the Broyden-Fletcher-Goldfarb-Shanno approach. The detection system is capable of accelerating detection by information gain based feature selection or principle component analysis based dimension reduction. By evaluating our system using the KDD99 dataset and the industrial control system datasets, we demonstrate that our design is highly scalable, e fficient and cost effective for securing SCADA infrastructures. Besides the theoretical IDS design, a testbed is modi ed and implemented for SCADA network security research. It simulates the working environment of SCADA systems with the functions of data collection and analysis for intrusion detection. The testbed is implemented to be more flexible and extensible compared to the existing related work on the testbeds. In the testbed, Bro network analyzer is introduced to support the research of anomaly-based intrusion detection. The procedures of both signature-based intrusion detection and anomaly-based intrusion detection using Bro analyzer are also presented. Besides, a generic Linux-based host is used as the container of different network functions and a human machine interface (HMI) together with the supervising network is set up to simulate the control center. The testbed does not implement a large number of traffic generation methods, but still provides useful examples of generating normal and abnormal traffic. Besides, the testbed can be modi ed or expanded in the future work about SCADA network security. / Graduate

The decrease in price and the ease of use of wireless network devices make them an attractive alternative to standard wired networks. However, the intrinsic insecurity of wireless media and weaknesses in the standards for use of wireless media leave wireless networks vulnerable to attacks from unauthorized users. The intrinsic insecurity of wireless media results from radio signals extending beyond the networks intended coverage area and the weaknesses in the standards result from the methods used for authorization and privacy. These insecurities restrict the use of wireless networks by entities that need a high level of security. This paper describes a Wireless Intrusion Detection System (WIDS) that provides additional security for 802.11b wireless networks. WIDS provides intrusion detection that can react to potential threats and locate an intruder through the use of intelligent access points equipped with rotating directional antennas.

Most existent protocols, applications and services for Mobile Ad Hoc NET-works (MANETs) assume a cooperative and friendly network environment and do not accommodate security. Therefore, Intrusion Detection Systems (IDSs), serving as the second line of defense for information systems, are indispensable for MANETs with high security requirements. Central to the research described in this dissertation is the proposed two-level nonoverlapping Zone-Based Intrusion Detection System (ZBIDS) which fit the unique requirement of MANETs. First, in the low-level of ZBIDS, I propose an intrusion detection agent model and present a Markov Chain based anomaly detection algorithm. Local and trusted communication activities such as routing table related features are periodically selected and formatted with minimum errors from raw data. A Markov Chain based normal profile is then constructed to capture the temporal dependency among network activities and accommodate the dynamic nature of raw data. A local detection model aggregating abnormal behaviors is constructed to reflect recent subject activities in order to achieve low false positive ratio and high detection ratio. A set of criteria to tune parameters is developed and the performance trade-off is discussed. Second, I present a nonoverlapping Zone-based framework to manage locally generated alerts from a wider area. An alert data model conformed to the Intrusion Detection Message Exchange Format (IDMEF) is presented to suit the needs of MANETs. Furthermore, an aggregation algorithm utilizing attribute similarity from alert messages is proposed to integrate security related information from a wider area. In this way, the gateway nodes of ZBIDS can reduce false positive ratio, improve detection ratio, and present more diagnostic information about the attack. Third, MANET IDSs need to consider mobility impact and adjust their behavior dynamically. I first demonstrate that nodes?? moving speed, a commonly used parameter in tuning IDS performance, is not an effective metric for the performance measurement of MANET IDSs. A new feature -link change rate -is then proposed as a unified metric for local MANET IDSs to adaptively select normal profiles . Different mobility models are utilized to evaluate the performance of the adaptive mechanisms.

Network access control (NAC) systems have a very important role in network security. However, NAC policy configuration is an extremely complicated and error-prone task due to the semantic complexity of NAC policies and the large number of rules that could exist. This significantly increases the possibility of policy misconfigurations and network vulnerabilities. NAC policy misconfigurations jeopardize network security and can result in a severe consequence such as reachability and denial of service problems. In this thesis, we choose to study and analyze the NAC policy configuration of two significant network security devices, namely, firewall and IDS/IPS. In the first part of the thesis, a visualization technique is proposed to visualize firewall rules and policies to efficiently enhance the understanding and inspection of firewall configuration. This is implemented in a tool called PolicyVis. Our tool helps the user to answer general questions such as ‘‘Does this policy satisfy my connection/security requirements’’. If not, the user can detect all misconfigurations in the firewall policy. In the second part of the thesis, we study various policy misconfigurations of Snort, a very popular IDS/IPS. We focus on the misconfigurations of the flowbits option which is one of the most important features to offers a stateful signature-based NIDS. We particularly concentrate on a class of flowbits misconfiguration that makes Snort susceptible to false negatives. We propose a method to detect the flowbits misconfiguration, suggest practical solutions with controllable false positives to fix the misconfiguration and formally prove that the solutions are complete and sound.

A variety of intrusion prevention techniques, such as user authentication (e.g.: using passwords), avoidance of programming errors, and information protection, have been used to protect computer systems. However, intrusion prevention alone is not sufficient to protect our systems, as those systems become ever more complex with the rapid growth and expansion of Internet technology and local network systems. Moreover, programming errors, firewall configuration errors, and ambiguous or undefined security policies add to the system’s complexity. An Intrusion Detection System (IDS) is therefore needed as another layer to protect computer systems. The IDS is one of the most important techniques of information dynamic security technology. It is defined as a process of monitoring the events occurring in a computer system or network and analyzing them to differentiate between normal activities of the system and behaviours that can be classified as suspicious or intrusive. Current Intrusion Detection Systems have several known shortcomings, such as: low accuracy (registering high False Positives and False Negatives); low real-time performance (processing a large amount of traffic in real time); limited scalability (storing a large number of user profiles and attack signatures); an inability to detect new attacks (recognizing new attacks when they are launched for the first time); and weak system-reactive capabilities (efficiency of response). This makes the area of IDS an attractive research field. In recent years, researchers have investigated techniques such as artificial intelligence, autonomous agents, and distributed systems for detecting intrusion in network environments. This thesis presents a novel IDS distributed architecture – Collaborative Distributed Intrusion Detection System (C-dIDS), based on lightweight IDS modules – that integrates two main concepts in order to improve IDS performance and the scalability: lightweight IDS and collaborative architecture. To accomplish the first concept, lightweight IDS, we apply two different approaches: a features selection approach and an IDS classification scheme. In the first approach, each detector (IDS module) uses smaller amounts of data in the detection process by applying a novel features selection approach called the Fuzzy Enhanced Support Vector Decision Function (Fuzzy ESVDF). This approach improves the system scalability in terms of reducing the number of needed features without degrading the overall system performance. The second approach uses a new IDS classification scheme. The proposed IDS classification scheme employs multiple specialized detectors in each layer of the TCP/IP network model. This helps collecting efficient and useful information for dIDS, increasing the system’s ability to detect different attack types and reducing the system’s scalability. The second concept uses a novel architecture for dIDS called Collaborative Distributed Intrusion Detection System (C-dIDS) to integrate these different specialized detectors (IDS modules) that are distributed on different points in the network. This architecture is a single-level hierarchy dIDS with a non-central analyzer. To make the detection decision for a specific IDS module in the system, this module must collaborate with the previous IDS module (host) in the lower level of the hierarchy only. Collaborating with other IDS modules improves the overall system accuracy without creating a heavy system overload. Also, this architecture avoids both single point of failure and scalability bottleneck problems. Integration of the two main concepts, lightweight IDS and a distributed collaborative architecture, has shown very good results and has addressed many IDS limitations.

Intrusion detection systems (IDSes) are an important security measure that network administrators adopt to defend computer networks against malicious attacks and intrusions. The field of IDS research includes many challenges. However, one open problem remains orthogonal to the others: IDS evaluation. In other words, researchers have not yet succeeded to agree on a general systematic methodology and/or a set of metrics to fairly evaluate different IDS algorithms. This leads to another problem: the lack of an appropriate IDS evaluation dataset that satisfies the common research needs. One major contribution in this area is the DARPA dataset offered by the Massachusetts Institute of Technology Lincoln Lab (MIT/LL), which has been extensively used to evaluate a number of IDS algorithms proposed in the literature. Despite this, the DARPA dataset received a lot of criticism concerning the way it was designed, especially concerning its obsoleteness and inability to incorporate new sorts of network attacks. In this thesis, we survey previous research projects that attempted to provide a system for IDS offline evaluation. From the survey, we identify a set of design requirements for such a system based on the research community needs. We, then, propose Algorizmi as an open-source configurable virtual testbed for generating datasets for offline IDS evaluation. We provide an architectural overview of Algorizmi and its software and hardware components. Algorizmi provides its users with tools that allow them to create their own experimental testbed using the concepts of virtualization and cloud computing. Algorizmi users can configure the virtual machine instances running in their experiments, select what background traffic those instances will generate and what attacks will be launched against them. At any point in time, an Algorizmi user can generate a dataset (network traffic trace) for any of her experiments so that she can use this dataset afterwards to evaluate an IDS the same way the DARPA dataset is used. Our analysis shows that Algorizmi satisfies more requirements than previous research projects that target the same research problem of generating datasets for IDS offline evaluation. Finally, we prove the utility of Algorizmi by building a sample network of machines, generate both background and attack traffic within that network. We then download a snapshot of the dataset for that experiment and run it against Snort IDS. Snort successfully detected the attacks we launched against the sample network. Additionally, we evaluate the performance of Algorizmi while processing some of the common usages of a typical user based on 5 metrics: CPU time, CPU usage, memory usage, network traffic sent/received and the execution time.