Optimal Attack Path Generation Based on Supervised Kohonen Neural Network

Attack graphs are important tools for analyzing security vulnerabilities in enterprise networks. Previous work on attack graphs has not provided an account of the scalability of the graph generating process, and there is often a lack of logical formalism in the representation of attack graphs, which results in the attack graph being difficult to use and understand by human beings. Pioneer work by Sheyner, et al. is the first attack-graph tool based on formal logical techniques, namely model-checking. However, when applied to moderate-sized networks, Sheyner's tool encountered a significant exponential explosion problem. This paper describes a new approach to represent and generate attack graphs. We propose logical attack graphs, which directly illustrate logical dependencies among attack goals and configuration information. A logical attack graph always has size polynomial to the network being analyzed. Our attack graph generation tool builds upon MulVAL, a network security analyzer based on logical programming. We demonstrate how to produce a derivation trace in the MulVAL logic-programming engine, and how to use the trace to generate a logical attack graph in quadratic time. We show experimental evidence that our logical attack graph generation algorithm is very efficient. We have generated logical attack graphs for fully connected networks of 1000 machines using a Pentium 4 CPU with 1GB of RAM.

An attack graph models the causal relationships between vulnerabilities. Attack graphs have important applications in protecting critical resources in networks against sophisticated multi-step intrusions. Currently, analyses of attack graphs largely depend on proprietary implementations of specialized algorithms. However, developing and implementing algorithms causes a delay to the availability of new analyses. The delay is usually unacceptable due to rapidly-changing needs in defending against network intrusions. An administrator may want to revise an analysis as soon as its outcome is observed. Such an interactive analysis, similar to that in decision support systems, is desirable but difficult with current approaches based on proprietary implementations of algorithms. This paper addresses the above issue through a relational approach. Specifically, we devise a relational model for representing necessary inputs, such as network configurations and domain knowledge, and we generate attack graphs from these inputs as relational views. We show that typical analyses can be supported through different type of searches in an attack graph, and these searches can be realized as relational queries. Our approach eliminates the needs for implementing algorithms, because an analysis is now simply a relational query. The interactive analysis of attack graphs becomes possible, since relational queries can be dynamically constructed and revised at run time. As a side effect, experimental results show that the mature optimization techniques in relational databases can transparently improve the performance of the analysis.

Attack Graphs have been widely used by the network security administrators to gain an understanding of possible attack paths, an attacker may follow to compromise critical resources. As networks get larger and more complex, one needs to use databases to perform iterative, interactive analysis tasks with attack graphs. In this paper we investigate how property graph based graph databases like Neo4j can be effectively used towards this application. We show that extending the standard property graph model with a constraint specification mechanism aids attack graph modeling and interactive analytics. We briefly sketch a preliminary attempt to implement a constraint layer over Neo4j and show how attack graph generation and analysis can be performed faithfully in Neo4j using the extended model.

Parallel Optical Burst Switching (POBS) is a variant of Optical Burst Switching (OBS) which is proposed as a new optical switching strategy for Ultra-Dense Wavelength Division Multiplexing (U-DWDM) to support the enormous bandwidth demand of the next generation Internet. As opposed to OBS, POBS transmits bursts in two dimensions: the wavelength dimension and the time dimension. POBS network uses an one-way resource reservation mechanism to set up the resources for each data burst transmission. The use of this mechanism may cause bursts to contend for the same resources at the same time at core (intermediate) nodes of the network. Therefore, the performance of POBS networks depends on the contention resolution policies to reach acceptable levels for bandwidth usage. These policies may increase both the cost and complexities of the core nodes in POBS networks. Most literatures on OBS networks apply contention resolution at the core nodes based on reactive strategies that are activated after contention takes place in core nodes. This study proposes the use of a proactive contention resolution technique at ingress nodes of POBS network as well as reactive contention resolution technique at core nodes for reducing the probability of burst drop in the network in order to increase the performance of the network. The simulation results show that the use of POBS network with the proposed Reactive Odd/Even Node ID Wavelength Assignment Technique (POBS-ROENIDWAT) shows a better performance in terms of reduced data loss rate and increased throughput compared to the performances of both POBS network with Sequential Wavelength Assignment Technique (POBS-SWAT) and OBS network with Burst Segmentation (OBS-BS).

To address the scalability problem in attack graphs generation, we propose a novel method to generate attack graphs automatically. Our approach constructs a two- tier attack graph framework, which includes a host access graph and some sub-attack graphs. A sub-attack graph describes concrete attack scenarios from one source host to one target host, while the host access graph describes the attacker’s privilege transition among hosts. Our sub-attack graphs and host access graph have remarkable smaller scales and can help network administrators to find the key hosts in attack sequences. Analysis shows that the upper bound computational cost of our model is O(N3), which could also be competed in real time. The following experiment validates our approach.

As computer networks continue to grow, it becomes increasingly more important to automate the process of evaluating their vulnerability to attacks. Despite the best efforts of software architects and developers, network hosts inevitably contain a number of vulnerabilities. Hence, it is not feasible for a network administrator to remove all vulnerabilities present in the network hosts. Therefore, the recent focus in security of such networks is on analysis of vulnerabilities globally, finding exploits that are more critical, and preventing them to thwart an intruder. When evaluating the security of a network, it is rarely enough to consider the presence or absence of isolated vulnerabilities. This is because intruders often combine exploits against multiple vulnerabilities in order to reach their goals (Abadi & Jalili, 2005). For example, an intruder might exploit the vulnerability of a particular version of FTP to overwrite the .rhosts file on a victim host. In the next step, the intruder could remotely log in to the victim. In a subsequent step, the intruder could use the victim host as a base to launch another exploit on a new victim, and so on. (Phillips & Swiler, 1998) proposed the concept of attack graphs, where each node represents a possible attack state. Edges represent a change of state caused by a single action taken by the intruder. (Sheyner et al., 2002) used a modified version of the model checker NuSMV (NuSMV, 2010) to produce attack graphs. (Ammann et al., 2002) introduced a monotonicity assumption and used it to develop a polynomial algorithm to encode all of the edges in an attack graph without actually computing the graph itself. These attack graphs are essentially similar to (Phillips & Swiler, 1998), where any path in the graph from an initial node to a goal node shows a sequence of exploits that an intruder can launch to reach his goal. (Noel et al., 2005) presented a number of techniques for managing attack graph complexity through visualization. (Mehta et al., 2006) presented a ranking scheme for the nodes of an attack graph. Rank of a node shows its importance based on factors like the probability of an intruder reaching that node. Given a ranked attack graph, the system administrator can concentrate on relevant subgraphs to figure out how to start deploying security measures. (Ou et al., 2006) presented logical attack graphs, which directly illustrate logical dependencies among attack goals and configuration information. Their attack graph generation tool builds upon MulVAL (Ou et al., 2005), a network security analyzer based on logical programming. The aim of minimization analysis of network attack graphs is to find a minimum critical set of exploits that completely disconnect the initial nodes and the goal nodes of the graph.

The ever-evolving capabilities of cyber attackers force security\nadministrators to focus on the early identification of emerging threats.\nTargeted cyber attacks usually consist of several phases, from initial\nreconnaissance of the network environment to final impact on objectives. This\npaper investigates the identification of multi-step cyber threat scenarios\nusing kill chain and attack graphs. Kill chain and attack graphs are threat\nmodeling concepts that enable determining weak security defense points. We\npropose a novel kill chain attack graph that merges kill chain and attack\ngraphs together. This approach determines possible chains of attacker's actions\nand their materialization within the protected network. The graph generation\nuses a categorization of threats according to violated security properties. The\ngraph allows determining the kill chain phase the administrator should focus on\nand applicable countermeasures to mitigate possible cyber threats. We\nimplemented the proposed approach for a predefined range of cyber threats,\nespecially vulnerability exploitation and network threats. The approach was\nvalidated on a real-world use case. Publicly available implementation contains\na proof-of-concept kill chain attack graph generator.\n

An attack graph is a succinct representation of all paths through a system that end in a state where an intruder has successfully achieved his goal. Today Red Teams determine the vulnerability of networked systems by drawing gigantic attack graphs by hand. Constructing attack graphs by hand is tedious, error-prone, and impractical for large systems. By viewing an attack as a violation of a safety property, we can use off-the-shelf model checking technology to produce attack graphs automatically: a successful path from the intruder's viewpoint is a counterexample produced by the model checker In this paper we present an algorithm for generating attack graphs using model checking as a subroutine. Security analysts use attack graphs for detection, defense and forensics. In this paper we present a minimization analysis technique that allows analysts to decide which minimal set of security measures would guarantee the safety of the system. We provide a formal characterization of this problem: we prove that it is polynomially equivalent to the minimum hitting set problem and we present a greedy algorithm with provable bounds. We also present a reliability analysis technique that allows analysts to perform a simple cost-benefit trade-off depending on the likelihoods of attacks. By interpreting attack graphs as Markov Decision Processes we can use the value iteration algorithm to compute the probabilities of intruder success for each attack the graph.

Path planning is a path used in guiding Unmanned Aerial Vehicles (UAVs), the three dimensional optimal path is calculated by A-star and Dijkstra algorithms proposed in the three dimensional environments. Simulation environments consist of a hypothetical environment and a realistic environment and including static obstacles or the shapes of modern buildings. This paper is to find the shortest path and avert obstacles to from the starting off node to the landing node and finding the shortest optimal path on the network nodes using algorithms proposed through the formation of a three dimensional network by connecting link nodes with adjacent starting off node and landing node so that robots can perform their tasks as quickly possible in the environment. The simulation environment has been established to evaluate the performance of the A-star and Dijkstra algorithms. The results proved that A-star algorithm outperformed Dijkstra’s algorithm in the execution time factor in finding the optimal path only, but they are similar in distance factor and passe nodes in the optimal path in all cases considered.

Cyber deception has great potential in thwarting cyberattacks [1, 4, 8]. A defender (e.g., network administrator) can use deceptive cyber artifacts such as honeypots and faking services to confuse attackers (e.g., hackers) and thus reduce the success rate and effectiveness of attacks. However, the attackers are often diverse and some attackers may be aware of the deception and adapt to the defender's strategy. Therefore, when the defender designs the deception strategy, he needs to take into account the attacker's strategic response. Game theory is suitable for such strategic interaction between the defender and the attacker. Building upon our previous work on security games,[2, 3] we developed a series of game-theoretic models for cyber deception, as well as algorithms to compute the equilibrium or the optimal defender strategy in the games. The first model we proposed is the Cyber Deception Game, a zero-sum Stackelberg game between the defender and an adversary. In this game, the defender is tasked to protect a set of targets, where each target corresponds to a system or a node in a network. Each target has a true configuration, which consists of a set of attributes, e.g., an operating system, services hosted, etc. The defender can choose an observed configuration for each target when responding to probes and scans that may be launched by the attacker. The observed configuration can be different from the true configuration and thus the deception. The attacker, after collecting information about the observed configuration, chooses which target to attack. We show that finding the defender's optimal strategy is NP-hard and provides mixed-integer linear programming (MILP)-based algorithms to compute the optimal deception scheme, and also several heuristic algorithms. In a follow-up work,[7] we extended this game model to a general-sum one, which captures the fact that the cost for the defender may not always be equal to the gain of the attacker. The general-sum nature of this new game model leads to new computational challenges. We provided a Fully Polynomial Time Approximation Scheme (FPTAS) for solving the game and designed a MILP-based algorithm together with several techniques to speed up the computation. In a recent work,[5] we proposed a new attack graph-based Stackelberg security game model and analyze the optimal deception strategy the defender can use. In contrast to previous models, the attacker in this new game can take sequential actions to reach the targets, which is modeled as taking a path on the attack graph. The defender can strategically manipulate the attack graph through deceptive actions in addition to allocating defensive resources to protect important targets from attackers. We provided a MILP-based solution for a special class of attack graphs and a neural architecture search-based method for general directed acyclic attack graphs. We empirically demonstrated the benefit of deception in all these game models and the scalability of the algorithms. This talk features an introduction to these models and algorithms, together with a discussion on future research directions for game theory-based cyber deception.

Attack Graphs are an important support for assessment and subsequent improvement of network security. They reveal possible paths an attacker can take to break through security perimeters and traverse a network to reach valuable assets deep inside the network. Although scalability is no longer the main issue, Attack Graphs still have some problems that make them less useful in practice. First, Attack Graphs remain difficult to relate to the network topology. Second, Attack Graphs traditionally only consider the exploitation of vulnerable hosts. Third, Attack Graphs do not rely on automatic identification of potential attack targets. We address these gaps in our MsAMS (Multi-step Attack Modelling and Simulation) tool, based on Mobile Ambients. The tool not only allows the modelling of more static aspects of the network, such as the network topology, but also the dynamics of network attacks. In addition to Mobile Ambients, we use the PageRank algorithm to determine targets and hub scores produced by the HITS (Hypertext Induced Topic Search) algorithm to guide the simulation of an attacker searching for targets.

Reinforcement learning (RL), in conjunction with attack graphs and cyber terrain, are used to develop reward and state associated with determination of optimal paths for exfiltration of data in enterprise networks. This work builds on previous crown jewels (CJ) identification that focused on the target goal of computing optimal paths that adversaries may traverse toward compromising CJs or hosts within their proximity. This work inverts the previous CJ approach based on the assumption that data has been stolen and now must be quietly exfiltrated from the network. RL is utilized to support the development of a reward function based on the identification of those paths where adversaries desire reduced detection. Results demonstrate promising performance for a sizable network environment.

Attack graph plays an important role in network security, as it directly shows the existence of vulnerabilities in network and how attackers use these vulnerabilities to implement an effective attack, the analysis on the attack graph or the simulation of dynamic attacks through attack graph can help us easily find out the vulnerabilities in network, and take corresponding security measures, in order to strengthen network security. Previous attack graph generation methods are generally not suitable for large network, because of their high complexity of time, high consumption of space, and the large scale of attack graphs. Based on substantive analysis of the vulnerabilities in network, this paper describes a model for automatically generating and analyzing network attack graph. Besides, a prototype system bases on this model has been designed. At last, this prototype system was tested by a model network we built, and it was proved to be simple, flexible, and efficient.

An attack graph is a method used to enumerate the possible paths that an attacker can take in the organizational network. MulVAL is a known open-source framework used to automatically generate attack graphs. MulVAL's default modeling has two main shortcomings. First, it lacks the ability to represent network protocol vulnerabilities, and thus it cannot be used to model common network attacks, such as ARP poisoning. Second, it does not support advanced types of communication, such as wireless and bus communication, and thus it cannot be used to model cyber-attacks on networks that include IoT devices or industrial components. In this article, we present an extended network security model for MulVAL that: (1) considers the physical network topology, (2) supports short-range communication protocols, (3) models vulnerabilities in the design of network protocols, and (4) models specific industrial communication architectures. Using the proposed extensions, we were able to model multiple attack techniques including: spoofing, man-in-the-middle, and denial of service attacks, as well as attacks on advanced types of communication. We demonstrate the proposed model in a testbed which implements a simplified network architecture comprised of both IT and industrial components.

: Report developed under SBIR contract for Topic No. N03-T010. In an ideal world, system and network administrators would keep all of their systems fully patched all the time. Unfortunately, for a variety of reasons few sites have the luxury of such an approach. Given that administrators cannot apply all possible countermeasures (e.g., patches), our next best strategy is to identify an ordering of countermeasures that will provide the optimal level of security for a given amount of changes to the network. One of the primary approaches to identify this optimized ordering is through attack graph analysis, and this approach has received considerable attention recently. Unfortunately, there are a number of critical problems with this strategy. This paper presents an approach that takes advantage of simple filtering capabilities in commodity routers that partially addresses the limitation of the attack graph strategy. Furthermore, in at least one problem, dealing with unknown vulnerabilities, the simple strategy presented in this paper can produce better results than the attack graph approach.