Using Code Bloat to Obfuscate Evolved Network

Size: px
Start display at page:

Download "Using Code Bloat to Obfuscate Evolved Network"

Transcription

1 Using Code Bloat to Obfuscate Evolved Network Traffic Patrick LaRoche, Nur Zincir-Heywood, and Malcolm I. Heywood Faculty of Computer Science, Dalhousie University Halifax, Nova Scotia, Canada Abstract. In this work, we investigate the ability of genetic programming techniques to evolve valid network patterns, while avoiding detectability by obfuscating the intent of the traffic. In order to validate our system s capabilities, we choose to evolve a port scan attack while running the packets through an Intrusion Detection System (IDS). In turn, the evolutionary process uses feedback such that it minimizes the alarms raised while port scanning across a network range. Results build off of previous work allow us to further analyze and understand what the role of introns, code bloat, play in the systems ability to reduce the detectability of it malicious behaviour. 1 Introduction Vulnerability testing, a mechanism in which one attempts to discover weaknesses in a security mechanism, is done both on the part of the attackers and the protectors. Fuzzing, a form of vulnerability testing where one explores the boundaries and variations of a systems designed input parameters, allows the fuzzer to discover the weaknesses of the targeted system. In this case, the aim of the fuzzer is to analyze the vulnerability and attempt to remedy the situation before it becomes exploited. Such an exploration is inherently limited by the breadth and depth of the expert s knowledge in the system. In this work, we apply paged based Linear Genetic Programming (GP) [1, 2] techniques in order to discover new variants of known vulnerabilities. By harnessing the natural exploratory nature of GP techniques, our work aims to start with base domain knowledge, and allow the GP to explore outside the realm of what an expert might conceive. In the work described here, we examine how GP learns to take advantage of introns (code bloat) to hide its desired network behaviour, reducing detectability. Also of importance, we look at the competition during evolution between attacking and remaining undetectable. By employing GP techniques, we are allowing for the discovery of mechanisms in which one can obfuscate the real intent and avoid detection. Previous work has shown the validity of evolutionary techniques by exploring mimicry attacks at the system call level against host based IDSs [3]. The mimicry aspect is that the actual attack consists of not only the core sequence,

2 2 Patrick LaRoche, Nur Zincir-Heywood, and Malcolm I. Heywood but also functionality surrounding it, which is used to obfuscate that actual attack. In [4], authors implemented a preliminary work towards port scans. In this work, we explore attacks at the network level that require the generation of network packets. As a validation of our technique, we have chosen to mimic a port scan attack against a block of hosts and ports, which are protected by the Snort IDS [5]. We focus on analyzing how GP achieves the attack, not the framework of the attack itself, which was analyzed in [4]. To this end, the proposed system evolves proper TCP/IP packets, including header values, such that it not only succeeds in creating port scan attacks, but also remain stealthy and evades a well know IDS, Snort [5]. The rest of this paper is organized as follows; Section 2 and 3 describe background information on both the networking and evolutionary methods, respectively. Section 4 details the goals and the design of the experiments and presents their results. Finally, conclusions are drawn and future work is discussed in Section 5. 2 Background This paper explores the use of code bloat for vulnerability testing. To achieve this, we employe GP in order to evolve malicious network behaviour by evolving TCP/IP network packets. The resulting network traffic is then used to perform fuzzing at the transport level. This allows us to investigate the behaviour of the individual solutions, comparing to known attack behaviours. 2.1 Port Scans The prevalence of port scanning, a method many employ to ascertain their next victims [6] continues to be on the rise [7]. Due to this increase, and the direct correlation with network attacks [8], many detection mechanisms have been developed. The basic detection mechanism for port scanning logs the number of packets X sent to Y host ports in time T. This establishes a threshold that has a temporal dependency, counting the number of events in a specific time frame. Snort, which is an open source IDS, implements such a mechanism. One can classify port scans into 4 generic models, in all cases the attacker is attempting to detect whether a port is open, closed, or managed. In our work we allow the GP to evolve solutions that fit into any one of these models. A high level summary of these models is given below: 1. Basic - The attacker scans a series of ports sequentially, attempting a full TCP connection. 2. Random - The attacker behaves in the same manner as the basic attack, except randomizing the port numbers scanned. 3. Stealth SYN- Instead of trying to establish a full TCP connection, the attacker sends a TCP packet with the SYN flag set. This attack is more difficult to detect due to not initiating a full TCP connection, it simply starts one.

3 Using Code Bloat to Obfuscate Network Traffic 3 4. NULL, FIN, or Xmas scans- Similar to the stealth SYN scan, this attack does not initiate a TCP connection, sends packets with any combination of flags set, looking for RST packet to be returned (or not). 2.2 TCP/IP Packets The intent of a port scan is to determine whether a port is open, closed or filtered. As such, we have selected the following TCP/IP features in order to control source and destination addresses and ports, as well as what control bits are set (also known as flags). All remaining values are calculated and implemented in a manner such that the packet fits requirements of the corresponding RFCs, 791 and 793. We use; IP Addresses, Port # (both source and destination) and Control Bits. It is the above feature set, that GP evolves in creating packets. This provides the appropriate tool set in order to create useful packets for port scanning, but does not exclude the ability to create non-sensical packets as well. We recognize that some combinations of these features would result in a packet that would not make it to its destination, for this work we still consider this a valid packet. 2.3 Code Bloat Code bloat, the rapid growth of Genetic Programs, without relationship to fitness, is well documented in literature [9,10]. Much work has been done on solving this growth issue [11]. Typically code bloat, in the form of introns or repeated sequences of operations that do not contribute to the overall solution, are considered to be extraneous and trimmed out either during the evolutionary process, or after the final generation, as they result in a solution that performs the desired task, but in an inefficient manner. However, in the realm of malicious network traffic, one could argue that the ability to run code that slows down the processing of the solution individual, might in fact not be a negative, but a positive quality. In our work, we focus on packets with temporal malicious behaviour, so introns that lead to additional computational time could perhaps be a benefit for avoiding detection by obfuscating the actual intent. 3 Evolutionary Model Using the knowledge gained from past works [3, 12], authors have developed a system that allows the GP to create valid TCP/IP packets while responding to feedback from Snort [4]. As such, a multi-objective evolutionary model is defined to evolve individual solutions that create valid network packets, send said packets such that a port scan is completed, and avoid detection. These objectives are defined as follows: Solutions determine which ports are available on a range of hosts Solutions raise a minimal number of Snort Alarms

4 4 Patrick LaRoche, Nur Zincir-Heywood, and Malcolm I. Heywood These two competing objectives are then summed into a scalar value. Thus the fitness evaluation not only assists us to rank the evolved attacks relative to the core attack but also provides feedback in terms of Snort s reaction. 3.1 Instruction Set In selecting an instruction set, we must provide the mechanisms in which our system will be successful in recreating the core attack, but also be able to do so in a way that remains undetected. In our work, the attack consists of a series of packets. As described in Section 2, we are concerned with specific fields within a TCP/IP packet, as such we define an instruction set that allows our model to create valid TCP/IP packets, table 1 lists this set. Table 1. Instruction Set for GP Function Parameter 1 Parameter 2 set flags Decimal 0-15 N/A set ips Source IP Destination IP set ports Source Port Destination Port send packet N/A N/A An example of creating a packet would be to call each function call once, in any order, with the last function being send packet, lines 1-4 in Fig 3.1. Each function call is considered a tick, thus we do not give the model an explicit NOP (No OPeration) or wait function in which to use up time. However, it is plausible that a series of functions not involving a send packet could exist to use up ticks (obfuscation), effectively adding a temporal component. An example of a code bloat scenario, Fig. 1, lines 5 and 6 show how the second call overrides the first before sending the packet in line 7. If the intent of the individual was to simply send the packet, then the first system call would be trimmed from the individual. However, in our case, we wish to harness these introns such that they allow the individual to either use up time (by using up processor cycles without sending a packet) or send multiple packets to non target hosts, shown in lines This is a good example of how code bloat obfuscates the intent of the attacker (for this example, assume is the target, i.e the victim). 3.2 Fitness Evaluation In order to assign a fitness metric to our individuals performance, we need to define what a successful port scan is, as well as incorporating feedback from the

5 Figure 1 Example Sequence 1: set flags(0) 2: set ips( , ) 3: set ports(80,22) 4: send packet() 5: set ports(1024,20) 6: set ports(22,31) 7: send packet() 8: set ips( , ) 9: send packet() 10: set ips( , ) 11: send packet() 12: set ips( , ) 13: send packet() Using Code Bloat to Obfuscate Network Traffic 5 detector in order to minimize alarms raised. For the purpose of this work, we consider the scan to be successful if a specific destination host:port pair is found to be open during the scanning process. The specific details of which port and which host varies on the experiment, as described in Section 4. In order to provide useful feedback to the fitness evaluation we have developed a metric that defines how close IP addresses are to each other. Our assumption when assigning this metric is that one considers the proximity of two IP addresses based on what class network they are on, and how similar those networks are to each other. A IPv4 address consists of 4 octets expressed in decimal with period as the delimiter between them. This describes a 32 bit address space, allowing for 4, 294, 967, 296(2 32 ) possible network address. For our testing we evaluate the closeness by looking from the left most octet to the right and measuring the difference between each octet, favouring closeness in the left most octet (which represents the network address) versus the right most octet (which represents host address). Eq. 1 describes this, where the distance is between two IP addresses, A.B.C.D and a.b.c.d. The result of this distance metric is that GP is encouraged to find IP addresses that have open ports and are on the same network. We then use this metric in computing the fitness of an individual, that is to say that an individual is rewarded for finding targets close to actual targets. This encourages exploration of similar networks, allowing for eventual discovery of targets. DistanceIP = ((A a) 4) + ((B b) 3) + ((C c) 2) + (D d) (1) The second part of our fitness function, the feedback from the detector - Snort, is provided in real time as the solutions are evolving. The feedback is offered in the form of an alarm count, where the worst case scenario is the number of alarms greater than or equal to the number of packets sent during the scan. Similarly, the best case is zero alarms being triggered during the execution of an individual solution.

6 6 Patrick LaRoche, Nur Zincir-Heywood, and Malcolm I. Heywood Thus, fitness function is then described as the combination of two metrics (Eq. 1 and 2), with each section being described as a percentage. In this case, the first objective is maximizing the percentage of packets sent that correspond to a targeted open host:port pair, whereas the second objective is the inverse of the percentage of alarms raised by all packets sent, as shown in Eq. 2. F it = ( ) #T argetip : P orthit T otalt argetip : P ort + #Alarms 1 T otalp acketssent Where # Target IP: Port Hit describes the number of target host:port pairs detected and Total Target IP: Port is the total number of open host:port pairs that were available for detection during the training process. # Alarms is the number of alarms triggered by Snort during the execution of the individual, and Total Packets Sent represents the total number of complete packets sent. Note that the size of the individual is not the same as the number of packets sent, as it requires at least four function calls to send one packet. (2) 3.3 Evolutionary Model In order to achieve our goals we have implemented a page based linear GP. The basis for such a representation is a (virtual) register machine [1, 2, 13]. Such a machine consists of a predefined set of general purpose registers, support for the execution of an instruction set and the ability to supply input features describing the state of the environment at any given time step. Thus, individuals in the population of possible solutions are defined using a fixed length representation, with each individual representing a linearly sequential set of instruction calls. Selection of individuals between each generation is performed using a steady state tournament over 4 individuals. Search operators are applied to all in the tournament and the resulting two best performers replace the worst two. The search operators employed in this work are: Crossover, single point: A single page at a location (chosen with uniform probability) in one parent is interchanged with the page located at the same position in other parent. Swap Selector: Two instructions are selected in an individual and swapped, individual length remains fixed. Instruction-wise Mutation: For each instruction, test for application of mutation operator. If mutation is to be applied, another instruction from the complete instruction set is chosen with uniform probability and used as replacement. In this work, we have employed instruction-wise mutation. With the linear annealing schedule, the probability of mutation linearly decreases over the generation count, reducing it to zero for the last iteration. Table 2 lists the specific evolutionary model parameters used during our experiments.

7 Using Code Bloat to Obfuscate Network Traffic 7 Table 2. Parameters for Evolutionary Model Parameter Value Parameter Value Population 1000 Mutation 0.5 with linear decay Page Count 100 Swap 0.5 Page Size 6 Crossover 0.9 Tournament Size 4 Stop Criteria Tournaments 4 Experiments and Results In these experiments, we explore the relationship between the available search space and the size of the target list, where the search space consists of all possible IP addresses while the target list varies in size, building off of the previous work s success. The experiments consist of three separate target lists, only used as a validation to the fitness of an individual, versus the entire IPv4 range. The first experiment consists of 10 targets, the second 100 and the final experiment being 1000 hosts. All three experiments are run 30 times using a different initial seed. Thus, we run 90 experiments in total. Experiments are performed such that there is a machine originating the attacks (running the GP) and a machine acting as a monitor for traffic generated by the attack machine. Snort is is deployed on the monitor, seeing all the traffic. Snort was configured such that all packets received are monitored and inspected for malicious behaviour. As such this testbed mimics the attacker and the targeted machines in a controlled manner, allowing us to examine alarms triggered solely by our work. It should be noted here that we use the default Snort configuration, and set pre-processor specific variables for the most robust port scan detection possible. Specifically, the flow-portscan pre-processor was enabled, and the sfportscan preprocessor was set to the highest scan level available. To assign a metric value to the performance of our system we rank our fitness out of 100 (perfect). A perfect solution would be a solution that sent a packet to each target machine without triggering any alarms. Also of importance, in order to realize the usage of code bloat towards obfuscating network traffic. This is measured by comparing the number of function calls (total size) versus the number of packets sent to the network, thus showing what percentage of an individual results in a packet being sent. We list the best solution as well as the average of each target list in Table 3. Also of interest is the tradeoff between the two competing metrics we use in our fitness evaluation, namely the metric of being able to find targets (open ports on hosts) versus staying stealthy. Moreover, how this develops over the evolutionary process is also important to understand the effect of code bloat in our experiments. Fig 2 presents the two components (finding targets, remaining stealthy) of the fitness for all experiments. By examining the fitness evaluation of each objective (component of the fitness function) separately, we can see how they evolve over each generation. Thus, Fig. 2 shows that the exploratory target objective evolves slowly at a linear rate, meanwhile the stealthy objective

8 8 Patrick LaRoche, Nur Zincir-Heywood, and Malcolm I. Heywood Table 3. Results 10 Targets 100 Targets 1000 Targets Best Solution s Fitness Best Solution s Length Best Solution s Packets Mean Population Solution Fitness Mean Population Solution Length Mean Population Packets Send evolves exponentially. This difference in learning rates has led to two further experiments where we modified the fitness function and re-ran the 100 Target experiments again. In this case, the fitness function was altered such that in one set of experiments once the IDS fitness component reached 90% we started tracking the target fitness (results shown as as Alt #2). Alt #1 shows the other additional experiment in which the Target fitness was weighted more (2 to 3) then the IDS fitness to encourage more improvement on the seemingly more difficult objective. Average results of each experiment at a given generation count are shown in Fig 2. Fig. 2. Fitness over Generation Count In both additional experiments we see an improvement in the rate in which the target fitness is learnt, with the alternative version #1 showing an actual non linear improvement after approximately generations. Further work in this area needs to be performed, but some promise is shown. 4.1 Analysis The system implemented in this work needed to concur two problems concurrently, scan a number of unknown targets to find a small subset that had available

9 Using Code Bloat to Obfuscate Network Traffic 9 services while also remaining stealthy in its activity. The system used feedback from the IDS (Snort) as well as distance metrics to determine how close it was to finding the available targets. In analyzing our results the system does indeed learn to achieve this by achieving the best solution fitness values of 91.57%, 87.39% and 85.47% in the three sets of experiments, Table 3. In looking at the size of the solutions as well as the breakdown of the fitness metrics, we see that the system is learning which problem is easier to solve: (i) finding all the targets, or (ii) avoiding detection. In all three sets of experiments, we can see that the avoiding of detection is the simpler problem to solve, Fig 2. The apparent difficulty of the problem can also be seen in the exponential increase in fitness with regard to detection avoidance versus the slow linear growth of the fitness associated with finding the targets. We explored this further by implementing two additional sets of experiments where we modified the relationship between the two objectives, first by artificially capping the IDS fitness at 90% until the Target Fitness improves, then by weighting the Target Fitness more then the IDS fitness. In both experiments, we show that we can further improve our results to best solution fitness values of 92.8% and 93.2% respectively, Fig 2. As stated earlier, the underlying motivation of our work is to examine the GP s ability to obfuscate it s core network intent using code bloat. To this end, we observe that the GP focuses on sending many packets, to many sources, while using a smaller amount of genes to change other packet header values. This is also an appropriate method for avoiding detection, as the GP is sending many packets, to many sources, raising no flags in Snort (as they are not being sent all to a specific IP range or port addresses, Snort does not flag this as an attack). 5 Conclusion and Future Work We have demonstrated the ability for GP to generate network traffic, while adding complexity to the problem by increasing the search space to real-life scenarios. To achieve this, we modified the fitness evaluation of [4] by adding a distance metric between targeted IP addresses, as well as allowing the GP to explore the complete IP address space. We performed 90 experiments with the new fitness function, where the malicious user would locate a small, medium, and large number of targets during the scanning process while remaining stealthy in its activity. Then, the ability for GP is explored to learn how it obfuscates the network traffic through the use of code bloat. The resulting solutions did not use code bloat in methods we had expected. Instead, solutions were focusing on maximizing packets sent by an individual to a large set of targets, probably due to the larger search space required in these experiments. However, the solutions remain stealthy in their activities. On one hand it is not surprising that finding 10 target hosts out of the full IPv4 range is difficult, but on the other hand we did not expect the stealthy component to be relatively easier to solve. In future work, we plan to further explore this apparent disparity in order to improve our system while also applying it to further network activities.

10 10 Patrick LaRoche, Nur Zincir-Heywood, and Malcolm I. Heywood 6 Acknowledgment This research is sponsored by the granting agencies NSERC and MITACS as well as the industrial partners TARA Inc. and SwissComm Innovations SA. References 1. Heywood, M.I., Zincir-Heywood, A.N.: Dynamic page based crossover in linear genetic programming. IEEE Transactions on Systems, Man, and Cybernetics: Part B - Cybernetics 32(3) (2002) Nordin, P.: A compiling genetic programming system that directly manipulates the machine code. In Kinnear, Jr., K.E., ed.: Advances in Genetic Programming. MIT Press (1994) Kayacik, H.G., Heywood, M.I., Zincir-Heywood, A.N.: Evolving buffer overflow attacks with detector feedback. In: Applications of Evolutionary Computing, EvoWorkshops2007: EvoCOMNET, EvoFIN, EvoIASP, EvoInteraction, Evo- MUSART, EvoSTOC, EvoTransLog. Volume 4448 of LNCS., Valencia, Spain, EvoStar, Springer Verlag (11-13 April 2007) Patrick LaRoche, Nur Zincir-Heywood, M.H.: Evolving tcp/ip packets: A case study of port scans. In: Second IEEE Symposium on Computational Intelligence for Security and Defense Applications. (2009) 5. Snort.org: Snort ids (March 2009) 6. Ashfaq, A.B., Robert, M.J., Mumtaz, A., Ali, M.Q., Sajjad, A., Khayam, S.A.: A comparative evaluation of anomaly detectors under portscan attacks. In Lippmann, R., Kirda, E., Trachtenberg, A., eds.: RAID. Volume 5230 of Lecture Notes in Computer Science., Springer (2008) Symantec: Global Internet Security Threat Report trends for July December 07. Volume XIII. Symantec (2008) 8. Panjwani, S., Tan, S., Jarrin, K.M.: An experimental evaluation to determine if port scans are precursors to an attack. In: DSN 05: Proceedings of the 2005 International Conference on Dependable Systems and Networks, Washington, DC, USA, IEEE Computer Society (2005) Soule, T., Professor, M., Foster, J.A., Foster, J.A., Alves-foss, J., Frenzel, J.F., Frincke, D., Jacobsen, R.T., Shreeve, J.M.: Code growth in genetic programming (1998) 10. Koza, J.R.: Genetic programming as a means for programming computers by natural selection. Stat. Comput. (UK 4 (1994) Nordin, P., Francone, F., Banzhaf, W.: Explicitly defined introns and destructive crossover in genetic programming (1995) 12. Kayacik, H.G., Heywood, M., Zincir-Heywood, N.: On evolving buffer overflow attacks using genetic programming. In: GECCO 06: Proceedings of the 8th annual conference on Genetic and evolutionary computation, New York, NY, USA, ACM (2006) Huelsbergen, L.: Toward simulated evolution of machine language iteration. In Koza, J.R., Goldberg, D.E., Fogel, D.B., Riolo, R.L., eds.: Genetic Programming 1996: Proceedings of the First Annual Conference, Stanford University, CA, USA, MIT Press ( )

Approach Using Genetic Algorithm for Intrusion Detection System

Approach Using Genetic Algorithm for Intrusion Detection System Approach Using Genetic Algorithm for Intrusion Detection System 544 Abhijeet Karve Government College of Engineering, Aurangabad, Dr. Babasaheb Ambedkar Marathwada University, Aurangabad, Maharashtra-

More information

Can the Best Defense be to Attack?

Can the Best Defense be to Attack? Can the Best Defense be to Attack? MITACS Digital Security Seminar Series at Carleton University Presenter: Dr. Nur Zincir-Heywood Dalhousie University, Faculty of Computer Science Arms Race Security engineers

More information

Evolving SQL Queries for Data Mining

Evolving SQL Queries for Data Mining Evolving SQL Queries for Data Mining Majid Salim and Xin Yao School of Computer Science, The University of Birmingham Edgbaston, Birmingham B15 2TT, UK {msc30mms,x.yao}@cs.bham.ac.uk Abstract. This paper

More information

Evolutionary Computation as an Artificial Attacker

Evolutionary Computation as an Artificial Attacker Noname manuscript No. (will be inserted by the editor) Evolutionary Computation as an Artificial Attacker Generating Evasion Attacks for Detector Vulnerability Testing Hilmi Güneş Kayacık A. Nur Zincir-Heywood

More information

Generating Mimicry Attacks using Genetic Programming: A Benchmarking Study

Generating Mimicry Attacks using Genetic Programming: A Benchmarking Study Generating Mimicry Attacks using Genetic Programming: A Benchmarking Study H. Güneş Kayacık 1, A. Nur Zincir-Heywood 1, Malcolm I. Heywood 1, Stefan Burschka 2 1 Dalhousie University, Faculty of Computer

More information

Investigating the Application of Genetic Programming to Function Approximation

Investigating the Application of Genetic Programming to Function Approximation Investigating the Application of Genetic Programming to Function Approximation Jeremy E. Emch Computer Science Dept. Penn State University University Park, PA 16802 Abstract When analyzing a data set it

More information

Automatic Programming with Ant Colony Optimization

Automatic Programming with Ant Colony Optimization Automatic Programming with Ant Colony Optimization Jennifer Green University of Kent jg9@kent.ac.uk Jacqueline L. Whalley University of Kent J.L.Whalley@kent.ac.uk Colin G. Johnson University of Kent C.G.Johnson@kent.ac.uk

More information

Detecting Specific Threats

Detecting Specific Threats The following topics explain how to use preprocessors in a network analysis policy to detect specific threats: Introduction to Specific Threat Detection, page 1 Back Orifice Detection, page 1 Portscan

More information

Lecture Notes on Critique of 1998 and 1999 DARPA IDS Evaluations

Lecture Notes on Critique of 1998 and 1999 DARPA IDS Evaluations Lecture Notes on Critique of 1998 and 1999 DARPA IDS Evaluations Prateek Saxena March 3 2008 1 The Problems Today s lecture is on the discussion of the critique on 1998 and 1999 DARPA IDS evaluations conducted

More information

Calc Redirection : A Structure for Direction Finding Aided Traffic Monitoring

Calc Redirection : A Structure for Direction Finding Aided Traffic Monitoring Calc Redirection : A Structure for Direction Finding Aided Traffic Monitoring Paparao Sanapathi MVGR College of engineering vizianagaram, AP P. Satheesh, M. Tech,Ph. D MVGR College of engineering vizianagaram,

More information

A LINEAR GENETIC PROGRAMMING APPROACH TO INTRUSTION DETECTION. Dong Song

A LINEAR GENETIC PROGRAMMING APPROACH TO INTRUSTION DETECTION. Dong Song A LINEAR GENETIC PROGRAMMING APPROACH TO INTRUSTION DETECTION by Dong Song Submitted in partial fulfilment of the requirements for the degree of Master of Computer Science at Dalhousie University Halifax,

More information

TESTING A COGNITIVE PACKET CONCEPT ON A LAN

TESTING A COGNITIVE PACKET CONCEPT ON A LAN TESTING A COGNITIVE PACKET CONCEPT ON A LAN X. Hu, A. N. Zincir-Heywood, M. I. Heywood Faculty of Computer Science, Dalhousie University {xhu@cs.dal.ca, zincir@cs.dal.ca, mheywood@cs.dal.ca} Abstract The

More information

Dynamic Page Based Crossover in Linear Genetic Programming

Dynamic Page Based Crossover in Linear Genetic Programming Dynamic Page Based Crossover in Linear Genetic Programming M.I. Heywood, A.N. Zincir-Heywood Abstract. Page-based Linear Genetic Programming (GP) is proposed in which individuals are described in terms

More information

Genetic Programming Part 1

Genetic Programming Part 1 Genetic Programming Part 1 Evolutionary Computation Lecture 11 Thorsten Schnier 06/11/2009 Previous Lecture Multi-objective Optimization Pareto optimality Hyper-volume based indicators Recent lectures

More information

Evolving Successful Stack Overflow Attacks for Vulnerability Testing

Evolving Successful Stack Overflow Attacks for Vulnerability Testing Evolving Successful Stack Overflow Attacks for Vulnerability Testing H. Güneş Kayacık, A. Nur Zincir-Heywood, Malcolm Heywood Dalhousie University, Faculty of Computer Science, 6050 University Avenue,

More information

A Hybrid Approach for Misbehavior Detection in Wireless Ad-Hoc Networks

A Hybrid Approach for Misbehavior Detection in Wireless Ad-Hoc Networks A Hybrid Approach for Misbehavior Detection in Wireless Ad-Hoc Networks S. Balachandran, D. Dasgupta, L. Wang Intelligent Security Systems Research Lab Department of Computer Science The University of

More information

Configuring attack detection and prevention 1

Configuring attack detection and prevention 1 Contents Configuring attack detection and prevention 1 Overview 1 Attacks that the device can prevent 1 Single-packet attacks 1 Scanning attacks 2 Flood attacks 3 TCP fragment attack 4 Login DoS attack

More information

intelop Stealth IPS false Positive

intelop Stealth IPS false Positive There is a wide variety of network traffic. Servers can be using different operating systems, an FTP server application used in the demilitarized zone (DMZ) can be different from the one used in the corporate

More information

Previous Lecture Genetic Programming

Previous Lecture Genetic Programming Genetic Programming Previous Lecture Constraint Handling Penalty Approach Penalize fitness for infeasible solutions, depending on distance from feasible region Balanace between under- and over-penalization

More information

Coevolving Functions in Genetic Programming: Classification using K-nearest-neighbour

Coevolving Functions in Genetic Programming: Classification using K-nearest-neighbour Coevolving Functions in Genetic Programming: Classification using K-nearest-neighbour Manu Ahluwalia Intelligent Computer Systems Centre Faculty of Computer Studies and Mathematics University of the West

More information

Genetic Image Network for Image Classification

Genetic Image Network for Image Classification Genetic Image Network for Image Classification Shinichi Shirakawa, Shiro Nakayama, and Tomoharu Nagao Graduate School of Environment and Information Sciences, Yokohama National University, 79-7, Tokiwadai,

More information

Genetic Programming for Data Classification: Partitioning the Search Space

Genetic Programming for Data Classification: Partitioning the Search Space Genetic Programming for Data Classification: Partitioning the Search Space Jeroen Eggermont jeggermo@liacs.nl Joost N. Kok joost@liacs.nl Walter A. Kosters kosters@liacs.nl ABSTRACT When Genetic Programming

More information

MIRROR SITE ORGANIZATION ON PACKET SWITCHED NETWORKS USING A SOCIAL INSECT METAPHOR

MIRROR SITE ORGANIZATION ON PACKET SWITCHED NETWORKS USING A SOCIAL INSECT METAPHOR MIRROR SITE ORGANIZATION ON PACKET SWITCHED NETWORKS USING A SOCIAL INSECT METAPHOR P. Shi, A. N. Zincir-Heywood and M. I. Heywood Faculty of Computer Science, Dalhousie University, Halifax NS, Canada

More information

Stochastic Analysis of Horizontal IP Scanning

Stochastic Analysis of Horizontal IP Scanning Stochastic Analysis of Horizontal IP Scanning Derek Leonard, Zhongmei Yao,, Xiaoming Wang, and Dmitri Loguinov Internet Research Lab Department of Computer Science and Engineering Texas A&M University

More information

Selecting Features for Intrusion Detection: A Feature Relevance Analysis on KDD 99 Intrusion Detection Datasets

Selecting Features for Intrusion Detection: A Feature Relevance Analysis on KDD 99 Intrusion Detection Datasets Selecting Features for Intrusion Detection: A Feature Relevance Analysis on KDD 99 Intrusion Detection Datasets H. Günes Kayacık, A. Nur Zincir-Heywood, Malcolm I. Heywood Dalhousie University, Faculty

More information

Basic Concepts in Intrusion Detection

Basic Concepts in Intrusion Detection Technology Technical Information Services Security Engineering Roma, L Università Roma Tor Vergata, 23 Aprile 2007 Basic Concepts in Intrusion Detection JOVAN GOLIĆ Outline 2 Introduction Classification

More information

Review on Data Mining Techniques for Intrusion Detection System

Review on Data Mining Techniques for Intrusion Detection System Review on Data Mining Techniques for Intrusion Detection System Sandeep D 1, M. S. Chaudhari 2 Research Scholar, Dept. of Computer Science, P.B.C.E, Nagpur, India 1 HoD, Dept. of Computer Science, P.B.C.E,

More information

Genetic Programming of Autonomous Agents. Functional Requirements List and Performance Specifi cations. Scott O'Dell

Genetic Programming of Autonomous Agents. Functional Requirements List and Performance Specifi cations. Scott O'Dell Genetic Programming of Autonomous Agents Functional Requirements List and Performance Specifi cations Scott O'Dell Advisors: Dr. Joel Schipper and Dr. Arnold Patton November 23, 2010 GPAA 1 Project Goals

More information

ANOMALY DETECTION IN COMMUNICTION NETWORKS

ANOMALY DETECTION IN COMMUNICTION NETWORKS Anomaly Detection Summer School Lecture 2014 ANOMALY DETECTION IN COMMUNICTION NETWORKS Prof. D.J.Parish and Francisco Aparicio-Navarro Loughborough University (School of Electronic, Electrical and Systems

More information

Genetic Programming Prof. Thomas Bäck Nat Evur ol al ut ic o om nar put y Aling go rg it roup hms Genetic Programming 1

Genetic Programming Prof. Thomas Bäck Nat Evur ol al ut ic o om nar put y Aling go rg it roup hms Genetic Programming 1 Genetic Programming Prof. Thomas Bäck Natural Evolutionary Computing Algorithms Group Genetic Programming 1 Genetic programming The idea originated in the 1950s (e.g., Alan Turing) Popularized by J.R.

More information

9. Security. Safeguard Engine. Safeguard Engine Settings

9. Security. Safeguard Engine. Safeguard Engine Settings 9. Security Safeguard Engine Traffic Segmentation Settings Storm Control DoS Attack Prevention Settings Zone Defense Settings SSL Safeguard Engine D-Link s Safeguard Engine is a robust and innovative technology

More information

Escaping Local Optima: Genetic Algorithm

Escaping Local Optima: Genetic Algorithm Artificial Intelligence Escaping Local Optima: Genetic Algorithm Dae-Won Kim School of Computer Science & Engineering Chung-Ang University We re trying to escape local optima To achieve this, we have learned

More information

A Comparative Study of Linear Encoding in Genetic Programming

A Comparative Study of Linear Encoding in Genetic Programming 2011 Ninth International Conference on ICT and Knowledge A Comparative Study of Linear Encoding in Genetic Programming Yuttana Suttasupa, Suppat Rungraungsilp, Suwat Pinyopan, Pravit Wungchusunti, Prabhas

More information

Configuring attack detection and prevention 1

Configuring attack detection and prevention 1 Contents Configuring attack detection and prevention 1 Overview 1 Attacks that the device can prevent 1 Single-packet attacks 1 Scanning attacks 2 Flood attacks 3 TCP fragment attack 4 Login DoS attack

More information

4-2 Rapid Analysis Technologies for Live Networks

4-2 Rapid Analysis Technologies for Live Networks 4 Cyber-Security Technologies: Live Network Monitoring and Analysis Technologies 4-2 Rapid Analysis Technologies for Live Networks Ichiro SHIMADA and Yu TSUDA In targeted cyberattacks, the attackers intrude

More information

Evolutionary form design: the application of genetic algorithmic techniques to computer-aided product design

Evolutionary form design: the application of genetic algorithmic techniques to computer-aided product design Loughborough University Institutional Repository Evolutionary form design: the application of genetic algorithmic techniques to computer-aided product design This item was submitted to Loughborough University's

More information

Genetic Programming in the Wild:

Genetic Programming in the Wild: Genetic Programming in the Wild: and orlovm, sipper@cs.bgu.ac.il Department of Computer Science Ben-Gurion University, Israel GECCO 2009, July 8 12 Montréal, Québec, Canada 1 / 46 GP: Programs or Representations?

More information

Evolutionary Algorithm Approaches for Detecting Computer Network Intrusion (Extended Abstract)

Evolutionary Algorithm Approaches for Detecting Computer Network Intrusion (Extended Abstract) Evolutionary Algorithm Approaches for Detecting Computer Network Intrusion (Extended Abstract) Kevin P. Anchor, Paul D. Williams, Gregg H. Gunsch, and Gary B. Lamont Department of Electrical and Computer

More information

Network Routing Protocol using Genetic Algorithms

Network Routing Protocol using Genetic Algorithms International Journal of Electrical & Computer Sciences IJECS-IJENS Vol:0 No:02 40 Network Routing Protocol using Genetic Algorithms Gihan Nagib and Wahied G. Ali Abstract This paper aims to develop a

More information

Chair for Network Architectures and Services Department of Informatics TU München Prof. Carle. Network Security. Chapter 8

Chair for Network Architectures and Services Department of Informatics TU München Prof. Carle. Network Security. Chapter 8 Chair for Network Architectures and Services Department of Informatics TU München Prof. Carle Network Security Chapter 8 System Vulnerabilities and Denial of Service Attacks System Vulnerabilities and

More information

Selection Based on the Pareto Nondomination Criterion for Controlling Code Growth in Genetic Programming

Selection Based on the Pareto Nondomination Criterion for Controlling Code Growth in Genetic Programming Genetic Programming and Evolvable Machines, 2, 61 73, 2001 2001 Kluwer Academic Publishers. Manufactured in The Netherlands. Selection Based on the Pareto Nondomination Criterion for Controlling Code Growth

More information

CSE 565 Computer Security Fall 2018

CSE 565 Computer Security Fall 2018 CSE 565 Computer Security Fall 2018 Lecture 19: Intrusion Detection Department of Computer Science and Engineering University at Buffalo 1 Lecture Outline Intruders Intrusion detection host-based network-based

More information

Network Discovery Policies

Network Discovery Policies The following topics describe how to create, configure, and manage network discovery policies: Overview:, page 1 Network Discovery Customization, page 2 Network Discovery Rules, page 3 Configuring Advanced

More information

Evolution of the Discrete Cosine Transform Using Genetic Programming

Evolution of the Discrete Cosine Transform Using Genetic Programming Res. Lett. Inf. Math. Sci. (22), 3, 117-125 Available online at http://www.massey.ac.nz/~wwiims/research/letters/ Evolution of the Discrete Cosine Transform Using Genetic Programming Xiang Biao Cui and

More information

Mapping Internet Sensors with Probe Response Attacks

Mapping Internet Sensors with Probe Response Attacks Mapping Internet Sensors with Probe Response Attacks John Bethencourt, Jason Franklin, and Mary Vernon {bethenco, jfrankli, vernon}@cs.wisc.edu Computer Sciences Department University of Wisconsin, Madison

More information

Transforming Security from Defense in Depth to Comprehensive Security Assurance

Transforming Security from Defense in Depth to Comprehensive Security Assurance Transforming Security from Defense in Depth to Comprehensive Security Assurance February 28, 2016 Revision #3 Table of Contents Introduction... 3 The problem: defense in depth is not working... 3 The new

More information

Intrusion Detection Systems Overview

Intrusion Detection Systems Overview Intrusion Detection Systems Overview Chris Figueroa East Carolina University figueroac13@ecu.edu Abstract Modern intrusion detection systems provide a first line of defense against attackers for organizations.

More information

Means for Intrusion Detection. Intrusion Detection. INFO404 - Lecture 13. Content

Means for Intrusion Detection. Intrusion Detection. INFO404 - Lecture 13. Content Intrusion Detection INFO404 - Lecture 13 21.04.2009 nfoukia@infoscience.otago.ac.nz Content Definition Network vs. Host IDS Misuse vs. Behavior Based IDS Means for Intrusion Detection Definitions (1) Intrusion:

More information

Intrusion Detection System with FGA and MLP Algorithm

Intrusion Detection System with FGA and MLP Algorithm Intrusion Detection System with FGA and MLP Algorithm International Journal of Engineering Research & Technology (IJERT) Miss. Madhuri R. Yadav Department Of Computer Engineering Siddhant College Of Engineering,

More information

Authentication System

Authentication System A Biologically Inspired Password Authentication System Dipankar Dasgupta and Sudip Saha Center for Information Assurance University of Memphis Memphis, TN 38152 Outline Motivation Position Authentication

More information

INTRUSION DETECTION SYSTEM USING BIG DATA FRAMEWORK

INTRUSION DETECTION SYSTEM USING BIG DATA FRAMEWORK INTRUSION DETECTION SYSTEM USING BIG DATA FRAMEWORK Abinesh Kamal K. U. and Shiju Sathyadevan Amrita Center for Cyber Security Systems and Networks, Amrita School of Engineering, Amritapuri, Amrita Vishwa

More information

The Genetic Algorithm for finding the maxima of single-variable functions

The Genetic Algorithm for finding the maxima of single-variable functions Research Inventy: International Journal Of Engineering And Science Vol.4, Issue 3(March 2014), PP 46-54 Issn (e): 2278-4721, Issn (p):2319-6483, www.researchinventy.com The Genetic Algorithm for finding

More information

EVALUATIONS OF THE EFFECTIVENESS OF ANOMALY BASED INTRUSION DETECTION SYSTEMS BASED ON AN ADAPTIVE KNN ALGORITHM

EVALUATIONS OF THE EFFECTIVENESS OF ANOMALY BASED INTRUSION DETECTION SYSTEMS BASED ON AN ADAPTIVE KNN ALGORITHM EVALUATIONS OF THE EFFECTIVENESS OF ANOMALY BASED INTRUSION DETECTION SYSTEMS BASED ON AN ADAPTIVE KNN ALGORITHM Assosiate professor, PhD Evgeniya Nikolova, BFU Assosiate professor, PhD Veselina Jecheva,

More information

A Study on Intrusion Detection Techniques in a TCP/IP Environment

A Study on Intrusion Detection Techniques in a TCP/IP Environment A Study on Intrusion Detection Techniques in a TCP/IP Environment C. A. Voglis and S. A. Paschos Department of Computer Science University of Ioannina GREECE Abstract: The TCP/IP protocol suite is the

More information

A Network Intrusion Detection System Architecture Based on Snort and. Computational Intelligence

A Network Intrusion Detection System Architecture Based on Snort and. Computational Intelligence 2nd International Conference on Electronics, Network and Computer Engineering (ICENCE 206) A Network Intrusion Detection System Architecture Based on Snort and Computational Intelligence Tao Liu, a, Da

More information

On the Impact of Class Imbalance in GP Streaming Classification with Label Budgets

On the Impact of Class Imbalance in GP Streaming Classification with Label Budgets On the Impact of Class Imbalance in GP Streaming Classification with Label Budgets Sara Khanchi, Malcolm I. Heywood, and A. Nur Zincir-Heywood Faculty of Computer Science, Dalhousie University, Halifax,

More information

IDS: Signature Detection

IDS: Signature Detection IDS: Signature Detection Idea: What is bad, is known What is not bad, is good Determines whether a sequence of instructions being executed is known to violate the site security policy Signatures: Descriptions

More information

CS395/495 Computer Security Project #2

CS395/495 Computer Security Project #2 CS395/495 Computer Security Project #2 Important Dates Out: 1/19/2005 Due: 2/15/2005 11:59pm Winter 2005 Project Overview Intrusion Detection System (IDS) is a common tool to detect the malicious activity

More information

The Reconnaissance Phase

The Reconnaissance Phase The Reconnaissance Phase Detecting the Enemy Before the Attack Carrie Gates PhD Candidate, Dalhousie University Visiting Scientist, CERT, Carnegie Mellon University Outline! Indicate a gap in our defences!

More information

Overview Intrusion Detection Systems and Practices

Overview Intrusion Detection Systems and Practices Overview Intrusion Detection Systems and Practices Chapter 13 Lecturer: Pei-yih Ting Intrusion Detection Concepts Dealing with Intruders Detecting Intruders Principles of Intrusions and IDS The IDS Taxonomy

More information

Scanning. Course Learning Outcomes for Unit III. Reading Assignment. Unit Lesson UNIT III STUDY GUIDE

Scanning. Course Learning Outcomes for Unit III. Reading Assignment. Unit Lesson UNIT III STUDY GUIDE UNIT III STUDY GUIDE Course Learning Outcomes for Unit III Upon completion of this unit, students should be able to: 1. Recall the terms port scanning, network scanning, and vulnerability scanning. 2.

More information

Adaptive Crossover in Genetic Algorithms Using Statistics Mechanism

Adaptive Crossover in Genetic Algorithms Using Statistics Mechanism in Artificial Life VIII, Standish, Abbass, Bedau (eds)(mit Press) 2002. pp 182 185 1 Adaptive Crossover in Genetic Algorithms Using Statistics Mechanism Shengxiang Yang Department of Mathematics and Computer

More information

Very Fast Containment of Scanning Worms. Nicholas Weaver, Stuart Staniford, Vern Paxson ICSI, Nevis Networks, ICSI & LBNL

Very Fast Containment of Scanning Worms. Nicholas Weaver, Stuart Staniford, Vern Paxson ICSI, Nevis Networks, ICSI & LBNL Very Fast Containment of Scanning Worms Nicholas Weaver, Stuart Staniford, Vern Paxson ICSI, Nevis Networks, ICSI & LBNL 1 Outline Worm Containment Scan Suppression Hardware Implementation Cooperation

More information

Evolutionary Algorithms. CS Evolutionary Algorithms 1

Evolutionary Algorithms. CS Evolutionary Algorithms 1 Evolutionary Algorithms CS 478 - Evolutionary Algorithms 1 Evolutionary Computation/Algorithms Genetic Algorithms l Simulate natural evolution of structures via selection and reproduction, based on performance

More information

Mapping Internet Sensors with Probe Response Attacks

Mapping Internet Sensors with Probe Response Attacks Mapping Internet Sensors with Probe Response Attacks Computer Sciences Department University of Wisconsin, Madison Introduction Outline Background Example Attack Introduction to the Attack Basic Probe

More information

A multi-step attack-correlation method with privacy protection

A multi-step attack-correlation method with privacy protection A multi-step attack-correlation method with privacy protection Research paper A multi-step attack-correlation method with privacy protection ZHANG Yongtang 1, 2, LUO Xianlu 1, LUO Haibo 1 1. Department

More information

ACS-3921/ Computer Security And Privacy. Chapter 9 Firewalls and Intrusion Prevention Systems

ACS-3921/ Computer Security And Privacy. Chapter 9 Firewalls and Intrusion Prevention Systems ACS-3921/4921-001 Computer Security And Privacy Chapter 9 Firewalls and Intrusion Prevention Systems ACS-3921/4921-001 Slides Used In The Course A note on the use of these slides: These slides has been

More information

Pyrite or gold? It takes more than a pick and shovel

Pyrite or gold? It takes more than a pick and shovel Pyrite or gold? It takes more than a pick and shovel SEI/CERT -CyLab Carnegie Mellon University 20 August 2004 John McHugh, and a cast of thousands Pyrite or Gold? Failed promises Data mining and machine

More information

Intrusion Detection Systems (IDS)

Intrusion Detection Systems (IDS) Intrusion Detection Systems (IDS) Presented by Erland Jonsson Department of Computer Science and Engineering Intruders & Attacks Cyber criminals Activists State-sponsored organizations Advanced Persistent

More information

Solving the Artificial Ant on the Santa Fe Trail Problem in 20,696 Fitness Evaluations

Solving the Artificial Ant on the Santa Fe Trail Problem in 20,696 Fitness Evaluations Solving the Artificial Ant on the Santa Fe Trail Problem in 20,696 Fitness Evaluations Steffen Christensen, Franz Oppacher School of Computer Science, Carleton University 25 Colonel By Drive, Ottawa, Ontario,

More information

A Structural Optimization Method of Genetic Network Programming. for Enhancing Generalization Ability

A Structural Optimization Method of Genetic Network Programming. for Enhancing Generalization Ability International Journal of Engineering Innovation and Management Vol.6, No.2, 2016 A Structural Optimization Method of Genetic Network Programming for Enhancing Generalization Ability Shingo Mabu, Yamaguchi

More information

Santa Fe Trail Problem Solution Using Grammatical Evolution

Santa Fe Trail Problem Solution Using Grammatical Evolution 2012 International Conference on Industrial and Intelligent Information (ICIII 2012) IPCSIT vol.31 (2012) (2012) IACSIT Press, Singapore Santa Fe Trail Problem Solution Using Grammatical Evolution Hideyuki

More information

Genetic Algorithm Performance with Different Selection Methods in Solving Multi-Objective Network Design Problem

Genetic Algorithm Performance with Different Selection Methods in Solving Multi-Objective Network Design Problem etic Algorithm Performance with Different Selection Methods in Solving Multi-Objective Network Design Problem R. O. Oladele Department of Computer Science University of Ilorin P.M.B. 1515, Ilorin, NIGERIA

More information

An empirical study of the efficiency of learning boolean functions using a Cartesian Genetic Programming approach

An empirical study of the efficiency of learning boolean functions using a Cartesian Genetic Programming approach An empirical study of the efficiency of learning boolean functions using a Cartesian Genetic Programming approach Julian F. Miller School of Computing Napier University 219 Colinton Road Edinburgh, EH14

More information

ISSN: [Keswani* et al., 7(1): January, 2018] Impact Factor: 4.116

ISSN: [Keswani* et al., 7(1): January, 2018] Impact Factor: 4.116 IJESRT INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH TECHNOLOGY AUTOMATIC TEST CASE GENERATION FOR PERFORMANCE ENHANCEMENT OF SOFTWARE THROUGH GENETIC ALGORITHM AND RANDOM TESTING Bright Keswani,

More information

Analysis of TCP Segment Header Based Attack Using Proposed Model

Analysis of TCP Segment Header Based Attack Using Proposed Model Chapter 4 Analysis of TCP Segment Header Based Attack Using Proposed Model 4.0 Introduction Though TCP has been extensively used for the wired network but is being used for mobile Adhoc network in the

More information

Chapter 9. Firewalls

Chapter 9. Firewalls Chapter 9 Firewalls The Need For Firewalls Internet connectivity is essential Effective means of protecting LANs Inserted between the premises network and the Internet to establish a controlled link however

More information

Artificial Immune System against Viral Attack

Artificial Immune System against Viral Attack Artificial Immune System against Viral Attack Hyungjoon Lee 1, Wonil Kim 2*, and Manpyo Hong 1 1 Digital Vaccine Lab, G,raduated School of Information and Communication Ajou University, Suwon, Republic

More information

ADAPTATION OF REPRESENTATION IN GP

ADAPTATION OF REPRESENTATION IN GP 1 ADAPTATION OF REPRESENTATION IN GP CEZARY Z. JANIKOW University of Missouri St. Louis Department of Mathematics and Computer Science St Louis, Missouri RAHUL A DESHPANDE University of Missouri St. Louis

More information

Parallel Linear Genetic Programming

Parallel Linear Genetic Programming Parallel Linear Genetic Programming Carlton Downey and Mengjie Zhang School of Engineering and Computer Science Victoria University of Wellington, Wellington, New Zealand Carlton.Downey@ecs.vuw.ac.nz,

More information

2. INTRUDER DETECTION SYSTEMS

2. INTRUDER DETECTION SYSTEMS 1. INTRODUCTION It is apparent that information technology is the backbone of many organizations, small or big. Since they depend on information technology to drive their business forward, issues regarding

More information

Monitoring the Device

Monitoring the Device The system includes dashboards and an Event Viewer that you can use to monitor the device and traffic that is passing through the device. Enable Logging to Obtain Traffic Statistics, page 1 Monitoring

More information

Using Threat Analytics to Protect Privileged Access and Prevent Breaches

Using Threat Analytics to Protect Privileged Access and Prevent Breaches Using Threat Analytics to Protect Privileged Access and Prevent Breaches Under Attack Protecting privileged access and preventing breaches remains an urgent concern for companies of all sizes. Attackers

More information

Evolution of Robot Leg Movements in a Physical Simulation

Evolution of Robot Leg Movements in a Physical Simulation Evolution of Robot Leg Movements in a Physical Simulation Jens Ziegler, Wolfgang Banzhaf University of Dortmund, Dept. of Computer Science, D-44227 Dortmund, Germany Abstract This paper introduces a Genetic

More information

A Virtual Laboratory for Study of Algorithms

A Virtual Laboratory for Study of Algorithms A Virtual Laboratory for Study of Algorithms Thomas E. O'Neil and Scott Kerlin Computer Science Department University of North Dakota Grand Forks, ND 58202-9015 oneil@cs.und.edu Abstract Empirical studies

More information

Towards Traffic Anomaly Detection via Reinforcement Learning and Data Flow

Towards Traffic Anomaly Detection via Reinforcement Learning and Data Flow Towards Traffic Anomaly Detection via Reinforcement Learning and Data Flow Arturo Servin Computer Science, University of York aservin@cs.york.ac.uk Abstract. Protection of computer networks against security

More information

Storage Efficient Capturing of Port Scanning Attack Traffic

Storage Efficient Capturing of Port Scanning Attack Traffic Storage Efficient Capturing of Port Scanning Attack Traffic Rajni Ranjan Singh Department of Computer Science and Engineering Maulana Azad National Institute of Technology, Bhopal, M.P., India Orcid Id:

More information

Module 19 : Threats in Network What makes a Network Vulnerable?

Module 19 : Threats in Network What makes a Network Vulnerable? Module 19 : Threats in Network What makes a Network Vulnerable? Sharing Unknown path Many points of attack What makes a network vulnerable? Unknown perimeter Anonymity Complexity of system Categories of

More information

Vulnerability Assessments and Penetration Testing

Vulnerability Assessments and Penetration Testing CYBERSECURITY Vulnerability Assessments and Penetration Testing A guide to understanding vulnerability assessments and penetration tests. OVERVIEW When organizations begin developing a strategy to analyze

More information

SYMANTEC ENTERPRISE SECURITY. Symantec Internet Security Threat Report September 2005 Power and Energy Industry Data Sheet

SYMANTEC ENTERPRISE SECURITY. Symantec Internet Security Threat Report September 2005 Power and Energy Industry Data Sheet SYMANTEC ENTERPRISE SECURITY Symantec Internet Security Threat Report September 00 Power and Energy Industry Data Sheet An important note about these statistics The statistics discussed in this document

More information

Host Identity Sources

Host Identity Sources The following topics provide information on host identity sources: Overview: Host Data Collection, on page 1 Determining Which Host Operating Systems the System Can Detect, on page 2 Identifying Host Operating

More information

Software Vulnerability

Software Vulnerability Software Vulnerability Refers to a weakness in a system allowing an attacker to violate the integrity, confidentiality, access control, availability, consistency or audit mechanism of the system or the

More information

A Rule-Based Intrusion Alert Correlation System for Integrated Security Management *

A Rule-Based Intrusion Alert Correlation System for Integrated Security Management * A Rule-Based Intrusion Correlation System for Integrated Security Management * Seong-Ho Lee 1, Hyung-Hyo Lee 2, and Bong-Nam Noh 1 1 Department of Computer Science, Chonnam National University, Gwangju,

More information

Evolutionary Lossless Compression with GP-ZIP

Evolutionary Lossless Compression with GP-ZIP Evolutionary Lossless Compression with GP-ZIP Ahmad Kattan and Riccardo Poli Abstract In this paper we propose a new approach for applying Genetic Programming to lossless data compression based on combining

More information

Sparse Matrices Reordering using Evolutionary Algorithms: A Seeded Approach

Sparse Matrices Reordering using Evolutionary Algorithms: A Seeded Approach 1 Sparse Matrices Reordering using Evolutionary Algorithms: A Seeded Approach David Greiner, Gustavo Montero, Gabriel Winter Institute of Intelligent Systems and Numerical Applications in Engineering (IUSIANI)

More information

Intrusion Detection - Snort

Intrusion Detection - Snort Intrusion Detection - Snort 1 Sometimes, Defenses Fail Our defenses aren t perfect Patches aren t applied promptly enough AV signatures not always up to date 0-days get through Someone brings in an infected

More information

Test Case Generation for Classes in Objects-Oriented Programming Using Grammatical Evolution

Test Case Generation for Classes in Objects-Oriented Programming Using Grammatical Evolution Test Case Generation for Classes in Objects-Oriented Programming Using Grammatical Evolution Jirawat Chaiareerat, Peraphon Sophatsathit and Chidchanok Lursinsap Abstract This paper proposes a dynamic test

More information

Intrusion Detection - Snort. Network Security Workshop April 2017 Bali Indonesia

Intrusion Detection - Snort. Network Security Workshop April 2017 Bali Indonesia Intrusion Detection - Snort Network Security Workshop 25-27 April 2017 Bali Indonesia Issue Date: [31-12-2015] Revision: [V.1] Sometimes, Defenses Fail Our defenses aren t perfect Patches weren t applied

More information

A Genetic Based Algorithm to Generate Random Simple Polygons Using a New Polygon Merge Algorithm

A Genetic Based Algorithm to Generate Random Simple Polygons Using a New Polygon Merge Algorithm A Genetic Based Algorithm to Generate Random Simple Polygons Using a New Polygon Merge Algorithm Ali Nourollah, Mohsen Movahedinejad Abstract In this paper a new algorithm to generate random simple polygons

More information

EVOLVING LEGO. Exploring the impact of alternative encodings on the performance of evolutionary algorithms. 1. Introduction

EVOLVING LEGO. Exploring the impact of alternative encodings on the performance of evolutionary algorithms. 1. Introduction N. Gu, S. Watanabe, H. Erhan, M. Hank Haeusler, W. Huang, R. Sosa (eds.), Rethinking Comprehensive Design: Speculative Counterculture, Proceedings of the 19th International Conference on Computer- Aided

More information