Symposium 2005 Materials

We will be posting more materials from the symposium as it becomes available.

Presentation Slides

Posters

Congratulations to the winners of the 6th Annual Information Security Symposium Poster Competition! We are proud to be associated with the following students whose posters best reflect the intellect and talent CERIAS fosters within its research and education programs.

First Place:

Abhi Bhargav Spantzel and Anna C. Squicciarini: Integrating Identity Management and Trust Negotiation

http://www.cerias.purdue.edu/news_and_events/events/symposium/2005/materials/#F949EF


Second Place

Michael Yang: ReAssure: Virtual Imaging Instrument for Logically Destructive Experiments

http://www.cerias.purdue.edu/news_and_events/events/symposium/2005/materials/#00B46C


Third Place

Mohamed Shehab: SERAT : SEcure Role mApping Technique for Decentralized Secure Interoperability

http://www.cerias.purdue.edu/news_and_events/events/symposium/2005/materials/#12FE0B

Identification, Authentication and Privacy

Incident Detection, Response, and Investigation

Assurable Software and Architectures

Enclave and Network Security

Security Awareness, Education and Training

Cryptology and Rights Management

Risk Management, Policies and Laws

Trusted Social and Human Interactions

 

Adaptive Intrusion Response using Attack graphs in an E-commerce environment

Yu-Sung Wu;
Area: Assurable Software and Architectures
Distributed systems with multiple interacting services, such as distributed e-commerce systems, are suitable targets for malicious attacks because of the potential financial impact. Intrusion detection in such systems has been an active area of research, while the problem of automated response has received relatively less attention. The thought often is that a system administrator will be included in the loop for troubleshooting once the alert about a possible intrusion has been raised. In this paper, we present the design of automated response mechanisms in an intrusion tolerant system called ADEPTS. The particular type of response we focus on enforces containment in the system, through which it localizes the effect of the intrusion thus allowing the system to provide service, albeit degraded. Containment can be very important in a large class of distributed systems in which a single compromised service can affect other services through the mutual interactions. ADEPTS uses a graph of intrusion goals, called I-GRAPH as the underlying representation in the system. In response to alerts from an intrusion detection framework, ADEPTS executes an algorithm to determine the possible path of spread of the intrusion and the appropriate response to deploy. A feedback mechanism evaluates the success of a deployed response and uses that in guiding future choices. ADEPTS is demonstrated on a distributed e-commerce system and evaluated using the survivability metric whose value depends on the operational services in the face of an intrusion.

Convenience and Security in the Safely Contained Experimentation Facility, Reassure.

Pascal Meunier (PI); Michael Yang;
Area: Assurable Software and Architectures
ReAssure is a work in-progress to create a virtual machine-based reconfigurable facility for efficient, reproducible, controlled and safely contained experimentation. Convenience and security in complex systems are traditionally seen as being at odds; what follows is a description of the challenges and solutions we adopted. The ReAssure human interface is a web-based application, with a separate utility for configuring networks, as well as access to experimental PCs. This poster describes how and why the data provided by the convenient but untrusted graphical network configuration tool can be trusted.

ReAssure: Virtual Imaging Instrument for Logically Destructive Experiments

Pascal Meunier (PI); Michael Wang
Area: Assurable Software and Architectures

Role-Based Access Control for Group Communication Systems

Jacques Thomas;
Area: Assurable Software and Architectures
In addition to basic security services such as confidentiality, integrity and data source authentication, a secure group communication system should also provide authentication of participants and access control to group resources. While considerable research has been conducted on providing confidentiality and integrity for group communication, less work focused on group access control services. In the context of group communication, specifying and enforcing access control becomes more challenging because of the dynamic and distributed nature of groups and the fault tolerance issues (i.e. withstanding process faults and network partitions).

In this work we analyze the requirements access control mechanisms must fulfill in the context of group communication and define a framework for supporting fine-grained access control in client-server group communication systems. Our framework combines role-based access control mechanisms with environment parameters (time, IP address, etc.) to provide policy support for a wide range of applications with very different requirements. While policy is defined by the application, its efficient enforcement is provided by the group communication system.

Security Based Testing of Access Control Systems

Ammar Masood;
Area: Assurable Software and Architectures
Software testing is essential to verify implementation conformance to the desired functional objectives. The high cost and extremely time consuming nature of software testing is a major hurdle in increasing its effectiveness, therefore it is very important to establish a structured process for generating tests. One important aspect of software testing is verifying that the software meets its specified security objectives, which is classified as security testing. The discovery of dangerous flaws in many security products signifies the importance of security testing. Using software specifications for test generation has several advantages as compared to code/structure based test generation. We present a model based testing strategy for security testing of access control systems. Our model based approach is formally guided by the system constraints in generating structural and behavior models of the system under test. The generated models are then used to derive test suites for performing the security testing.

The Poly^2 Project

David Ehrmann; Gregory Ose; Michael Armbrust; Jay Gengelbach; Keith Watson (PI);
Area: Assurable Software and Architectures
A significant challenge faced by many organizations today is the enforcement and evaluation of trust in complex networks built using a variety of commercial off-the-shelf and freeware components, and running on one or more general purpose operating systems. We will address this problem by simplifying, modularizing, and separating functionality to the extent that Poly^2 components have the minimum functionality needed for a particular network task, and interactions between components are confined to known, defendable contexts. The Poly^2 research project advances the understanding of building secure and reliable system architectures to support critical services in hostile network environments. A secure and reliable system architecture must only provide the required services to authorized users in time to be effective. The Poly^2 architecture we propose will be based on sound, widely acknowledged security design principles. It will form the basis for providing present and future network services while, at the same time, being highly robust and resistant to attack. A prototype of the new architecture will be developed that will provide traditional network services (e.g. web, FTP, email, DNS, etc.) using commodity hardware and an open source operating system. All along, security analyses will be conducted to see if there are flaws in the design or implementation and whether those flaws are the result of conflicting requirements or design objectives. The resulting implementation will serve as a secure and highly available platform from which organizations can deploy their own critical network services. Poly^2 research will provide the community with a better understanding of the application of sound security design principles, the decomposition of COTS software components to increase trust, separation of data based on design and policy, and the isolation of services to decrease commonality and contain failures and attacks. Further, it can provide a springboard for additional research in areas such as intrusion detection, security evaluation, and performance metrics.

Uncheatable Master-Worker Grid Computing and Its Application in Drug Discovery

Mummoorthy Murugesan;
Area: Assurable Software and Architectures
Drug discovery process has long been acknowledged as an expensive and computationally intensive procedure. This process starts with identifying a protein that causes a disease. Subsequently, millions of molecules are tested for finding lead molecules that will bind with the protein to modulate the disease. Instead of doing laborious experiments in labs, which normally takes years of analyzing all the lead molecules, molecular modeling is used to do the simulations of the same computations in few hours. Researchers have investigated methods to accelerate this matching process (also known as docking), and the most recent involves the PC Grid computing.

However, in grid computing, since the supervisor does not have any control over the participants' machines and cannot prevent them from manipulating the programs provided, participants may not have performed the necessary computations but claim to have done so. When participants are paid for their contributions, they have strong incentives to cheat to maximize their gains. This cheating behavior, if undetected, may render the results useless. Drug discovery process is a highly expensive procedure, and cheating may cause severe consequences to the whole process.

This presentation discusses how to detect cheating in grid computing for drug discovery projects. We use CBS (Commitment Based Sampling) scheme to detect whether a participant is cheating. We show that our scheme is efficient in terms of communication cost and extra computational effort required. We also present an experimental study on the prototype implementation of our scheme on two molecular docking programs: AutoDock and FTDock. We show that CBS scheme adds less than 7%computational overhead to the actual task.

Updating XML Documents in Distributed and Cooperative Systems

Yunhua Koglin; Yunhua Koglin;
Area: Assurable Software and Architectures
Securing data is becoming a crucial need for most internet-based applications. Whereas the problem of data confidentiality has been widely investigated, the problem of how to ensure that data, when moving among different parties, are modified only according to the stated policies has been so far not deeply investigated. In this paper, we propose an approach supporting parallel and distributed secure updates to XML documents. The approach, based on the use of a security region-object parallel flow (S-RPF) graph protocol, is particularly suited for all environments requiring cooperative updates to XML documents. It allows different users to simultaneously update different portions of the same document, according to the specified access control policies. Additionally, it supports a decentralized management of update operations in that a subject can exercise its privileges and verify the correctness of the operations performed so far on the document without interacting, in most of the cases, with the document server.

Ensuring Correctness over Untrusted Private Databases

Sarvjeet Singh (PI);
Area: Cryptology and Rights Management
Protecting the privacy of data is becoming increasingly important as is evidenced by recent interest in this area, especially for databases. We address the problem of ensuring the correctness of query results returned by an untrusted private database. The database owns the data and may modify it at any time. The querier is allowed to execute queries over this database; however it may not learn anything more than the result of these legal queries about the contents of the database. The querier does not necessarily trust the database and would like the owner to furnish proof that the data has not been modified in response to recent events such as the submission of the query. There are many practical scenarios that lead to this problem including the need for private databases to participate in collaboration with semi-trusted partners or to comply with legal requirements. This problem has not been addressed earlier. Special cases of this problem that trust the data owner,and do not protect the privacy of the data have been addressed earlier. Our work is directly applicable to these special cases too.

We have developed two metrics that capture the correctness of query answers and proposed a number of solutions to this problem that provide a trade-off between the degree of exposure of private data in order to prove correctness, and the cost of generating these proofs and executing the verification. Our final solution is able to provide minimal exposure of private data while ensuring a low overhead. Our proposed solutions are tested through implementation using PostgreSQL and real data. The results show that it is easy to implement our techniques and the overheads are acceptable.

Evaluation Methodologies for Internet Security Technology (EMIST)

Roman Chertov; Sonia Fahmy (PI);
Area: Enclave and Network Security
Evaluation Methodologies for Internet Security Technology (EMIST) is a multi-institution project to develop rigorous testing methodologies, tools, and benchmarks for Internet security attacks and defenses. EMIST is a companion project to Defense Technology Experimental Research Network (DETER): a remotely accessible experimental testbed that allows researchers to evaluate Internet cyber-security technologies in a realistic, but quarantined, environment. The team from Purdue has developed innovative models to characterize the impact of attacks on congestion control (TCP) and routing (BGP/OSPF) protocols on Internet infrastructure and hosts, both spatially and temporally. We have also identified a number of key limitations of analytical, simulation, and emulation models.

Global Purification of Internet Streams

Bhagyalaxmi Bethala;
Area: Enclave and Network Security

Scalable Infrastructure Protection and Performance Evaluation in Power-law Networks

Hyojeong Kim;
Area: Enclave and Network Security

WormHole: A Gigabit Worm Filter Using Network Processor Technology

Ikkyun Kim;
Area: Enclave and Network Security
We have made initial steps to address the fast worm filtering problem, designing a fast filtering architecture and implementing a prototype system in the IXP1200 network processor platform that handles 30 recent worms including Blaster, CodeRedI & II, Li0n, Nimda, Ramen, Sadmind, Sasser, Slammer, Slapper, and Welchia. The worm filter is able to filter these worms and their mutations-known and artificially generated-at near gigabit line speed (>>965 Mbps). We give a summary of what we have been able to accomplish and what challenges remain to achieve 10 Gbps and faster worm filtering -gigabit line speed filtering because of its worm-specificity. By this we mean: because worms implementing buffer overflow exploits do not generate the vast space of polymorphic variations that viruses can-i.e., ones that cannot be dealt with by length invariance-we are able to utilize design techniques that perform filtering in O(1) time. General purpose packet classifiers cannot make full use of this specialization, which limits their worm filtering performance. Our worm filter architecture realizes three principal features: (f1) workload sensitivity that puts frequent traffic first, (f2) Bloom filter hashing that rapidly identifies non-worm traffic for immediate pass-through, and (f3) multi-level caching that appropriates differentiated processing speed while allowing run-time programmability. Features (f1) and (f2) embody Amdahl's law and Bloom filtering provides an effective means to achieve identification of "important" traffic that dominate overall filtering performance. A beneficial consequence is that normal traffic is passed through with minimal footprint.

A GTRBAC Based System for Workflow Composition and Management

Basit Shafiq;
Area: Identification, Authentication and Privacy
We present an architecture for adaptive real-time workflow-based collaborative system. Such a system is needed to support communication and sharing of information among predefined or ad hoc team of users collaborating with each other for the execution of their respective tasks in the workflow. A key requirement for real-time workflow system is to provide the right data to the right person at the right time. In addition, the workflow needs to be reconfigured if a subtask of a workflow cannot be executed within the due time. We use the generalized temporal role-based access control (GTRBAC) model to capture the real-time dependencies of such workflow applications. In addition, support for triggers in GTRBAC allows dynamic adaptation of workflow based on the occurrence of certain events. Such adaptations may include rescheduling of workflow tasks, reassignment of users to scheduled tasks based on their availability and skill level, and abortion of incomplete tasks.

An Analysis of Privacy and Security Information Provided and Elicited by Different Types of Web Site

M Athar Ali; Robert Proctor (PI);
Area: Identification, Authentication and Privacy
Issues involving consumer privacy and the privacy policies of organizations are of concern for Web transactions that require users to provide personal information. Although several studies have examined users' reported concerns and preferences with respect to privacy issues, no systematic investigation has been conducted regarding the personal information requested by Web sites and the ways in which the sites incorporate privacy and security features to alleviate users' concerns. We report the results of an investigation of 42 Web sites, six each from seven different categories. The results show large differences in the amount and types of personal information requested, both between and within the different site categories. This variability may cause users to hesitate to provide information to a site that deviates from their prior experience with similar sites. Furthermore, although privacy certification logos are viewed favorably by users, many sites do not display them.

Biometric Feasibility Study: Hand Geometry at the Recreational Sports Center

Eric Kukula;
Area: Identification, Authentication and Privacy
Technology is impacting most components of our lives, and will continue to do such with the continuous development and passage of information security and privacy protection laws. An emerging field known as biometrics combines privacy, security, and audit control functionalities together to an individual's physical and behavioral characteristics for use in applications such as: time and attendance, access control, e-commerce, and identification systems.

This study explores the feasibility and performance related to implementing a hand geometry system for access and audit control purposes at the Purdue University Recreational Sports Center. This system will replace the existing system of magnetic stripe identification cards to enter the facility. The current system requires a user to swipe a student identification card through a reader to access the recreational center. Since the access decision is based solely on a token, which can be stolen, lost, handed to others, or copied; accurate auditing is not possible, causing a two fold problem: (1) an insurance risk as accurate records of who is in the facility is not available, and (2) a loss of revenue for the recreational center as multiple people could possibly use the same identification card. Hand geometry removes the token and replaces it with something you have, your hand, removing the possibility for lost or stolen tokens to be used for accessing the recreational center.

This evaluation discusses implementation issues of the hand geometry system into the current infrastructure, user interactions, including the habituation curve, user perceptions of the device, and the performance results, including the failure to enroll rate, failure to acquire rate, false match rate, false non-match rate, as well as analyzing the interclass relationships using collected demographic information including: gender, ethnicity, and handedness.

Digital Identity Management domain in Ontological Semantics

Evguenia Malaia (PI);
Area: Identification, Authentication and Privacy
This paper focuses on ontological efforts to support information security applications - more specifically, applied natural language processing technology - in the domain of Digital Identity Management (DIM).

Because digital identities have such varied uses and meanings, and because the social implications of digital identity management policies are far-reaching into domains of free speech, privacy and accountability online, it is necessary to develop a universal vocabulary for digital identity framework development and policy language. We propose the framework of ontological semantic processing and text meaning representation through the ontology as a possible solution for this problem. The existing ontology (about 49,000 concepts, including 400 pertaining to information security) would need to be supplemented by ~500-700 more concepts and ~1,500 lexical items to allow adequate support for the domain of digital identity management. The ontological semantic approach to natural language processing as such consists of using both language-independent static knowledge sources (the ontology, fact database) and static language-dependent sources (lexicons, onomasticons, morphological processor, syntactic processor, etc.), as well as dynamic processing algorithms. In summary, the dynamic algorithms (which include the tokenizer, syntactic, ecological and morphological analyzers) process the text and match the lexical items in the text with appropriate ontological concepts through the lexicon-ontology connections, the output being text meaning representation, or TMR. As a language-independent semantic "concept web", it can then be used (with appropriate follow-up processing) for machine translation, data mining, information extraction, question answering, text summarization, etc. The present paper deals with the following methodological questions in domain acquisition for two of the static knowledge sources, the ontology and the lexicon: -Delimitation of the expanding DIM textual corpus with volatile vocabulary;
-Extraction of lexical items pertaining to the domain;
-Building ontological support for lexical items; introduction of necessary attributes and relations;
-Semantic representation of lexical items in the domain.

Integrating Identity Management and Trust Negotiation

Abhilasha Bhargav-Spantzel (PI);
Area: Identification, Authentication and Privacy

Ontological Semantics Support for Handling Privacy Policies

Olya Krachina;
Area: Identification, Authentication and Privacy
CyberTrust Project purposes to provide a more robust system for handling privacy policies as compared to the currently available solutions. One of the tasks is to convert user-defined natural language into machine-readable language necessary for further implementation. NLP presents itself as a powerful processing tool, given the fact that privacy policies are formulated in natural language initially. Furthermore, Ontological Semantics successfully accommodates the afore-mentioned need. This poster demonstrates fundamental features of the framework through topic-specific examples.

Privacy-Preserving Distributed k-Anonymity

Wei Jiang (PI);
Area: Identification, Authentication and Privacy
k-anonymity provides a measure of privacy protection by preventing re-identification of data to fewer than group of k data items. While algorithms exist for producing k-anonymous data, the model has been that of a single source wanting to publish data. This paper presents a k-anonymity protocol when the data is vertically partitioned between sites. A key contribution is a proof that the protocol preserves k-anonymity between the sites: While one site may have individually identifiable data, it learns nothing that violates k-anonymity with respect to the data at the other site. This is a fundamentally different distributed privacy definition than that of Secure Multiparty Computation, and it provides a better match with both ethical and legal views of privacy.

Protecting Consumer Privacy in Reusable Digital Devices

Brad Moseng (PI);
Area: Identification, Authentication and Privacy
The increasing use of disposable digital devices is leading to expanding vulnerabilities in personal privacy. In an effort to bring digital devices to the mass populace, manufacturers have started to market single-use, recyclable digital devices. One such device, the Dakota single-use digital camera, is designed to compete in the disposable photography market. However, with reusable digital devices, there is often sensitive residual data left behind from previous users. The purpose of this research is to determine if Pure Digital, the makers of the Dakota camera, are providing enough data security in their device recycling process.

Purpose Based Access Control for Privacy Protection

Ji-Won Byun;
Area: Identification, Authentication and Privacy
In this project, we investigate a comprehensive approach for privacy preserving access control based on the notion of purpose. Purpose information associated with a given data element specifies the intended use of the data element, and our model allows multiple purposes to be associated with each data element. A key feature of our model is that it also supports explicit prohibitions, thus allowing privacy officers to specify that some data should not be used for certain purposes. Another important issue is the granularity of data labeling, that is, the units of data with which purposes can be associated. We address this issue in the context of relational databases and propose four different labeling schemes, each providing a different granularity. We also explore an approach to representing purpose information, which results in very low storage overhead, and we exploit query modification techniques to support data access control based on purpose information.

Querying Private Data in Moving-Object Environments

Reynold Cheng;
Area: Identification, Authentication and Privacy
Location-based services, such as finding the nearest gas station, require users to supply their location information. However, a user's location can be tracked without her consent or knowledge. Lowering the spatial and temporal resolution of location data sent to the server has been proposed as a solution. Although this technique is effective in protecting privacy, it may be overkill and the quality of desired services can be severely affected. In this paper, we investigate the relationship between uncertainty, privacy, and quality of services. We propose using imprecise queries to hide the location of the query issuer and evaluate uncertain information. We also suggest a framework where uncertainty can be controlled to provide high quality and privacy-preserving services. We study how the idea can be applied to a moving range query over moving objects. We further investigate how the linkability of the proposed solution can be protected against trajectory-tracing.

Securing the Manufacturing Environment using Biometrics

Stephen Elliott (PI); Shimon Modi;
Area: Identification, Authentication and Privacy
Computer integrated manufacturing systems have changed the interaction of Computer integrated manufacturing systems have changed the interaction of industrial manufacturing equipment with different systems within and outside the manufacturing environment. The increase in the sophistication of the manufacturing equipment, along with increased connectivity with internal and external systems has changed the way that manufacturing security is designed. As manufacturers move towards a more connected collaborative environment in order to compete in global businesses and geographically disparate facilities, concerns that their proprietary manufacturing processes and intellectual property could be exposed to damaging compromise on a worldwide scale are increasing. The US government has also passed several regulations so that companies take into account general concerns like physical and logical security. The Sarbanes-Oxley Act of 2002 and FDA's 21 CFR Part 11 are two such regulations which require that companies have specific controls to ensure authenticity, integrity and auditability of electronic records. As part of the compliance guidelines, biometrics is indicated as a preferred means of security. The general problem that the manufacturing environment is facing is that operation of most industrial manufacturing equipment does not require any strong form of authentication or identification when some transaction related to product manufacturing takes place. Most manufacturing systems require a password to log onto the system, after which the system is open to anyone on the manufacturing floor to operate. The manufacturing systems are sophisticated enough to provide remote operation capability, but the only form of authentication is a password. There are no means to ascertain who was operating the machine and if they were authorized to do so. In an event of a malfunction or accident, the audit trail does not provide adequate information. Biometrics can solidify the authority checks and operator entry checks since the authentication is no longer based only on passwords or security cards/tokens. The main aim of this project is to demonstrate the integration of several biometric technologies into the manufacturing environment. This project proposes a unique application of biometrics and computer integrated technology as part of providing an applied solution for the problems of security and auditability in the manufacturing environment.

Security and Privacy in Healthcare Environments

Bharat Bhargava; Leszek Lilien; Yuhui Zhong
Area: Identification, Authentication and Privacy
The objectives of our work on security and privacy in healthcare environments include assuring security, privacy, and safety for patients and staff, as well as processes and facilities in healthcare institutions. Our work goes in three major directions. First, we are concerned with vulnerabilities due to malicious behavior, hostile settings, terrorist attacks, natural disasters, and tampering with data. Second, we study how reliability, security, and privacy issues affect timeliness and precision of patient information. Third, we investigate methods and techniques to assure secure, private, trustworthy, reliable, consistent, correct and pseudonymous collaboration over networks among healthcare professionals.

SERAT : SEcure Role mApping Technique for decentralized secure interoperability

Mohamed Shehab (PI);
Area: Identification, Authentication and Privacy
Multi-domain application environments where distributed domains interoperate with each other are becoming a reality in internet-based and web-services based enterprise applica- tions. The secure interoperation in a multidomain environment is a challenging problem. In this paper, we propose a distributed secure interoperability protocol that ensures secure interoperation of the multiple collaborating domains without compromising the security of collaborating domains. We introduce the idea of access paths and access paths constraints. Furthermore, we device a path discover algorithm that is capable of querying interoperating domains for the set of secure access paths between different domains.

Translation-based Steganography

Christian Grothoff; Krista Bennett;
Area: Identification, Authentication and Privacy
We are investigating the possibilities of steganographically embedding information in the "noise" created by automatic translation of natural language documents. An automated natural language translation system is ideal for steganographic applications, since natural language translation leaves plenty of room for variation. Also, because there are frequent errors in legitimate automatic text translations, additional errors inserted by an information hiding mechanism are plausibly undetectable and would appear to be part of the normal noise associated with translation. Significantly, it should be extremely difficult for an adversary to determine if inaccuracies in the translation are caused by the use of steganography or by perceptions and deficiencies of the translation software.

VNCs, Web Portals, Biometrics and the Verification of Distance Education Students

Nathan Sickler (PI);
Area: Identification, Authentication and Privacy
Technologies exist which allow Higher Education Institutions to extend their classrooms beyond the traditional lecture hall to virtual learning environments. Technologies such as Virtual Network Connections (VNCs), and Web Portals (WebCT Vista & CE) are useful tools in the transfer of knowledge to distance students. However, these technologies offer weak methods of student verification - passwords. To improve the confidence of Universities, Instructors, and other Students that the proper distance students are answering exams, quizzes, assignments, and lab practicals, biometric technologies (i.e. face, fingerprint, iris, keystroke dynamics) can be employed.

Wireless Encryption Practices: Social Capital Factors and Diffusion of Innovation

Sorin Matei; John Hooker (PI);
Area: Identification, Authentication and Privacy
The poster is based on a study that explores how social capital effects the diffusion of the social practice of residential wireless computer network encryption over time. It is important to note that we are examining the influences on and practice of using encryption, not the technology itself. As a practice, choosing to use encryption is similar to choosing whether to use other computer technologies like the internet and e-mail. Thus, our results show that formal social capital (membership in community organizations) influences people to employ encryption in their home wireless computer networks. Informal social capital (level of neighborhood belonging) was found to not significantly affect encryption practices.

Causality-based Intrusion Analysis

Sundararaman Jeyaraman;
Area: Incident Detection, Response, and Investigation
Intrusion Analysis has traditionally been a very time-consuming and manual task. We take the view that the current failings of intrusion analysis tools can be attributed to the lack of data that identify causal relationships between system events. Causal relationships, once identified, are not only useful for intrusion-analysis, but also in a variety of other areas like Intrusion Detection, Forensics, Information-flow tracking and Intrusion-alert correlation. In our work, we design and study a host of techniques for capturing causal relationships.

Discussion of Defense against Phishing

mercan topkara (PI);
Area: Incident Detection, Response, and Investigation

Foundations of Digital Forensic Investigations

Brian Carrier;
Area: Incident Detection, Response, and Investigation
Digital investigations frequently occur, but there is little theory behind them. The lack of theory becomes apparent when one tries to define tool requirements for development and testing. There is currently an umbrella term of "computer forensic tools," but there are no formally defined categories of tools. This is similar to including IDS, firewalls, and anti-virus in the category of "computer security tools." In this work, we formally examine how data and evidence are created and define a framework accordingly. This framework can be used to more rigorously define procedures and tools.

iPod Forensics

Christopher Marsico (PI);
Area: Incident Detection, Response, and Investigation
The iPod is the most popular digital music device. The newest versions of the iPod have become more PDA like then ever before. With this new functionality the iPod has recently found its way into the criminal world. With the continued growth of the digital music device market, the iPod's use in criminal activity will only continue to increase. This research discusses some of the features of the iPod and how a criminal could use them. Literature review found little or no documentation or discussion on the forensic analysis of the iPod or similar devices. Therefore, this research outlines what should be considered when an iPod is found at the crime scene, and offers a critical analysis of some common forensic tools and their ability to collect and analyze data from an iPod.

Psychological Profiling and Computer Forensics: Locard's Principle in the Digital World

Marc Rogers (PI);
Area: Incident Detection, Response, and Investigation
The presentation discusses the need to extend psychological crime scene analysis from its current supportive role in physical crime scene analysis, to an identical role in digital and cyber crime scenes. The fundamentals of crime scene analysis are discussed and a focus on the ability of psychological cyber crime scene analysis to answer the FBI's critical elements is presented. A model illustrating the analogous physical and cyber crime scene elements is provided. The importance of cyber victimology in profiling and target hardening is also briefly examined, as is the importance of not being fearful of the seeming uniqueness of computer crime scenes. Finally, suggestions for future study are offered

The Trojan Horse Defense In Cybercrime Cases

Brian Carrier;
Area: Incident Detection, Response, and Investigation
In a United Kingdom prosecution in 2003, Aaron Caffrey, an admitted hacker, was acquitted of hacking into the computer system of the Port of Houston and shutting it down. Caffrey was acquitted despite the fact that no one disputed it was his laptop which committed the crime. His defense was that his computer was taken over, courtesy of a Trojan horse, and used to commit the crime without his knowledge. The jury acquitted even though no evidence of a Trojan horse was found on Caffrey's computer.

Caffrey's is only one of several cases in which defendants who invoked the "Trojan horse defense" have been acquitted of charges ranging from hacking to tax fraud. This article examines the defense from both legal and technical perspectives; while its primary focus is on how the prosecution can respond to the invocation of a Trojan horse defense, it also explains that such a defense can, in fact, be well-grounded. It would be quite possible for someone to use a Trojan horse program to seize control of another's computer and frame them.

Using process labels to obtain forensic and traceback information

Florian Buchholz;
Area: Incident Detection, Response, and Investigation
Many questions that are of interest in digital forensics, intrusion detection, network traceback, and access control can not be sufficiently answered by today's computing systems. In particular the question of who or what caused certain actions on a system or from where they originated can not be answered by information currently available. In this paper we present a model that makes it possible to propagate arbitrary meta-information bound to subjects -- active principals and passive objects -- based on the exchange of information among them. The goal is to bind a meaningful label, for instance user or location information, to an active principal on a system at a certain event and then be able to propagate and track that label as information is being exchanged.

Virtual Playgrounds For Worm Behavior Investigation

Xuxian Jiang (PI);
Area: Incident Detection, Response, and Investigation
To better understand worms' dynamic and possibly camouflaged behavior, researchers have long hoped to have a safe and convenient environment to unleash and run real-world worms. There are, however, major challenges in realizing such "worm playgrounds", including the playgrounds' fidelity, confinement, scalability, as well as convenience in infrastructure setup and worm experiments. In particular, worm playgrounds that use physical hosts as playground nodes may not be effective in addressing these challenges.

In this work, we present a virtualization-based approach to creating virtual worm playgrounds, called vGrounds, on top of a general-purpose infrastructure. A vGround is an all-software virtual environment dynamically created on top of a physical infrastructure. The infrastructure can be a single physical machine, a local cluster, or a multi-domain overlay infrastructure such as PlanetLab. A vGround contains realistic end-hosts and network entities, all realized as virtual machines (VMs) and confined in a virtual network (VN). The salient features of vGround include: (1) high fidelity supporting real worm codes exploiting real vulnerable services, (2) strict confinement making the real Internet totally invisible and unreachable from inside a vGround, (3) high resource efficiency providing worm experiments with a scale magnitudes larger than the number of physical machines in the infrastructure, and (4) flexible and efficient worm experiment control enabling fast (tens of seconds) and automatic generation, re-installation, and final tear-down of vGrounds. Our experiments with real-world worms have successfully reproduced their probing and propagation patterns, exploitation steps, and malicious payloads, demonstrating both research and education value of vGrounds.

Open Source Vs Proprietary Software: Vulnerabilities and Patch Response

Sanjay Sridhar (PI);
Area: Risk Management, Policies and Laws
Software selection is an important consideration in risk management for information security. Additionally, the underlying robustness and security of a technology under consideration has become increasingly important in total cost of ownership and other calculations of business value. Open source software is often touted as being robust to many of the problems that seem to plague proprietary software. This study seeks to empirically investigate, from an information security perspective specific security characteristics of open source software compared to those of proprietary software. Software vulnerability data spanning several years are collected and analyzed to determine if significant differences exist in terms of inter-arrival times of published vulnerabilities, median time to release 'fixes' (commonly referred to as patches), type of vulnerability reported and the respective severity of the vulnerabilities. It appears that both open source and proprietary software are each likely to report similar vulnerabilities and that open source software is only marginally quicker in releasing patches for problems identified in their software. The arguments favoring the inherent security of open source software do not appear to hold up to scrutiny. These findings provide evidence to security managers to focus more on holistic software security management, irrespective of the proprietary-nature of the underlying software.

Securing the National Plant Diagnostic Network

Keith Watson (PI);
Area: Risk Management, Policies and Laws
The National Plant Diagnostic Network (NPDN) is a USDA-funded plant biosecurity program. The mission of the Network is to enhance national agricultural security by quickly detecting introduced pests and pathogens. This is achieved through a nationwide network of public agricultural institutions with a cohesive, distributed system to quickly detect deliberately introduced, high consequence, biological pests and pathogens into agricultural and natural ecosystems by providing means for quick identifications and establishing protocols for immediate reporting to appropriate responders and decision makers. The Network will allow land grant university diagnosticians and faculty, state regulatory personnel, and first detectors to efficiently communicate information, images, and methods of detection throughout the system in a timely manner.

CERIAS is assisting the NPDN in developing an effective information security program. Security assessments were conducted at each regional center and the national database facility to identify security issues. Security policy review is underway. Technical training is being developed for the NPDN regional IT personnel, as well as security awareness materials for all personnel associated with the NPDN.

An Object-Oriented Multimedia System for Delivering K-12 Information Security ‎Curriculum

Nathan Bingham (PI);
Area: Security Awareness, Education and Training
Educational research has shown that the use of multiple sensory stimuli facilitates and ‎extends the information processing ability of learners with a variety of learning styles; in ‎addition, dual-encoding theory posits that multiple representations are more effective for ‎purposes of long-term retention (Smith and Ragan, 1999). Specifically, the use of video ‎is effective for addressing attitudinal change and allows humanization of seemingly ‎abstract and impersonal topics such as information security. These factors, along with ‎practical concerns such as sustainability and efficiency, highlight the utility of interactive ‎multimedia as an ideal tool for end-user training and awareness initiatives. For the K-12 ‎audience who may have little or no experience with the topic, engaging and exciting the ‎prospective learner is the key to changing behavior. ‎

Unfortunately, despite the great benefit to the user, the burden of creating interactive ‎multimedia materials can be both time consuming and costly. Using an object-oriented ‎design approach to both the system and the educational materials, these costs can be ‎reduced over time. Object-oriented systems allow for reusability of multimedia ‎components and coding such as exercises, screen layouts, and graphics. Ideally, once a ‎learning object is created it can be reused or repurposed to a particular audience. The ‎system created for CERIAS's K-12 initiative takes advantage of benefits of both ‎interactive multimedia and object-oriented design. Using Macromedia Flash and XML in ‎an object-oriented design, the K-12 Information Security Modules deliver multimedia ‎lessons to the user using predefined XML instructions created by an instructor. The ‎system, graphics, and exercises used in the module can be reused or repurposed allowing ‎the instructor to extend an existing lesson plan when new topics arise or even tailor the ‎system to a particular audience easily and quickly.‎

CERIAS K-12 Outreach

Matt Rose (PI);
Area: Security Awareness, Education and Training
The K-12 Outreach program has developed a portfolio of educational materials and resources targeted at the needs of K-12 teachers, technology coordinators, administrators, parents, and students. The K-12 Outreach program provides support, staff development programs, student workshops, professional collaborations, and this online reference. To date, more than 5,200 educators have participated in K-12 Outreach program activities. This poster highlights educational and research initiatives in K-12 outreach.

Creating Awareness of Computer Security Concepts Using Multimedia Teaching Aids

Jeff Hennis (PI);
Area: Security Awareness, Education and Training
As the amount of cyber crimes and identity theft steadily increases, there has been a growing need for awareness and training in the areas of computer securities. The purpose of this project was to develop multimedia elements that will fulfill this need to parents and guardians with minors 6 - 17 years of age, in the areas of computer securities. The target audience was tested not only on their awareness and knowledge before viewing the developed product, but also tested how well it created awareness, the effectiveness, and the motivation to look further into the areas of computer securities. The testing method of the product was available in a PDF file, on the internet. The results received, showed a possible increase in awareness and motivation based on the effectiveness of the product. With knowledge of this outcome, Group 12 will show how their form of reintroduced technologies can be used as an effective as a teaching aid.

Internship "Living Lab" Projects

Ed Finkler;
Area: Security Awareness, Education and Training
To provide real world experience in Networking and Security to students by supporting the Computer and Information Technology Department Students, Classes and Faculty.

National Capacity Building

Melissa Dark (PI);
Area: Security Awareness, Education and Training
This poster reports the impact of the Center for Faculty Development in Information Assurance Education which is housed at CERIAS.

Network Security Certificate Program

Connie Justice (PI); Ed Finkler;
Area: Security Awareness, Education and Training
To provide Information Assurance education and training to students and professionals.

Open Seminar

Annie Anton;
Area: Security Awareness, Education and Training

Refactoring Secure Programming Classes

Pascal Meunier (PI);
Area: Security Awareness, Education and Training
Undergraduate curricula most often focuses on achieving functional objectives, and assignments are graded accordingly. Security objectives are typically not emphasized, so most software vulnerabilities are repeated, well-understood mistakes. Secure programming classes have been taught at Purdue since the Fall 2002 in order to address this problem. However, the scope of the classes was limited to that of an associated "sister" class (e.g., operating systems), and biased towards UNIX. We expanded the scope of the previously developed material to include Windows issues and created exercises and labs that are independent of any "sister" class. We also added coverage of trust, threat, risk issues as well as software engineering considerations. In the process we uncovered Windows threats that have been neglected in the litterature. The material is freely available at http://www.cerias.purdue.edu/secprog/ as sets of slides and is divided into three modules, which can be taught to different audiences.

Social Engineering Defense Architecture

Michael Hoeschele (PI); Michael Armbrust (PI);
Area: Trusted Social and Human Interactions
This paper proposes a theoretical solution to the problem of Social Engineering (SE) attacks perpetrated over the phone lines. As a byproduct real time attack signatures are generated, which can be used in a cyber forensic analysis of such attacks. Current methods of SE attack detection and prevention rely on policy and personnel training, which fails because the root of the problem, people, are still involved. The proposed solution relies on computer systems to analyze phone conversations in real time and determine if the caller is deceiving the receiver. This Social Engineering Defense Architecture (SEDA) is completely theoretical as the technologies employed are only in the proof of concept phase, but they are all proven to be tractable problems.