|
All MITRE Projects (with summaries and presentations where available)
Listing of project titles in alphabetical order
Pages: 1234567891011121314151617
Interactive ISR Data Exploitation and Sensor Operation
Primary Investigator:Witkoskie, James B.
Problems:
GMTI data exploitation requires both signal processing by computers and contextual reasoning by human operators/analysts. The operator has a more natural understanding of vehicle behaviors in terms of contextual information such as terrain and road networks, while the algorithms have a speed advantage and can better process noisy data.
By creating novel algorithms that match an operator's hierarchical approach to data fusion and displays that give insight into the algorithm's inner workings, the semi-automated system can recognize problems that an operator can correct, visualize information to allow the operator to address these algorithmic ambiguities, and prompt the user for input.
This interactive ISR exploitation environment will be able to create more reliable information faster than the current operator centered activities, while maintaining the robustness of the products expected from operator exploitation.
Objectives:
Developed prototype environment in MATLAB: Uses a novel non-Kalman Filter based track estimation. These methods allow hierarchical data exploitation, which fits naturally into human reasoning.
Demonstrated utility for disadvantaged radars: These methods are robust for small aperture radars (radars with large cross range errors), where Kalman Filter based methods have difficulties.
Demonstrate utility in forensic applications: Current focus is on forensic applications. We are attempting to speed up this exploitation process. Special emphasis will be paid to exploiting data from multiple assets.
Demonstrate utility in real time tracking/targeting application: Many future radar systems have a real time automated target tracking component. We plan to demonstrate that a semi-automated solution may be more appropriate.
Quantify Uuility: We want to define metrics to measure the performance of this system against manual exploitation as well as automated exploitation.
Transition to operational software platforms: We are targeting ISR Forenics Tool and DotMatrix software platforms, as well as the modernized JSTARS system.
Fuse GMTI data with SIGINT (Signals Intelligence): Need to fuse this information with other data sources, such as SIGINT.
Activities:
1) Restructure Interactive ISR MATLAB environment: An initial prototype was developed in FY08 that does not have the required flexibility to address the wide array of problems we wish to tackle.
2) Develop algorithms to produce higher level information using non-Kalman filtering algorithms to look for patterns (track segments and convoy clustering).
3) Develop algorithms to identify algorithmic ambiguities and prompt user for corrective action.
4) Allow human to override algorithmic decisions and catalogue these decisions to develop better understanding of algorithms missing contextual information.
5) Address MULTI-INT issues, specifically combining MOVINT with SIGINT.
Impact:
1) Disadvantaged systems (proposed Space Radar and UAV radars): Our novel track initiation approach is able to find tracks in noisy data, where traditional Kalman Filter based approaches failed. Technology is being transitioned to DotMatrix.
2) Supporting JSTARS modernization decisions by demonstrating methods of exploiting data from different types of radars (such as X-band vs. Ku-band radars).
3) Tactical radar users will be able to track more targets since they can designate several targets to track and then concentrate on situations where they are prompted to address algorithmic ambiguity.
4) Forensic users will also benefit from ability to track more objects, and use fewer mouse clicks to vet a track. Our novel method deals extremely well with time gaps in radar coverage.
5) Our approach seamlessly addresses issues with fusing data from multiple platforms (including difficulties due to registration and other interplatform issues).
Public Release No:09-0590
[Presentation]
Exhibit Date(s):May 6
Internet Enabled Social Patrol
Primary Investigator:Crossler, Jay A.
Problems:
The goal of the I-SPI project is to develop a MITRE-specific social networking tool for mobile applications. The two primary challenges associated with this project revolve around identifying a set of social networking capabilities that make sense for MITRE employees, and implementing these in a manner that maintains information security while still allowing data to be passed outside the corporate firewall to a mobile device. We hope to create a tool that integrates with existing MITRE-internal social networking tools while still providing new capabilities.
Objectives:
1. Enable secure transmission of data over a firewall
2. Create a social networking tool that facilitates collaboration between MITRE employees
3. Design a system that is interoperable with, and that can easily be updated to include, existing and newly developing social networking tools
Activities:
We are using polling mechanisms, encryption, and the implementation of an XML firewall to allow for sufficient security. There are three primary components to the architecture, as envisioned: a relay server outside the firewall,an eventing server, and an internal scripting server.
For social networking, we plan to implement a contact list, a recommender service, tools for facilitating ad hoc meeting planning, and tools that make it easier to take advantage of unscripted collaboration, like using a QR code reader to quickly add someone to a contact list.
Impact:
Although this application is being developed for a MITRE-internal application, the architecture was designed to allow for relatively easy porting to MITRE’s sponsor organizations. The isolation of the scripting server from the external relay server and internal eventing servers means that the process of passing information across a security boundary is independent of the process by which social networking services are provided. This has two implications: first, this component can be used without the associated scripting server to handle information security in a variety of contexts—for example, passing information between different clearance levels, from command and control centers to tactical positions, etc. Second, the scripting software can easily be populated with organization-specific scripts to manage application branding, data source management, and integration with internally-supported tools. Thus, when completed, this project will not only provide a new social networking application to facilitate collaboration within MITRE, but also provide a technical capability for passing information across security boundaries that can easily be implemented by MITRE’s sponsors.
Public Release No:09-0906
[Presentation]
Exhibit Date(s):May 7
Intervention Strategies for Managing Performance in National Airspace
Primary Investigator:Moreno-Hines, Felipe
Problems:
Over 30 years ago, the Airline Deregulation Act removed government control over airline fares, routes, and other aspects of the airline industry. However, on occasion the FAA has intervened to resolve issues of congestion caused by over-scheduling at capacity constrained airports. This research seeks to understand the impact of FAA regulatory interventions to manage congestion in the National Airspace System. Analysis of past interventions can shed light on the effectiveness and broader impact of the actions taken to fix a delay problem in different market environments. The research will rely on applied economic theory and case studies of recent FAA schedule interventions. The objective is to develop a conceptual framework through which the benefits and costs of various delay mitigation strategies can be evaluated to more systematically determine whether, where, when, how, and how much to intervene.
Objectives:
The project will explore several questions: What outcome and social "good" is a particular intervention intended to manage? How do schedule restrictions affect competitive dynamics? How does the operating environment affect airline response to the use (or anticipated use) of restrictions? How effective are restrictions at achieving their stated objectives? What is the impact of alternative strategies, including doing nothing?
Activities:
This research will start by conducting a review of FAA-imposed flight limits at historically capacity-constrained airports (examples include Chicago’s O’Hare and New York’s JFK). We will study these efforts in terms of their stated rationale and justification, applying theoretic principles concerning the role and effectiveness of government market interventions in promoting socially efficient outcomes. Through empirical case studies, we will analyze the impact of historical interventions on market structure and consumer welfare under the assumption of competitive airline behavior. We will evaluate these case studies in the context of their environment and attempt to generalize the effects of contextual factors on the effectiveness of the applied intervention strategy. These generalizations will serve as the basis of a decision support framework for intervention.
Impact:
While a decision support framework for intervention would not replace the consensus-building process that the FAA currently employs, it would potentially complement it by providing a more objective and consistent method for considering the need for, and nature of, future FAA actions. In the long term, this work will promote the FAA’s ability to craft socially efficient outcomes, and enhance the transparency and equitability through which the agency implements congestion management policies in a multi-stakeholder environment.
Public Release No:09-1017
[Presentation]
IP Multicast Dynamic Secure Session Establishment
Primary Investigator:Tahan, Thomas
iSPE Overview, Impact and Importance
Primary Investigator:Crawford, Greg
Exhibit Date(s):May 6
Laika EHR Testing Framework
Primary Investigator:Gregorowicz, Andrew J.
Problems:
Laika analyzes and reports on the interoperability capabilities of Electronic Health Record (EHR) systems. This includes the testing for certification of EHR software products and networks.
Laika is a collaborative effort with the Certification Commission for Healthcare Information Technology (CCHIT), the sole body recognized by U.S. Health and Human Services for EHR certification. CCHIT started using Laika to certify EHR systems for continuity of care data interoperability, beginning in July 2008. To date, over 21 vendor systems have been certified by CCHIT using the Laika toolkit.
Laika offers more than just support for certification processes. It provides an "interoperability report card" of a candidate EHR system. This analysis will assist health information technology organizations in developing interoperable solutions. Using Laika, vendors, non-profits and government agencies can test their EHR systems for compliance to standards.
Laika is a successful, mature open source project. There have been several major source code contributions to the project from external organizations. Through the combined efforts of the community, Laika offers freely available tools for testing across a number of Health IT domains including continuity of care, Health Information Exchanges (HIE) technologies, and patient identity management.
Objectives:
Establish external positive recognition of MITRE's system engineering work in the Health Informatics domain
Automate the CCHIT health information exchange certification process of U.S. Electronic Health Record systems with an operational system by March 2009
Provide access to the Laika testing software to anyone via an open source, Apache 2.0 distribution license
Secure direct funding from the United States Department of Health and Human Services to support MITRE's system engineering work of Laika, by October 2009.
Activities:
MITRE is collaborating with CCHIT, the sole body recognized by U.S. Health and Human Services to certify EHRs. The Laika software is freely available to anyone via an Apache open source license. Anyone can download, use, and modify the Laika software.
Through Laika, MITRE is helping to build an open source healthcare IT community which includes government, commercial, and non-profit organizations.
Impact:
As of July 2008, CCHIT has started to use Laika to test and certify U.S. EHR systems for data interoperability. To date, 21 EHR vendors have successfully been certified by CCHIT. Laika is the only automated part of the CCHIT interoperbility testing process that these 21 vendors have had to support.
MITRE is providing an open source tool to allow anyone to assess data interoperability of EHR systems.
MITRE is combining technical software capabilities with an understanding of the clinical health care culture to increase use of interoperable EHR systems in the U.S.
Public Release No:09-1159
[Presentation]
Exhibit Date(s):May 7
Lightweight Beacon System for UAS and Other Aviation Applications
Primary Investigator:Strain, Robert C.
Problems:
We will perform research on low-cost, lightweight radio technology capable of providing Automatic Dependent Surveillance-Broadcast (ADS-B) services to small unmanned aircraft systems (UASs) and supporting other aviation applications. These applications may include rapidly deployable surveillance for search and rescue, aerial firefighting, or disaster/national emergency relief efforts.
Objectives:
The goals of this research are to demonstrate a small, lightweight transceiver appropriate for less-maneuverable or special-use aircraft. Such a system would potentially allow small UASs to operate in national airspace where other aircraft are also suitably equipped. This cooperative approach to surveillance will improve visibility of UASs to proximate aircraft. The prototype system being developed will use the Universal Access Transceiver (UAT) waveform developed by MITRE in the mid-1990s, which is now an international standard for ADS-B. The small transceiver, referred to as the UAT Beacon Radio, will have a modular architecture, enabling it to operate as a stand-alone unit or to be integrated with other aircraft electronics. By proving the technology, demonstrating the benefits to small aircraft operators, and transferring the technology, we expect the UAT Beacon Radio will stimulate the ADS-B avionics marketplace for many facets of aviation.
Activities:
We have developed a small, self-contained, battery-operated transmitter unit, and tested it on a small commercial UAS, a sail plane, and a private rocket. Transmissions from this low-power device can be received at ranges in excess of 20 miles, and the device can operate on battery power for up to 14 hours. Recently we added a receiver, and other modifications further reduced the size and weight of the radio. The technology is being made available for licensing to commercial companies and is being used in cooperation with other organizations to investigate new operational applications of ADS-B.
Impact:
Our efforts through 2008 reduced the design risk for manufacturers using the UAT Beacon Radio technology. The market for devices based on our work is maturing. We continue to receive inquiries from commercial companies about our work and how they might benefit. We have attracted industry attention by developing working units that demonstrate and validate the design within strict space, weight, and power constraints. Through these proof-of-concept demonstrations, and MITRE's technology transfer activities, we have emboldened potential radio manufacturers to explore this promising technology.
This project has benefited our FAA and DoD sponsors by significantly increasing awareness of ADS-B within the UAS, commercial space, and general aviation communities. In particular, this demonstrates the potential of minimizing cost, which has been a risk area for the FAA's ADS-B program. MITRE's work helps to reduce this risk by offering a solution appropriate for many airspace users. MITRE's DoD sponsors have expressed interest in this technology, and a number of MITRE's technology licensees are developing potential DoD applications.
Public Release No:09-1018
[Presentation]
Exhibit Date(s):May 6, May 7
Live Hotspotting of VoIP
Primary Investigator:Hu, Qian
Exhibit Date(s):May 5, May 6
Machine Translation for Foreign Language Science and Technology Analysis
Primary Investigator:Burger, John D.
Problems:
Foreign language science and technology intelligence analysis requires several valuable sets of skills, making it a high-impact target for any force-multiplier technology. The most obvious such technology is machine translation, but it is unclear whether recent dramatic improvements in newswire translation will apply equally well to other genres and domains, especially highly technical ones.
We are exploring ways to tune general-purpose machine translation to specific scientific domains, especially in S&T analysis scenarios.
Objectives:
Build on our success in our first domain, semiconductors, and transfer this to a new domain, chem-bio
Extend our evaluation and assessment methodology to ensure that system improvements are analytically relevant
Carry out in-depth discussion with sponsor-side S&T analysts, to better understand analysis workflows involving machine translation.
Activities:
Preparing a development corpus in a new domain, chem-bio, including supervising an external translation contractor.
Testing and extending our existing tuning techniques in new domain.
Additional human assessment exercises:
• Using in-house microelectronics experts for existing domain
• Replicating analytically relevant document triage task with machine output on new domain.
Scaling system up to even larger volumes of general-purpose training data, while also making better use use of limited domain data.
Impact:
Improved sponsor effectiveness and efficiency for document triage and higher-level analytical tasks
Better understanding of relationship between translation quality and analytic utility, allowing MITRE to better advise sponsors on MT-related matters.
Public Release No:09-1208
[Presentation]
Exhibit Date(s):May 5
Malware Phylogenetics
Primary Investigator:Chase, Melissa P.
Problems:
The nature of malware threats has evolved from widespread outbreaks for the sake of notoriety to large numbers of targeted attacks motivated by economic gain. In this environment it is critical for end-users, researchers, investigators, and security tool vendors to have a better understanding of the relationships between malware families and variants in order to improve detection, protection, and response.
Objectives:
We will apply phylogenetic and comparative genomic techniques to malware to cluster malware threats and understand the evolutionary relationships within and between clusters.
We will use bionformatics and machine learning techniques to identify differences in development and tradecraft across instances of malware.
Activities:
We created a data set of malware, extracted features from these samples, and used these features to create phylogenetic models, using a variety of algorithms (distance-based, character-based, and complete composition vectors). We applied these techniques to cluster and classify packed malware. We used comparative genomic techniques (synteny analysis) to trace the development history of modules. We used different machine learning and language modeling approaches to identify different compilers, versions, and parameter settings. We have begun to explore different ways to visualize the results of our analyses.
Impact:
Applying bioinformatics techniques to malware analysis improves the classification of malware and suggests evolutionary relationships. This helps malware analysts and criminal investigators understand their malware collections as a whole and determine how new malware fits into the collection, which aids in triage and reuse of prior analysis. Characterizing the development environment and tradecraft may suggest attribution leads. Understanding the development history of malware may lead to better prediction of how malware will evolve in the future, and in turn lead to improved protection for end-users.
Public Release No:09-1265
[Presentation]
|
