A -- Tools to Identify an Enemy's Command Organization and Manage its Disruption

This is a Request for Information (RFI) from interested sources by the Defense Advanced Research Projects Agency (DARPA) Information Exploitation Office (IXO) regarding the ability of firms to conduct research in the area of tools to identify an enemy’s command organization, estimate its performance, and manage its disruption. DARPA is exploring the availability of computational techniques relevant to the identification and disruption of decision-making processes in enemy command organizations. To this end, DARPA invites brief white papers. Upon review of the white papers, DARPA will select a limited subset of papers that are judged as particularly relevant to DARPA's interests. The authors of the selected papers will be invited to discuss their concepts at a workshop to be held in late January 2005, location to be announced later. Attendance at the workshop is by invitation only. The workshop will be UNCLASSIFIED. DARPA is interested in white papers even if the authors are unable to attend the workshop. The motivation for this topic is the US military’s growing emphasis on Information Warfare. Understanding and affecting the human decision-making processes within an enemy command organization is a key aspect of Information Warfare [1]. This RFI focuses specifically on managing the efforts to degrade the timeliness and quality of the decisions produced by an enemy command organization. A command organization consists of individuals and sub-organizations at different levels of responsibility, linked by a variety of relations and procedures, both hierarchical and peer-to-peer, both formal and informal. Organizations of interest range from the more structured command systems of a conventional military to the less structured and fluid leadership of guerillas and other irregular combatant forces. The enemy command organization's structure and processes are often little known to the friendly forces, and difficult to observe. Such organizations evolve and change rapidly, especially during a military operation, due in part to physical attrition as well as to other factors. The flows and interactions of decisions and information in such a dynamic system can cause a range of complex phenomena. Thus, by applying a proper set of dynamic, time-dependent impacts (either physical or informational) on elements and connections of the enemy command organization, it may be possible to induce delays and errors in the enemy decision-making processes that are advantageous to the friendly forces, at the right place and time. Tentative illustrations of some such effects are in discussed at references 1 through 5 at the end of this announcement. Within this overall capability, consider the following four sub-problems. DARPA is interested in computational techniques and tools that would assist the US armed forces to solve these problems in real-time, during the conduct of military operations. 1. Identify the enemy command organization. By exploiting limited, sparse observables, such as observations of enemy movements, attacks, communications (often non-electronic and/or encoded) between some of the command nodes, determine the most likely topology, characteristics, relations and decision-making processes of the enemy command organization. Account for rapid adaptations and changes (often on a scale of hours and even minutes) that are likely to be occurring in that organization and processes. 2. Plan a coordinated course of actions against the enemy command organization. Given the available (usually constrained) means of physical and informational impacts on the enemy command, devise an effective combination and sequence of actions. Provide for probing. Leverage dynamic system effects. Account for significant uncertainty in the friendly knowledge of the enemy command organization, as well as its time-dependency and adaptability. 3. Predict the impact of the planned actions. Given the knowledge of the enemy command organization (such as obtained in the identification process, above) and the proposed friendly course of action, determine the likely impact these actions would have on the enemy command decision-making, particularly on its timeliness and accuracy. Time and effort involved in such modeling and prediction should be consistent with its use in real-time. Address the issue of validating a predictive model with respect to the real world. 4. Control the unfolding execution of the selected course of action. In real-time, exploit the available observables (generally sparse and partially inaccurate), assess the state of the unfolding operation and suggest corrective adjustments. Response Instructions DARPA does not intend to award any type of award instrument based on responses to this RFI, or otherwise provide compensation for the information solicited. However, responses to a DARPA RFI can be the basis for a new program or solicitation. Any information submitted in response to this RFI is provided voluntarily. Respondents are advised that DARPA is under no obligation to acknowledge receipt of information received, or provide feedback to respondents with respect to any information submitted under this RFI. This RFI is open to any capable and qualified sources, including but not limited to universities, private or public companies, Government Laboratories, and Government research establishments. Responses shall be no more than 5 pages in length and no more than 2500 words. A single response must address only one computational technique, or one coherent integration of several techniques, that focuses on one or more of the four problems specified above. Any proprietary concepts or information should be clearly identified as such. Format specifications include 12 pitch or larger type, single spaced, single-sided, and 8.5 by 11 inches with 1 inch margins all around the page. The page limitation includes all attachments, etc. Preferred file types are .doc/.pdf/.ppt. The application will NOT accept Zip files or .exe file types. Outline of the white paper: • Name, affiliation, and email of the author(s). • Title: one sentence description of the technique/approach. • The problem (one or more of the four described above) addressed by the technique. • Brief technical description: input, output, model, algorithm. • Arguments in support of the technique’s applicability to the target problem, and of its feasibility, preferably of experimental nature. • Current applications, if any. • Extensions required to solve the target problem. • Up to three references to technical papers, preferably with URLs. DARPA will employ an electronic upload process for response submissions. Responses to this RFI are due no later than 1:00pm, Local Time, Arlington, VA, on Friday, 10 December 2004. Responders may find submission guidance at: http://www.darpa.mil/ixo/solicitations/SN05-09/index.htm. Organizations must register at: http://www.tfims.darpa.mil/baa to respond. One registration per white paper should be submitted. Organizations wishing to submit multiple papers should complete a single registration for each response. The deadline for registration is 8 Dec 2004 at the URL listed above. By registering, the Responder has made no commitment to submit. The Government anticipates that responses to this RFI may be either unclassified or classified. Unclassified responses should be submitted in accordance with submission guidelines at: http://www.darpa.mil/ixo/solicitations/SN05-09/index.htm. Classified submissions shall be in accordance with the following guidance: When registering for this RFI, responders with classified white papers must enter the following in the field titled, “Comments Regarding this Effort”: Classified Response. In submitting a classified response, you must first receive permission from the Original Classification Authority to use their classified information and you must identify in the response the applicable Security Classification Guide used to classify the information in the response. Collateral Classified Information: Use classification and marking guidance provided by previously issued security classification guides, the Information Security Regulations (DoD 5200.1-R), and the National Industrial Security Program Operating Manual (DoD 5220.22-M) when marking and transmitting information previously classified by another original classification authority. Classified information at the Confidential and Secret level may only be mailed via U.S. Postal Service (USPS) Registered Mail or U.S. Postal Service Express Mail. All classified information will be enclosed in opaque inner and outer covers and double wrapped. The inner envelope shall be sealed and plainly marked with the assigned classification and addresses of both sender and addressee. The inner envelope shall be addressed to: Defense Advanced Research Projects Agency ATTN: SN05-09, DARPA/IXO 3701 North Fairfax Drive Arlington, VA 22203-1714 The outer envelope shall be sealed with no identification as to the classification of its contents and addressed to: DARPA Security & Intelligence Directorate, Attn: CDR 3701 North Fairfax Drive Arlington, VA 22203-1714 All Top Secret materials should be hand carried via an authorized, two-person courier team to the DARPA CDR. Responses may NOT contain Special Access Program (SAP) Information or Sensitive Compartmented Information (SCI) Data. Proprietary Data: All information submittals containing proprietary data should be appropriately marked. It is the respondent's responsibility to clearly define to the Government what is considered proprietary data. Important Dates: • Register by 8 Dec 2004 • Submit white papers by 1300 local time, 10 Dec 2004 • Invitations to workshop sent by 17 Dec 2004 • Workshop will occur approximately 18-20 or 25-27 Jan 2005 Original Point of Contact Technical questions may be emailed to: Dr. Alexander Kott, Program Manager ([email protected]). Other questions may be directed to Michael Blackstone, Contracting Officer, Phone (571)218-4804, Fax (703)696-2208, Email [email protected] References [1] Department of the Army, “Field Manual 3-13, Information Operations,” 2003 http://globalsecurity.org/military/library/policy/army/fm/3-13/ [2] Hubbard, P., et al., “Managing Responsibility and Information Flow in Dynamic Team Decision-Making," Proceedings of Command and Control Research and Technology Symposium, 2002. http://citeseer.ist.psu.edu/hubbard02managing.html [3] Kott, A., et al., “Instability of Distributed Decision Making in Command and Control Systems,” Proceedings of the 2001 American Control Conference, Arlington, VA, June 2001 [4] Carley, K., et al. 2003, “Destabilizing Dynamic Covert Networks,” in Proceedings of the 8th International Command and Control Research and Technology Symposium. http://www.casos.ece.cmu.edu/resources_others/a2c2_carley_2003_destabilizing.pdf [5] Kott, A., and Krogh, B., “Pathologies in Control: How C2 Systems Can Go Wrong,” in Kott, A. (ed.), Advanced Technology Concepts for Command and Control, 2004