Army Transformation to the Future ForceA Race for Speed and Precision

Dr. John ParmentolaDirector for Research and Laboratory ManagementSeptember 23, 20037th Annual High Performance Embedded Computing WorkshopLincoln Lab/MIT

Purpose Describe Armys vision of the Future Force

Address Army needs and challenges in High Performance Embedded Computing (HPEC) to enable Transformation to the Future Force

Increased strategicresponsivenessFuture Force for Full Spectrum of MissionsHighLow Urban

Open rolling terrainMajor Theater WarEnvironmentalComplexitySpectrum of ConflictStability and Support OperationsSmall Scale Contingencies

Brigade in 96 hrs; Division in 120 hrs; Five Divisions in 30 daysFight immediately upon arrivalSimultaneous air and sea liftRender Previous Ways of Warfighting Obsolete

Seeking A Revolution in Capabilities . . . Smaller, Smarter, Lighter & FasterTodayFuture Force~100 lb. load< 40 lb.effective load< 20 tons> 40 mphS&T Mission -- Accelerating the Pace of Army Transformation70+ tons0 mphFit the C-130 Crucible Implement System of Systems

Future Combat Systems (FCS) Maneuver Unit of Action (Brigade Equivalent)Manned Systems: MCS60 ICV 84 NLOS-Mortar24 NLOS-Cannon18 C2V 49 R&SV 30 Medical Vehicle29Medical VehicleUnmanned Systems: UAV (CL III/IVa)56 UAV (CL I) 54 NLOS-LS 24 MULE & ARV-A (L) 54, 27 UGS 157 SUGV 81Infantry Carrier Vehicle (ICV)

Mounted Combat System (MCS)Non-Line-of-Sight (NLOS) - MortarNon-Line-of-Sight (NLOS) -CannonCommand & ControlVehicle (C2V)Land Warrior (LW) FCS 2550Recon & SurvVehicle (R&SV)Small UnmannedGround Vehicle (SUGV)Unattended Ground Sensor (UGS)MULE & Armed RoboticVehicle Assault (Light)(ARV- A (L))Non-Line-of-Sight (NLOS) - Launch System

Unmanned Aerial Vehicle (UAV)(CL I)Unmanned Aerial Vehicle (UAV)(CL III/IVa)Network

Unit of Action Networked Battle Command Sensors Information gatherers

Network Architecture Information management (processing, routing, dissemination)

Nodes or platforms (soldiers, ground vehicles, aerial vehicles) Information receivers, gatherers and users (FCS ~ 3300)

Shooters Information receivers and users

FCS Key HPEC ChallengesNetwork-centric/Collaboration-centricLocal HPEC capabilities to manage the complexity of large amounts of data and informationHPEC will enable local nodes to perform complex analysis and data management functions with reduced use of bandwidthRobotics integrated into forceReal time scene understanding for maneuver and threat analysisReal time RISTA from multiple sensors and sensing modalities Increased reliance on extended range engagementHPEC crucial for smart munitions accurate target IDWide range of distributed sensors, each needing HPECCapable of air-mobile operations - DoD strategic and tactical liftNew smaller sized force elements require small embedded processors to meet demanding computing requirements

Objective Force Warrior (OFW)Integrated Combat SuitHead Borne Vision EnhancementPhysiological Status MonitoringPersonal NavigationRobotic MuleSituational AwarenessNetworking Digital RadioWarrior Team collaborationHorizontal data fusionMicro UAVRobotic Mule

OFW Key HPEC ChallengesReal time situational awarenessConnection and exploitation of information from FCS networkAccess to Common Operating PictureVertical/horizontal position/navigationStatus of physiological readiness and vital signsTwo-way language translationEmbedded trainingPlanning and rehearsal of complex missionsImmediate access to Tactics, Techniques and Procedures (TTPs)

Low power HPEC is an essential OFW need

HPEC Challenge: Communication FunctionsNetwork may be source of surprise computational problems and system congestion / bottlenecksWill be the largest, most ad hoc, dynamic and mobile Unit of Action network ever deployedA ground traffic control system integrated with an air traffic control system that will work in all environments and conditionsMix of SATCOM, platform and soldier radio networksRapidly formed and broken links rates, i.e., chaotic networkRouting and network management may require:Active network technologyVery advanced routing methods heavy computationDemand for fully distributed computing Power-aware routing, low power computing

HPEC Challenge: Communications SecurityHPEC can assist in data encryption Process is very complex, i.e., multilevel security, and computationally intensive especially for authentication.Challenge is striking a balance between security and performance, interoperability, reliability, . . .Main processing challenge is complexity of decoding encryption techniques used in authentication Symmetric ECC RSA Computational Encryption Key Size Key Size Complexity Key Size 56 112 512 80 160 1024 112 224 2048 128 256 3072 192 384 7680 256 512 15360 Increases with key size

HPEC Challenge: Aided / Automatic Target RecognitionTypical ATR systems analyze a digital representation of a scene and locate/identify objects of interestWhile conceptually simple, ATR has extremely demanding I/O and computational requirementsImage data are large, can be generated in real-time, and must be processed quickly so that results remain relevant in a dynamic environmentFuture Force ATR will incorporate more than one sensor and use more data from other sources

HPEC Challenge: Data FusionFunctions needed for multi-source fusionDistributed, real time fusion is needed to minimize command center message inundationFCS will increase the volume and complexity of data more local fusion is needed

Army S&T Program That Uses HPEC Eye in the SkyDemonstrate onboard automated payload management functions to facilitate tasking and cross-cueing of sensorsOnboard, autonomous sensor management and data fusionSoftware to integrate RF Tags SA data into the COP for Blue Force trackingMultiple Sensors - Radars (GMTI, SAR, GPEN, FOPEN) - EO/IR - Hyper-Spectral - Electronic Support Measures - LADARIntegrated on a Class IVA UAV

HPEC Challenge: Affordability101,00010020021998199510,0002005INTEL12 GFLOPS/$M100x In 7 YearsSKY225 GFLOPS/$MProjection:2007: 60 trillion flops/$M2010: 360 trillion flops/$MCHALLENGE Develop and incorporate the most affordable embedded information technology availableReconfigurableHPC10XExponentially Improving HPEC Affordability Transitioned to DoD Users34 TOPSXeon Cluster3.1 TFLOPS/$M APPROACH Leverage commercial investments in computer architectures Develop portable embedded DoD applications using middleware standards Leverage DARPA and other DoD efforts in emerging architecturesSource: Dr. Rich Linderman, RADC

Trend in Computer SizeSource: Dr. Ray Kurzweil, Kurzweil TechnologiesPhoto courtesy of Cris Pedregal-Martinhttp://www-ccs.cs.umass.edu/~shri/iPic.htmliPic

Trend in Growth of ComputingSource: Dr. Ray Kurzweil, Kurzweil Technologies

SummaryThe Armys Future Force will have a critical need for HPEC technologies throughout its system of systemsAutonomous sensing and sensor fusionComplex communications tasks The acute Army challenges for HPEC are cost, power consumption and physical size The Army is looking to academia and industry for advances in HPEC to enable its vision of the Future ForceArmy TransformationA Race for Speed and Precision

Excerpts from TRADOC PAM 525-3-90 Maneuver O&O: The Army operates as part of a joint and often multi-national force. Units of Employment (UE) are tailorable (comparable to a Division composed of 3 to 6 brigades), higher level echelons that integrate and synchronize Army, Joint and Multinational forces for full spectrum operations at higher tactical and operational levels of war. The Unit of Action (UA) is the tactical warfighting echelon of the Future Force and comprises echelons at brigade and below. It will be part of the joint team. The UA is not a fixed organization. It has the capability to command and control up to six Combined Arms (CA) Battalions. A UA can serve as an Army Forces (ARFOR) component headquarters for the Joint Task Force. The UE fights battles; the UA orchestrates multiple engagements to win battles.

The Objective Force Warrior system will provide the soldier with a heads-up capability that will display Infrared (IR) and Thermal images and situational awareness information on digital map backgrounds.Active networks are a novel approach to network architecture in which the switches of the network perform customized computations on the messages flowing through them.

This approach is motivated by both lead user applications, which perform user-driven computation at nodes within the network today, and the emergence of mobile code technologies that make dynamic network service innovation attainable.

Active networks allow their users to inject customized programs into the nodes of the network.

Active architectures permit a massive increase in the sophistication of the computation that is performed within the network. They will enable new applications, especially those based on application-specific multicast, information fusion, and other services that leverage network-based computation and storage. Furthermore, they will accelerate the pace of innovation by decoupling network services from the underlying hardware and allowing new services to be loaded into the infrastructure on demand.

Given bandwidth constraints, advanced routing methods will be necessary to maintain low latency and fast upload and download times.Network security measures add an additional computational burden.Organize the discrete pieces of data for useIdentify what is physically out thereCorrelate like informationAggregate discrete entitiesResolve Information ConflictsInterpret events and actionsDetermine how entities are working together & what they are doingPredict what the enemy will do next and how it will affect your plansAssess what you must do to improve the fidelity of the informationA Reason for Optimism

Portable embedded software applications are developed using standard middleware implementations such as the Common Object Request Broker Architecture (CORBA), an open, vendor-independent architecture and infrastructure that computer applications use to work together over networks.

CORBAs Internet Inter-ORB Protocol (IIOP) is at the heart of network-based distributed computing. Using this standard protocol, a CORBA-based program from any vendor, on almost any computer, operating system, programming language, and network, can interoperate with a CORBA-based program from the same or another vendor, on almost any other computer, operating system, programming language, and network.

Java is being standardized by the International Standards Organization (ISO). Java is another approach to middleware that is complimentary to CORBA. Java technology and Extensible Markup Language (XML) are a natural match for creation of applications that exploit the web of information where different classes of clients from a traditional phone to the latest smart refrigerator consume and generate information that is exchanged between different servers that run on varied system platforms. The portability and extensibility of both XML and Java technology make them the ideal choice for flexibility and wide availability and wide availability of this new web.(Photo courtesy of Cris Pedregal-Martin)

IPic - A Match Head Sized Web-Server

The single chip computer in the above picture runs the iPic web-server, the world's tiniest implementation of a TCP/IP stack and a HTTP web-server. The chip above is a complete micro-computer, and it includes all components of a complete computer on a single tiny micro-chip (this includes the CPU (central processing unit), memory, serial port interface circuitry, and clock oscillator). The chip is connected directly to an Internet router, which is essentially the same as an Internet connection from an ISP. When you visit the iPic Web-server, by clicking on this demonstration link, (or this mirror), your web-browser connects to the chip shown in these photographs and the web-pages you see are sent to your web-browser directly from the tiny chip. Source: http://www-ccs.cs.umass.edu/~shri/iPic.html