Intelligence Family
Amid all of the discussion in recent years about the transition
to network-centric warfare, the Distributed Common Ground
System is one area where talk is being matched by action.
By Peter A. Buxbaum
Amid all of the discussion in recent years about the transition to network-centric warfare, the Distributed Common Ground System, a family of programs with common elements designed to meet the intelligence, surveillance and reconnaissance needs of each of the armed services, is one area where talk is being matched by action.
DCGS, as it is being developed, provides an interoperable architecture for the collection, processing, exploitation, dissemination and archiving of geospatial and other forms of intelligence. While DCGS includes common elements, it is not, strictly speaking, a joint program. The Army, Navy, Air Force and Marine Corps are each working on their own DCGS platforms.
A key upcoming development in the related programs is the anticipated delivery in the coming months of capabilities under the latest Air Force version of the program, which will focus on video and imaging intelligence.
DoD’s Net-Centric Operations Warfare Reference Model posits several key system elements that define network-centricity. Chief among these are a service-oriented architecture (SOA), which does away with system-to-system integration in favor of reusable, loosely coupled application elements, and a data strategy that promotes visibility and interoperability. DCGS’s integration backbone and the Air Force’s DCGS Block 10.2 exemplify this network-centric approach, analysts say.
“DCGS is at the cusp of legacy and future systems,” said Jeffrey Roncka, vice president of CRA International, a defense consultancy. “The focus is on coming up with a family of common grounds stations that take multisource intelligence feeds using different types of data from 10 different intelligence pipes and putting them all in one pipe with a set of common standards that everyone can read off of.”
Backbone Integration
The DCGS Integration Backbone (DIB) provides a common operating environment that represents a transition from the old DoD federated architectures, in which systems were integrated individually on a point-to-point basis. “It’s the vision of the Global Information Grid,” said Roncka. “We’re not quite there yet, but DCGS is a window to a future of truly seamless information and intelligence processing.”
“The DIB provides a standard scheme for doing metadata tagging and includes a common data thesaurus,” explained Chris Jackson, chief of integration at the Joint Transformation Command for Intelligence, a unit of the Joint Forces Command, “so that all the service DCGS systems recognize each others’ data.”
Legacy systems tend to collect intelligence data from sensors and other sources in a stovepiped fashion, Jackson added. “The process of stovepiping continues all the way to the dissemination and exploitation of the intelligence,” he said. “The intent of DCGS is to allow data to be shared early on in the process. The DIB allows raw and rudimentary data to be discovered by systems and capabilities that are not native to collector of that data, so that data coming from the Air Force, for example, could be accessed and processed across all DCGS systems. The service-oriented architecture allows data at various processing stages to be discovered by other intelligence analysts and users.”
The DIB, which began delivering capabilities two years ago, has been declared a joint element of DCGS by the Joint Transformation Command for Intelligence. It slices across all of the service DCGS programs.
“Since then we’ve gone through a couple of spiral upgrades and added additional capabilities,” said Mark Bigham, director of business development for tactical intelligence systems at Raytheon, the Air Force’s prime contractor for DCGS and DoD’s prime contractor for the DIB.
DCGS does not eliminate the use of legacy systems, however. Instead, each service must adapt its legacy systems to DCGS’s network-centric environment. “Often these systems go back multiple decades,” said John Beck, head of tactical intelligence business development at Lockheed Martin. “They were purpose-built and not designed to connect to others. In some cases, a service has to adapt its database so others can get access to it. In other cases, they need to allow other systems access to data and processes, and that sometimes requires re-architecting systems.” Lockheed is working on all four DCGS service teams.
The Air Force’s Block 10.2 is the largest ongoing DCGS project in terms of dollars and scope, according to Bigham. The first capabilities for DCGS Air Force Block 10.2 are about to be delivered. Following several months of testing, they are expected to become operational next year. “The first spin of Block 10.2 will be focusing on video and imaging intelligence,” said Bigham.
Block 10.2 will replace the earlier iteration, which was based on the old computer architecture of point-to-point interfaces. “Those are expensive and difficult to maintain,” said Bigham.
The more modular design of a service-oriented architecture and the extensive meta-tagging of data for visibility and searchability will also provide dividends when it comes to intelligence analysis, according to Bigham. “An intelligence tool called Workflow allows users to rapidly reconfigure applications,” he said. “The metadata catalog provides for Google-like searches of information.”
Enhancements to DCGS’s search capabilities are currently under development.
Legacy Adaptation
One of the challenges DCGS faces is in adapting legacy networks, applications and data to a service-oriented and interoperable environment. Failing to do so would render the data and information resident in these systems of little or no use.
Some of the nuts and bolts that will enable DCGS to exploit legacy capabilities are still under development, with a company called Modus Operandi working on several.
“In the current environment, legacy systems operate in isolation and don’t share information well,” said Tod Hagan, Modus Operandi’s director of advanced technology. “One of the reasons is that different systems store data in different formats.”
Older systems often store data in binary files that aren’t well understood by modern systems. “A document might be 30 years old, but it may still hold critical information,” said Hagan. “That information needs to be liberated and it needs to be plugged into the service-oriented architecture so that we can find the data.”
Modus Operandi is working on a system that would automatically reconfigure information from legacy systems into an intermediate format, from which it would be pulled by intelligence users and retranslated it into whatever format the user required.
Once the information is unlocked and exposed as a service on the new architecture, another related challenge is to facilitate the communication and exchange of that information among different user communities. The challenge here is to synchronize and harmonize data across legacy systems by standardizing the vocabulary used for specific data elements.
“The U.S. military has over 50 formats for storing geolocation data,” said Hagan. “Once all of them are exposed to the same network, they need to be speaking with the same vocabulary.”
Another challenge involves extracting business logic and algorithms held in legacy systems for use in a network-centric system such as DCGS. “The Army might have an algorithm for pinpointing an emitter on the battlefield,” said Hagan. “This piece of mission-critical information could be written with many thousands of lines of code, written in Fortran or some other older software language, and embedded in a legacy system. We are developing ways to unwrap those algorithms and migrate them to the service-oriented architecture. The alternative is to lose the information altogether.”
The huge volumes of data to be processed through DCGS presents a challenge to intelligence analysts and users to discover data and information relevant to their needs. “Each of the legacy systems has six to 20 data services,” said Hagan. “Multiply that by 13 systems and that becomes a lot of data to search through and find critical information.”
Modus Operandi is working on a project that will allow users to search for intelligence using keywords that are also qualified by time and space parameters and aided by a graphical display of the discovered information. “This allows the analyst to constrain a search geospatially and temporally,” said Hagan. “Keyword searches come back in terms of time and geolocation on a map. The analyst can scroll through time and see the information change with reference to a time line or he can scroll through space and the information available will change on the map.
“So, for example, in a high tempo environment like Iraq, an intelligence analyst can quickly determine what happened on a specific street in Baghdad this week or a specific building yesterday,” Hagan continued. “The alternative is to pull that information out of all the legacy intelligence systems individually. The point is to retrieve the information quickly because operations are constrained by time.”
Another tool Hagan and his team are working on involves the development of autonomous semantic agents that act on behalf of intelligence analysts to find actionable intelligence automatically within a tactically useful timeline. “With all the data exposed now the analyst will be overwhelmed,” he said. “The autonomous semantic agent will do some of the analyst’s work for him. There will always be a human in the loop, but the analyst cannot be constantly querying the system.”
The semantic agents will work by notifying the analyst when something important comes up on the system. “We can make agents extremely smart and based on current mission requirements and tactical objectives,” said Hagan. “The system will automatically provide actionable intelligence in a tactically useful time line. It could tell an analyst, for example that something just happened and you have few minutes to act.”
The point of developing all of these tools, said Hagan is to increase the velocity of the intelligence cycle.
Standardized Capabilities
Given the fact that each armed service is developing its own DCGS systems, the question arises whether the overarching DCGS program achieves the level of interoperability, and therefore network centricity, that it promises. “DCGS actually started as a joint program,” said Bigham. “But early on the services discerned significant differences in their missions, so they all went their separate ways.”
For example, the Air Force, which views targets from tens of thousands of feet, could not use the same visualization tools as the Army, which needs to look at terrain from the ground level. “There are differences in mission and of scale that couldn’t be accommodated in a joint program,” said Bigham.
“In theory, by following the same standards, you end up with an interoperable architecture,” said Beck. “Acquisition of each of DCGS systems is still under the control of each specific armed service, but they are able to share data and services more readily now.”
The point is not to standardize capabilities across all the armed services, explained Jackson, but to allow the cross-exploitation of data derived from diverse systems.
“Each service has unique requirements with respect to their capabilities to process and utilize data from ISR systems,” he said, “but there is a common implementation capability to all of the DCGS programs. When fully enabled, the DCGS Integration Backbone will permit users of the Navy DCGS to discover and have free access to information from another DCGS. JFCOM also has the responsibility to identify what the joint requirements are across all DCGS systems, so that common components can be developed.”
“There is commonality in certain areas,” said Thomas Hennies, director of DCGS enterprise programs at BAE Systems. “The best way to characterize DCGS is that each service is trying to satisfy its own users’ needs and accomplish its own missions while taking into consideration the methods and approaches being used to achieve interoperability.”
BAE supports both the Navy and the Air Force DCGS programs.
“Image storage and dissemination is a key component of every DCGS system,” Hennies added. “They support storage and organization of image data wherever it might come from. We are working on some tools that will allow analysis of image intelligence as well as annotation of images so that users can describe what they saw. There are also tools being developed that support search based on geolocation, date, quality and suitability for targeting.”
DCGS is already exhibiting attributes of interoperability, according to Hennies. “The different DCGS systems are able to share Web services,” he said. “The point is not to have the individual services build or deploy the same capabilities, but rather to allow them to share and make them usable by other elements of the military and to allow intelligence users and analysts to be able to collaborate and work on problems together versus independently.”
All this is made possible with common data standards, advanced search capabilities, and the service-oriented architecture. “The vision is to have deployed forces in Bahrain, for example, using the DCGS capability to reach back and touch folks sitting at Naval Operations or at Langley Air Force Base to collaborate with intelligence analysts to accomplish missions,” Hennies said.
To that extent, developing a family of related programs yields greater capabilities and better reaches network-centric goals than taking a joint approach. “There is a money issue when you do the joint program thing,” said Bigham. “Everything gets lumped into one bucket and feeds off of a single budget. The way things are working now, each armed service is able to set its own priorities for the development of DCGS capabilities. They don’t have to coordinate with other services and get held up by their requirements. Having a single focal point can end up being a choke point when it comes to developing capabilities at the applications level. By developing separate platforms, all they have to do is to plug into the DIB.”
“You’ve already got a robust DCGS up and running in Iraq and to a lesser degree in Afghanistan,” said Jackson. “A big part of DCGS is making better use of the information that is collected. We can collect data once and use it many times. The armed services are getting a lot better at that and this is being enabled by the technology that we are deploying.”
Imagery Management
One of the ways the Army is looking to make better use of DCGS information collected, for example, is through testing at its DCGS Futures Lab of capabilities focused on enterprise data management of satellite imagery and other complex digit formats.
Commercial-off-the-shelf geospatial data management and upstream processing technologies from spatial information management software provider Intergraph, known as TerraShare and Auto Terra Ingest, are currently being evaluated and tested at the Futures Lab with customer sets against the requirements for applicability in future configurations.
TerraShare supports enterprise data management of satellite imagery, aerial photos and other complex file formats in a manner that makes the data easily discoverable and supports dissemination to analyst desktop applications or Web browsers. “It goes a long way towards simplifying the process of discovering relevant intelligence information and delivering that to an application, so that the analyst does not have to spend time doing manual searches, filtering the searches and finding the data and bringing it into his environment,” said Rob Mott, executive director, defense and intelligence solutions, for Intergraph. “The analyst’s preferences dictate automated data feeds of imagery and other data coming from TerraShare to those desktop applications.”
A key technology that couples with TerraShare is a product called Auto Terra Ingest, which automates the preprocessing of the data so that it is more readily usable. “We’re collapsing the decision cycle and eliminating a lot of the mundane tasks associated with putting raw data through its paces in order to generate properly referenced imagery or intelligence products for the analyst. With the use of these two products, we’re able to simplify and create an environment that is much more reliable and provides better sources of information to support improved analysis and the decision-making process,” Mott explained.
Both of those technologies are currently in use by the Defense Intelligence Agency’s Missile and Space Intelligence Center. They have led to dramatic improvements in overall operations, helping analysts save two to three hours a day of manual search time, according to Mott.
Another important aspect of what is being done with these technologies at the DCGS Software Engineering Center, Technology Test Team, is the use of Open Web Services for dissemination of this information. “We’re working with protocols such as the OGC’s Web Feature Service and Web Map Service, so that you have an open standard, vendor-neutral format for transmitting that information from TerraShare across the backbone to other components within the DCGS architecture that would benefit. We’re getting some valuable feedback about the importance of these open Web services and how they support some of the key requirements of the DCGS program,” Mott said. ♦
