Defense Intelligence Carves a New Niche
Traditional functions may spin off from warfighters to information mavens.
Unified military operations are leading to a redistribution of intelligence functions as the U.S. Defense Department transitions into a network-centric world. Sensors and shooters once belonged to the same family of operators. Now, sensing, analysis and dissemination of intelligence information are moving into a realm apart from the weapons delivery process.
The traditional approach of dedicated sensors or observers guiding munitions from designated delivery systems has been used in warfare since the invention of gunpowder. In modern times, land-based artillery, aircraft and naval guns would bring fire to bear on targets identified and located by either onboard or remote, but mission-specific, sensor systems. The advent of multimission targeting platforms, such as the Joint Surveillance Targeting Attack Radar System (Joint STARS), deployed in joint operations accelerated a new trend—the use of various generic sensors to supply data to many different weapons systems.
This portends changes in procurement for both warfighting platforms and intelligence assets. Dedicated sensor platforms will combine data with input from other land, sea and air systems throughout the battlespace. And, their intelligence product will be more widely disseminated than envisioned even a few years ago.
The challenge over the next few years will be for the intelligence community to evolve from a production orientation to a user orientation, predicts John P. Stenbit, assistant secretary of defense for command, control, communications and intelligence. Speaking in a SIGNAL interview, Stenbit says he foresees increased funding for defense intelligence, but it will not be allocated along traditional lines. A broader variety of decentralized intelligence systems will result in increased numbers of smaller contracts. The defense intelligence effort will be more collaborative.
Historically, the elements of tactical intelligence were ceded to the military, while political intelligence had been the purview of other organizations. Not only does Stenbit observe that this separation has become increasingly awkward, he also believes that this awkwardness pales in comparison to what lies ahead if intelligence is not reformed.
This is a major issue in the Defense Department’s transformation, he notes. The new emphases on jointness, long-range operations, precision strike, and global positioning system targeting are generating enormous pressures on the defense intelligence infrastructure.
“We were used to a concept where the person finding the target at which we would like to shoot belonged to the same bureaucracy as the shooter,” he relates. “Now, we are moving away from that concept at a very rapid rate. This changes the definition of tactical intelligence.
“The transformation of going from a common bureaucracy for the target finder and shooter to independent bureaucracies for each one is the heart and soul of command and control issues,” Stenbit declares. The degree of pressure on the intelligence community to handle these kinds of issues has not been recognized yet, he states.
“Since we are not going to predict very well what the military problem is that we are going to fix, it is much more important to have some ability to do everything rather than lots of ability to do one thing,” he says. “In intelligence terms, that is the difference between search and operations.”
Stenbit envisions a ubiquitous intelligence future in which information is collected everywhere by everyone for dissemination to virtually every person involved in a battle. Information must be available not only to the commander and the “production shop,” but also to every potential user, he declares.
He cites Intelink (SIGNAL, October 2000, page 27) as a good model of an open system wherein any authorized user can obtain good information. However, while characterizing Intelink as a noble beginning, he allows that it is nowhere near the omnipresence needed in the community.
For example, the Joint Strike Fighter, with its advanced radar system, could serve as a significant intelligence platform by transmitting radar data back to a processing platform or center. Stenbit notes that tactical operators, who are not accustomed to shipping their data into the intelligence pool, will have to adjust to this new sharing approach. “We are going to be sweeping more square miles per day of hostile territory with pretty good sensors,” he says. This will contribute to the blurring of the lines between tactical data and intelligence, which Stenbit describes as part of the symmetrical revolution currently underway.
Throughout the U.S. military, many units and organizations have been taxed by an increased amount of force deployments. Stenbit relates that this also applies to the Defense Department’s intelligence assets. They have been affected severely by the number of commitments imposed on them relative to the amount of funding they have received. “The size of the [intelligence] system has been ill-matched to the number of places assets have had to be in simultaneously,” he adds.
Platforms such as Rivet Joint aircraft are being used more often than expected, and this contributes to hardware wear and tear as well as personnel stress. This “low-density, high-demand” syndrome is becoming more prevalent. Stenbit is calling for increased funding for modernizing these platforms and their technologies. “All big-picture defense intelligence systems need modernization programs to occur,” he emphasizes.
Increasing the flow of intelligence information will require greater bandwidth access, but Stenbit is not worried about bandwidth capacity. Of bandwidth shortage, he says “it is an interesting historical perspective that is no longer correct. Ten years ago you could have said the same thing about processing power, but Moore’s Law continues, and processing power continues to double every 18 months. For bandwidth, we now can get 80 colors per fiber, and bandwidth capacity is doubling about every four months.
“There always will be users who are bandwidth starved,” he continues. “But, in the 10- to 15-year timeframe we are talking about [to provide advanced intelligence connectivity], it is a bad assumption that bandwidth will be a problem in general.”
Nonetheless, bandwidth does remain a key enabler for defense intelligence, and Stenbit cites the ability to transport optical switched data as one important area. Technologies involving archiving and distribution of information also are vital. With Web-based information flow increasing in usage, tools that improve data searching and retrieving will be “very applicable,” he says.
Sensors are another area with room for improvement. “I think we have the technology available to do a lot better in sensing where we have more variety of sensors [that are] less production-oriented,” Stenbit suggests. Technologies have improved to the point where building prototypes is not as expensive as it was just a few years ago, he offers, and the department can take advantage of that aspect.
“The changing dynamics of intelligence and operations, combined with available technologies such as processing and sensing, will lead us to a broader set of sensing opportunities,” he adds.
The trend toward information technology and services outsourcing is likely to continue, but Stenbit cautions against taking this path without proper study. “Outsourcing … is going to be a very important issue, but you must understand where that revolution is going before you do it. It’s bad business for the government to start contracting without knowing what it wants to do,” he states.
This cautionary approach also extends to government funding of commercial operations. For example, purchasing remote sensing imagery from commercial sources for intelligence applications “is a wonderful idea,” Stenbit offers. However, some commercial imagery companies want the government to provide venture capital funding to develop and deploy assets that could serve the intelligence community effectively. The government in turn would be purchasing imagery from technology that it helped develop.
“That’s not commercialization; that’s government funding of companies,” Stenbit charges. “I’d love for the commercial imagery companies to be successful, just as I’d love for the commercial electronics firms to be successful.” He is opposed, however, to the government paying for commercial products that it helps develop. For example, this could lead to a situation in which the government finds itself the prime customer for a product that is not generating sales elsewhere, and the commercial provider teeters on the brink of bankruptcy and requests additional government funding. Without that additional funding, the firm would collapse and the government would lose its source of goods or services. “That’s the world’s worst position to be in for the Defense Department,” he declares.
Stenbit suggests that several steps can be taken to increase satisfaction with both Defense Department demand and commercial sector supply. Instead of buying an 80-percent solution and transforming it to meet military needs, the department could adopt a more flexible approach to defining its requirements.
“If we could focus on how we could tell what we want instead of on meetings, we might be able to open up some paths that would allow us to use the commercial advantage—it is here today. It has a business base beyond the military customer; its maintenance and research and development have been spread out; and more importantly, it will be there when we want it,” he explains.
Achieving this will require the defense intelligence community to describe “what acceptance is” in enough detail, Stenbit offers.
He suggests a new approach to contracting that could entail both risk and reward for contractors while benefiting the defense intelligence community. “Suppose I had a contract that spelled out the acceptance criteria, and another company emerged—before the contract was finished—with something better. Then, I could cancel the contract and buy the new material. By definition, I’m better off.
“But that requires that we spend money defining the metrics that measure results. And, I don’t know whether we currently can do that,” he allows.