As budgets retreat from wartime levels, the tight fiscal environment may force the U.S. Department of Defense and the intelligence community to re-evaluate how to best use, assign and plan geospatial and ISR data. We suggest that the government may be able to do more with less by using mission planning to determine, on a prioritized basis, the information needed to accomplish a specific mission.
For decades, large numbers of assets, such as manned or unmanned aircraft, collected massive amounts of data for analysts to reduce to actionable intelligence. Often, only a small part of this information is used; the rest is stored and rarely, if ever, accessed again. The goal should not be to collect terabytes of raw data, but to more efficiently determine what data needs to be collected.
A different, mission-oriented approach is needed to provide more data processing and analysis power at the sensor/platform level. Sifting and meta-tagging geospatial and ISR data at the scene allows analysts to work more efficiently and provides more timely, usable material for war fighters. In addition, data collected at one point in the mission may drive an additional tasking to improve the tactical or intelligence value of the mission.
The DoD and intelligence communities already are moving toward placing more processing power at the sensor and platform levels. However, software with the algorithms to sift and catalog the data, and widely accepted federal and commercial standards for these algorithms, still need to be developed. Such standards would not just take into account where data was collected and tagged, but perhaps also highlight the specific image or scene of which it was a part.
New software tools are also needed because tighter budgets make acquiring additional hardware platforms difficult. Besides modifying equipment, technology and advanced-sensor processing may facilitate changes to existing doctrine and promote a more flexible approach.
Traditionally, data collected at the strategic level was sifted by hundreds of analysts and then stored for possible reuse. Pulling in large amounts of data and keeping it is optimal from an analyst’s perspective. However, as sensors have become sharper and more complex, the amount of information collected in a single mission has often exceeded data collected only a few years ago.
Information-gathering needs at the tactical level are different. In this realm, timeliness is vital, as war fighters have to be able to act on real-time data. At the tip of the spear, bandwidth and time are in short supply. Because commanders do not have the access or time to wait for material being processed by teams of analysts, it becomes necessary to work smarter with existing assets.
At the tactical level, users must define their needs to developers: target types, probabilities of detection, probability of false alarms, when to engage and the ability to track targets. This represents a different type of intelligence collection that requires lots of front-end processing in the sensors. In-sensor processing, backed up by the right software and standards, will improve detection, identification and tracking.
To meet this goal, mission requirements and needs must drive the process, not the push for more data. A change to a more mission-oriented approach could provide war fighters with critical and actionable data in real time.
Many missions could benefit from in-sensor processing. For example, there are different operational needs and priorities for locating enemy air defenses and tracking troop movements versus locating terrorists or identifying hostages.
Any new system for commanders must be able to evaluate the capabilities of the available sensors and platforms against an ideal template for that mission. Such a software tool would have to consider important variables, such as terrain, which could affect sensor detection.
Modeling and simulation also would take a key role. Apart from providing the software tools to plan out a mission, simulations could take advantage of tagged metadata, which allows commanders to use pertinent, real-world information for training.
A number of templates are available for modern military radar systems to define detection probabilities for different target types in different operating environments. It is possible to make such tools for all sensor types, including radar, electro-optical and infrared systems. The important part is that users will have to optimize the equipment to match the mission.
While there are many ways to better leverage existing technologies, DoD and intelligence community planners will have to look at how they can maximize all of their assets. By rethinking the traditional approach that emphasized collecting a maximum amount of data, commanders and agency heads can get more use out of what they have to help meet their missions.
Scott Goldstein, senior vice president and manager of QinetiQ North America’s National Systems Sector.