DHS ‘virtual fence’ and Army’s FCS test limits of networked sensors

Remote sensors that can accurately detect and track people and vehicles passing through a certain area have always been of interest to the military and are increasingly a focus for U.S. border surveillance.

They also are a factor in the Army’s network-centric vision for the battlefield as part of Future Combat Systems (FCS). But as recent efforts by the Homeland Security Department have demonstrated, turning remote-sensor technology into an effective sentry is no mean feat.

DHS’ Project 28, a proof-of-concept program, used a combination of mobile and static sensors to erect a virtual fence along a 28-mile stretch of the border between the United States and Mexico. It was meant to highlight new ways of detecting illegal crossborder movements as part of DHS’ SBInet.

After some delays to tackle technicaldeficiencies in the $20 million project, DHS Secretary Michael Chertoff announced the department’s acceptance of the Project 28 prototype from lead contractor Boeing in February. The technology worked, he said, but he added that it “wasn’t as good as it could be.”

The core of Project 28 is a network of nine mobile towers that carry a range of radar systems, cameras, and satellite and wireless communications. Unattended ground sensors (UGS) scattered along the border are meant to catch movement and alert the towers for a more detailed look.

The towers and UGS were accompanied by an array of other resources, such as border agent vehicles fitted with specialized communications; rapid-response vehicles; mobile command, control and communication units; and software to operate the system and provide the map-based Common Operating Picture (COP), a near real-time view of what was happening along the border.

None of this is necessarily new technology, said Jayson Ahern, deputy commissioner at Customs and Border Protection. However, he told a recent House panel looking into the lessons learned from Project 28 that the components are typically deployed singly, making them resource-intensive.

For example, he said, deployment and operation of truck-mounted mobile surveillance systems requires a border agent to drive the
truck, monitor activity on the radar and relay that information to dispatchers or other agents.

Likewise, he said, UGS and remote video surveillance systems were more useful if they could be linked together and more of their actions automated.

Project 28 was the first proof of concept of this integrated, linked approach [that] we believe in the long run will make our front-line personnel more efficient and effective by delivering an integrated package of sensor technology with an enhanced user interface,” he told lawmakers.

At an earlier briefing on border security, he said the system provided a way for border agents to detect movement and identify its source at the same time. Previously, agents could detect something using ground sensors and cameras but couldn’t tie the reports together.

The problems with Project 28 weren’t with the individual sensor and communications components, which are proven commercial technologies. One of the biggest problems was with the COP software.

The initial version of COP was based on a law enforcement dispatch system that proved unable to process and distribute the type of information collected by the cameras, radars and sensors, according to the Government Accountability Office. As a result, it said, remote operators in Tucson, Ariz., couldn’t lock the tower cameras onto their targets.

Other problems included the system taking too long to display radar information in command centers, and newly deployed radars were sometimes activated by rain or blowing leaves, making the whole system unusable.

In December, DHS awarded Boeing another $65 million contract to replace the COP software package with one based on military command-and-control software.

Project 28 is conceptually similar to the way the Army employs its newest generation of Remotely Monitored Battlefield Sensor System Version II (Rembass-II) UGS, which is most often used in combination with ground surveillance radar.

Rembass-II uses a range of seismic, acoustic, magnetic and infrared sensors to detect motion and then reports it to the operator of a laptop PC. The radar is then used to get a broader view of any targets that are moving in the vicinity of the UGS. However, Rembass-II UGS will be replaced during the next few years by yet another generation of systems designed to fit into the more network-centric battlefield envisioned under FCS.

These UGS, which were recently delivered to the Army Evaluation Task Force at Fort Bliss, Texas, for evaluation, are designed to act as nodes in a sensor network that can not only detect the motion of people or vehicles but also closely track them over a wide area. “What differentiates these FCS UGS from current and past UGS is the networking of the sensors,” said Mike Beltier, director of intelligence, surveillance and reconnaissance at Boeing, the FCS lead systems integrator. “The reports from those sensors are fused in a network fashion and then progressively fused with other sensor data.”

“It’s the networking that provides the force multiplier” for these systems, he said.

The FCS UGS also apply commercial technology as much as they can, he said. The imaging sensors are based on the megapixel advances of the cell-phone industry.As Boeing has found in its DHS activities, however, bringing together the data from these sensors is the hard part. “Data fusion is certainly one of the challenges,” Beltier said. “You have to strike a balance between where you are doing it, communication latencies, the frequency with which you update the information and the bandwidth you have available.”

The algorithms used in these FCS UGS, which were developed by Textron Systems, selectively filter data associated with animals, blowing leaves and other effects behind so many of the false-identification problems that have bedeviled older sensors. The FCS UGS can classify the people and vehicles they detect in addition to tracking them. Besides the more advanced algorithms, advances in digital signal processing electronics have helped make that possible, said John McQuiddy, president of McQ,, a developer of remote surveillance, security and environmental monitoring technologies. It provides UGS for the Army and DHS.

DHS still uses ground sensors based on Vietnam-era technology that can only report that something has moved through an area, he said.
The new generation UGS “are so dramatically different from what has gone before, and they can significantly alter the way you control
the border,” he said. There could be tremendous overlap in the use of FCS UGS in the military and DHS, Beltier said, because they have similar requirements.

The only real difference is in the details of their implementation and what sensor technologies are used. CBP has about 7,500 UGS deployed along U.S. borders, Ahern said, with another 2,500 on order for the current fiscal year. It’s unclear what the eventual mix of sensor technologies used in SBInet will be. At the initial award of the SBInet contract in 2006, Chertoff said that the network was intended to be a flexible tool that would have a different look along different stretches of the border. The only common feature would be how data from all sensors would be integrated into COP.

Ahern reiterated that in his February congressional testimony. Referring specifically to Project 28, he said a version of the solution
could be used where it makes sense for selected portions of the border. “Different segments of the border require different approaches, and a P-28-like system would neither be cost-effective nor necessary everywhere,” he said.

For many locations, he said, existing tools will be more than sufficient to cover the nature and level of the threat, and for others, even more
advanced tools will be needed.