Connecting the Border

The Homeland Security Department is turning to industry for answers with its SBInet contract.

Naco, Ariz., is a small and dusty town, steps away from the Mexican border. It's also a hot spot for illegal immigrants crossing into the United States. Groups daily slip through the barbed wire fences that mark the frontier near Naco, crouching and running through the desert scrub into America.

In 1997, the Immigration and Naturalization Service began to set up a network of ground sensors and cameras to better patrol the northern and southern borders. INS chose Naco as one of the sites. If the Integrated Surveillance Intelligence System sensor detected motion, Border Patrol agents were to remotely aim the camera and assess the situation.

In summer 2004, auditors from the General Services Administration traveled to Naco to assess the acquisition. Investigators found equipment worth millions of dollars that never was hooked up, lying on the desert floor or bundled away in storage. Even when ISIS equipment functioned properly, a sample data review by the Homeland Security Department inspector general showed that more than 90 percent of the sensor alerts were false alarms. Passing animals or a train could set off alarms.

"The high-tech surveillance system guarding our border is more like a low-tech mess," the Arizona Republic grumbled in a June 2005 editorial. Between ISIS and America's Shield, which was to be the program's successor before DHS scrapped it last year, the federal government budgeted more than $450 million through fiscal 2005 on remote border surveillance efforts.

DHS is getting ready to start a new enforcement project. As part of the Secure Border Initiative, the latest effort to counter illegal immigration and secure America's porous perimeter, DHS in April issued a request for proposals for SBInet.

The department says the approach will be different this time. SBInet "is not about simply buying gizmos," Michael Jackson, DHS' deputy secretary, told an industry audience in Washington earlier this year. The performance-based, indefinite-delivery indefinite-quantity contract is about reshaping the way government polices the frontier. "We're asking you to come back and tell us how to do our business," Jackson said.

And the government wants industry to be quick about coming up with ideas. DHS expects to award a contract to a single prime integrator no later than Sept. 30. "There's really not one answer to the border issue," says Bruce Walker, director for homeland security at Northrop Grumman. In densely populated areas, agents have only minutes to respond to illegal crossings while in remote areas they could have hours. Patrolling the Great Lakes is different from the Rio Grande. Northrop, along with Lockheed Martin and Raytheon, is preparing to bid as a prime integrator. (DHS expects primes to team with other companies.)

Some needs are readily apparent. ISIS sensors and cameras are not linked to automatically respond together. Control centers in each of the Border Patrol's 20 sectors aren't connected.

"Clearly, the government is asking for improved situational awareness," says Jay Dragone, a Lockheed Martin vice president for homeland security programs.

Also, just monitoring the border probably isn't enough. "We need to detect the people before they cross, so we can be there to intercept them," says an industry official, who asked for anonymity. The number of people in an illegal crossing should control how many agents are sent to intercept them. And the time it takes border agents to process apprehended immigrants must be decreased, says Gene Blackwell, vice president of Raytheon's rapid initiatives group. Dealing with administrative tasks means an agent is "off the border-he's no longer able to control his assigned sector." The essence of SBInet is combining data from many evolving technology sources with business processes that can adjust to a constantly changing border.

When it considers industry proposals, DHS should avoid falling into the trap of hiring systems integrators to wire together a fixed set of components with proprietary technology, says Matt Walton, chairman of the Emergency Interoperability Consortium, a public-private effort to create public safety data-sharing standards.

New technology, unanticipated requirements and a need to share information with other homeland protection entities make adoption of open data standards imperative, Walton adds. Proprietary standards admittedly offer an easier way to ensure information security, he says, but they would be self-defeating. "We've got to make all these disparate systems out there better able to communicate," Walton says. Proprietary standards block information sharing; open standards encourage it. Open systems also make network components replaceable, which means that as innovation occurs, the network can adapt. DHS' quick turnaround time makes open standards even more crucial. "What it provides you is the ability to be wrong on the margin," he says.

Some observers worry whether DHS has the procurement chops to pull off SBInet. Understaffing, changes in leadership and a clash of cultures among the throng of previously separate procurement organizations have added up to delays and confusion, industry sources say. Plus, "Half of DHS is going to be working on hurricane response, so where do these other programs go?" the industry official asks. And in general, the government's ability to supervise large acquisitions hasn't always been great.

Others say DHS is taking steps to ensure the contract's success. The Secure Border Initiative program management office is working to make sure it gets procurement office resources. DHS officials have gone out of their way to tout the program, too. "Large parts of the border . . . are not going to be addressed by putting fencing or roads in; it's going to be addressed by getting what we call SBInet," DHS Secretary Michael Chertoff told the House Appropriations Homeland Security Subcommittee on Feb 15.

DHS officials acknowledge that SBInet cannot succeed by itself. The contract is just one leg of a three-pronged effort to control the border, the other two being more robust immigration enforcement in American workplaces and a temporary worker program. Especially without some way to incorporate immigrant labor into the American economy, the hard pull of better jobs here could prove stronger than any enforcement effort.