What are you looking at?

Technology and integration choices dominate Secure Border competition

How the SBI bidders stack up

Ericsson Inc.

Stockholm, Sweden

Experience: Norway-Russia border surveillance

Team: Computer Sciences Corp., L-3 Communications Inc.; Fluor Daniels, Camber Corp., MTC Corp., Sarnoff Corp., University of Texas

Comments: "The big focus we have learned is to be intensively focused on how the border patrol's working process is, how they actually use it, and making sure it's integrated. It's a focus on the people and the process, not the sensors," said Douglas Smith, director of government sales.

Lockheed Martin Corp.

Bethesda, Md.

Experience: In August 2005, won a $212 million contract to create an integrated video surveillance system for the New York Metropolitan Transit Authority

Team: Not announced

Comments: "What is most technically challenging in Secure Border is to balance the amount of investment you need to get to the operational control of the borders," said Jay Dragone, vice president of Integrated Border Security Systems. "You have to manage the resources and how to best address the threat."

Northrop Grumman Corp.

Los Angeles

Experience: In January 2005, won a $12 million Pentagon contract in to develop a system that uses low-flying, unmanned aerial vehicles to deliver real-time surveillance data to individual warfighters on the battlefield. Named finalist in January, with Motorola Corp., for contract to build a high-speed wireless network in New York City

Team: Not announced

Comments: Collecting surveillance data and delivering it where it is needed in the field draws on Northrop Grumman's expertise in geospatial and tactical intelligence systems and wireless capabilities, said Sidney Fuchs, president of the civilian agencies group. Regarding selection as a finalist for the wireless initiative in New York, he said: "It's pretty significant that Northrop Grumman, with a history in shipyards and aircraft, beat out companies whose business model is based on wireless communications."

Raytheon Co.

Waltham, Mass.

Experience: On March 22, won a two-year, $102 million contract to design and deploy a perimeter intrusion detection system for four airports operated by the Port Authority of New York and New Jersey
Team: IBM Corp., BAE Systems Inc., Bechtel National Inc.

Comments: Raytheon has done many field evaluations for border security solutions in the last 18 months that it has tested near El Paso, Texas, said Gene Blackwell, vice president of rapid initiatives. "Our approach is to go to the border, experience it, and find out what works and what doesn't work," Blackwell said. "We try to walk in the shoes of the agents."

Video surveillance grows up

Video surveillance expert Joe Freeman has watched in recent years as cameras joined the IT world, as formerly analog footage became digitized and cameras were integrated into IT networks.

As 21st century as such digital imagery may be, it's not technology that may be the biggest issue for the Secure Border Initiative-Net, the Homeland Security Department's comprehensive border surveillance system, said Freeman, president of consulting and market research firm J.P. Freeman Co. Inc., Newtown, Conn.

"Probably one of the main considerations is the environment ? intense heat and hurricanes to the south, and intense cold up north," Freeman said. "You want these cameras to be working 24/7/365, so they have to be protected."

If the cameras are to handle facial recognition, they must be high resolution. But that presents difficulties in getting adequate bandwidth and processing power, especially for wireless networks, he said. "High resolution is a bandwidth hog," he said.

The project's main difficulties stem from the amount of integration required to connect the cameras with data and voice networks for the border patrol and other agencies, he said.

In routine applications such as building surveillance, video surveillance data often is discarded after 30 days. It's likely to be held much longer in a border surveillance project, as agents pursue intelligence leads, Freeman said.

"Suppose something happened 42 days ago? You'll need more storage," he said.

Many commercially available video systems, such as those used in buildings, airports, stadiums and shopping malls, have proprietary software and links to networks, but a federal surveillance system for the border most likely will need an open architecture to integrate and share information with agencies, Freeman said.

But there may be obstacles in buying an off-the-shelf system. "The camera manufacturers don't necessarily want to have open standards; they are resisting this," Freeman said. "They want everyone to keep using their proprietary equipment."?Alice Lipowicz

Chris Josephs, director of homeland security of the global government solutions group for Cisco Systems Inc., says "Surveillance is easy where there is infrastructure. The hard part is the other 90 percent."

Rick Steele

Beyond cities and highways, miles upon miles into the desert, the U.S.-Mexico border stretches to the horizon over sun-baked land.

"In the desert, there is no power, no towers, no anything," said Chris Josephs, director of homeland security of the global government solutions group for Cisco Systems Inc., Herndon, Va.

In this harsh environment, several of the largest IT systems integrators in the United States are taking on one of the biggest federal technology projects of the decade: creating Secure Border Initiative-Net, an electronic surveillance system that will cover the U.S.-Mexico border, 2,000 miles long. The other portion will span the 4,000-mile-long northern border with Canada.

IT experts consider Secure Border daunting, mainly because of the scope and complexity of the solutions and integration needed to achieve the project's goal: Give border patrol agents operational control of U.S. borders.

Establishing an infrastructure of power and communications to support a large-scale surveillance system in such vast and remote areas unquestionably presents challenges.

"Surveillance is easy where there is infrastructure. The hard part is the other 90 percent," Josephs said.

The agency and contractor will need to make many choices, not only on sensors and video, but also concerning infrastructure support elements such as high or low bandwidth, wired or wireless networks, power lines or alternative power and on-site or off-site data storage.

Prospective prime contractors must aim for infrastructure that is resilient, reliable, scalable, flexible, secure and interoperable with other systems, IT industry executives said.

"The biggest challenge is the infrastructure," said Glenn McGonnigle, chief executive of VistaScape Security Systems Corp. of Atlanta, makers of automated wide-area surveillance systems. "They will need to move the data over wide-open spaces and mountainous regions."

Years in the making

Homeland Security Secretary Michael Chertoff announced Secure Border in November, but the program's debut follows several years of anticipation for a predecessor border surveillance program, America's Shield Initiative, which has been folded into Secure Border. That merged program was to expand upon the border patrol's $430 million video surveillance system, construction of which started in 1998.

Secure Border is expected to be one of the department's largest contracts and be worth $2 billion. Ericsson Inc., Lockheed Martin Corp., Northrop Grumman Corp., and Raytheon Co. are each putting together a team to compete for the prime contract.

A request for proposals is anticipated within days as of press time, and an award will be made in September.

A chief feature of Secure Border will be sensor and surveillance technology, possibly including advanced video cameras with high-resolution images, infrared viewing, night vision, low-light capabilities and "intelligent" capabilities to detect unusual events.

Detectors for heat, motion and biological and chemical agents likely will be included, and ground radar may play a role. Facial recognition software also could come into play; sensors would be mounted on stationary and moving platforms, such as unmanned aerial vehicles.

Power lines, networks and systems will support the sensors and cameras, and that's where some of the prospective solutions could differ.

One key decision for each team competing will be its blend of wired and wireless infrastructure, Josephs and others said.

Wireless networks are fast and flexible, but few offer adequate bandwidth to support high-resolution video images and streaming video. And they don't work well in high humidity or in heavily forested and mountainous areas, where radio signals can be blocked.

Although wired networks, laid with fiber-optic cables underground, are an optimal technical solution and can handle large amounts of video and data, they also are expensive and time-consuming to install.

"I'm not sure you need to run wires all around the borders of the United States," said Gene Blackwell, vice president of Raytheon's rapid initiative group. "It will be a mix of wireless and wired."

"It could take two decades to do towers and fiber along the borders," Josephs said.

Alternatively, wireless networks can effectively transmit video content if it is compressed, Josephs said. "Five years ago, video was very bandwidth-intensive, but now we can minimize its bandwidth consumption," he said.

Another choice affecting bandwidth and networking will be whether to support video analytics in cameras or in a network. Each approach has its promoters.

By placing the analytical capabilities on a processing chip in a camera, "it will save money, and make the cameras more hardened" and better able to operate independently, said Joe Freeman, president of J.P. Freeman Co. Inc., a consulting and market research firm. That could be especially important, he said, in harsh environments such as the hot, dry desert and the extremely cold northern border regions, where cameras need to be well-protected from the elements, possibly making them more difficult to access.

A remote wireless camera that operates on its own may be more secure from sabotage than a wired, accessible camera attached to a tower, Freeman said.

"What's to stop someone from climbing up and cutting the access wire?" he asked.

On the plus side

Placing those capabilities in the network instead of in the camera also has advantages, said McGonnigle of VistaScape, which produces video analytic software.

"If the threat profile changes, you can change it on the network instead of changing it on every camera," he said.

For example, if the analytics are wired to sound an alarm to indicate the movement of vehicles or people on or near the ground, the alarm settings and possibly camera positions may need to be adjusted if border crossers use gliders or parachutes to circumvent the cameras and enter the country.

Another unresolved aspect of the system is whether to include facial recognition capabilities.

"With high-resolution cameras, you get more information, but it slows down the processing," Freeman said. "But if you're going to do facial recognition, you need high resolution."

Storage requirements for the surveillance system would be huge. The amount of storage is indicated by operational needs, which for forensic intelligence may lean toward indefinite. It also depends on the amount of compression and analytics. Intelligent video with analytics, for example, may begin shooting high-resolution footage for storage only after sensors detect an unusual event.

The surveillance videos are designed for 24/7 operation. A standard digital video compression format generates about 13G of data per hour, and even if that is reduced with additional compression or analytics, storage requirements are massive.

"The amount of data is tremendous," Clara Conti, chief executive of Ipix Inc., a high-resolution video camera maker in Reston, Va. "Fortunately, the government and military are very accustomed to [large] storage requirements."

About that spectrum

Managing radio spectrum will be another infrastructure challenge. Will the system use public safety, military or commercial spectrum? Must it be spectrum-compatible with the $10 billion Integrated Wireless Network being built by the Homeland Security, Justice and Treasury departments?

IT executives involved in Secure Border said the answers are unclear, though presumably Secure Border will need to coordinate with IWN.

Executives from Raytheon, Lockheed Martin and Northrop Grumman said there will be difficulties installing and integrating the Secure Border technologies and infrastructures, but there are no significant risks to reaching success.

"It's a classic systems-engineering problem," said Jay Dragone, vice president of integrated border security solutions for Lockheed Martin. "I don't see any really strong technical challenges."

"We'd like to give representatives in the field the technology they need to get a common operational picture," said Sidney Fuchs, president of the civilian agencies group for Northrop Grumman IT.

The integrators aren't starting completely from scratch. The controversial Integrated Surveillance Intelligence System is still in operation, albeit with many imperfections and gaps. A host of federal audits have documented problems, including cameras purchased but not installed and lack of competitive bidding.

The surveillance system will need to blend old and new in mind-bogglingly harsh and vast environments.

"This is almost like the Internet was in the 1980s," Cisco's Josephs said.

As are some of the questions.

"How will it look?" he asked. "Who will service it?"

Staff Writer Alice Lipowicz can be reached at alipowicz@postnewsweektech.com.

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above.

What is your e-mail address?

My e-mail address is:

Do you have a password?

Forgot your password? Click here

Washington Technology Daily

Sign up for our newsletter.

Terms and Privacy Policy consent

I agree to this site's Privacy Policy.


contracts DB