After nearly 15 years of development work, more than eight years of delays and billions of dollars in cost overruns, the first of the U.S. Defense Department's new early missile warning satellites is finally poised for launch.

The Space-Based Infrared System (Sbirs) geosynchronous (GEO) satellites will provide a new generation of IR sensors designed to detect ballistic missile launches—including “dim,” short-range boosts—faster than today's Defense Support Program (DSP) constellation.

A launch success will be a step to help move forward from more than a decade of dismal performance in space programs by the U.S. Air Force and Lockheed Martin.

A Sbirs failure would be a stunning turn for the worse for military space programs, which have struggled through quality-control problems, management mishaps and multibillion-dollar overruns. In short, the Air Force's credibility in delivering precious spaceborne capabilities for the nation is on the line.

Less than a month remains for an 11th-hour snag to arise for Sbirs GEO-1, which is poised to lift off from Cape Canaveral. The flight date is set for May 5 on a United Launch Alliance Atlas V, but Col. Roger Teague, the U.S. Air Force's Sbirs program manager, says May 4 is an option depending on when the space shuttle Endeavour returns from its latest mission.

The satellite must also undergo the launch, outgas, shed its protective sensor cover, and point and focus its highly sophisticated infrared payload before military commanders will be at ease.

Although Air Force officials say the average cost of a Sbirs satellite is $1.3 billion (without factoring in the large development price), GEO-1 will be by far the most expensive of the series owing to all of the delays incurred while building the first one. So it's not just the Air Force's credibility on the line; a massive amount of funding is at stake.

Total program cost was estimated to be $15.1 billion for six units; this includes a mix of GEO satellites and separate sensors placed on classified host satellites in highly elliptical orbit (HEO). Air Force officials refused to provide a per-unit cost for the GEO spacecraft, including development fees; the average is based on a 2010 report to Congress is $2.5 billion. However, the cost growth may not be over. Government auditors found in a March report that the Sbirs contract estimated cost at completion has grown about $600 million, more the twice the overage projected last year.

Aside from the nearly four-fold cost increase over the program's life and substantial delay, the failure of the last DSP satellite, made by Northrop Grumman (formerly TRW), to operate properly after its 2007 launch underscores the need for the Air Force to make sure the forthcoming Sbirs flight is successful so that a gap in missile warning coverage can be avoided.

For years, successive Strategic Command chiefs have publicly complained that the missile warning constellation is too fragile to withstand an in-orbit failure or launch problem. The primary concern is not only for the immediate future, but in keeping DSPs functioning long enough for Sbirs GEO-3 to be lofted and certified for use. Lockheed Martin is working on that spacecraft, but its production suffered a gap owing to a lack of government funding.

Sbirs GEO-1 arrived at Cape Canaveral on March 3 for processing; the satellite was slated to be fueled on April 8.

Once in orbit, the satellite will bring dramatically new missile detection and intelligence capabilities into the U.S. arsenal even though the nation's plans for a missile defense system have changed substantially since Sbirs was conceived 20 years ago. Army Lt. Gen. Patrick O'Reilly, Missile Defense Agency (MDA) director, says his primary focus now is to defend against a “raid”—10 or more short- and medium-range missiles launched nearly simultaneously against U.S. interests. “What we see with all the intel trends are that large raid sizes are something that a prudent defense architecture needs to not only understand but also embrace,” O'Reilly said last month.

The primary focus today is in building robust regional defenses, specifically through the Obama administration's Phased Adaptive Approach aimed at protecting much of Europe from an Iranian ballistic missile attack.

Although regional defenses were considered a priority in the 1990s, much of the MDA's efforts were directed toward fielding the Ground-Based Midcourse Defense (GMD) System, a global series of ground- and space-based sensors with interceptors in Alaska and California designed to protect North America from an ICBM attack.

“Because of the improvement in accuracy and the improvements in the [sensor] revisit rates, . . . Sbirs will help us in being able to provide information down to short- and medium-range threats for the missile defense architecture,” says Air Force Lt. Gen. (ret.) Henry “Trey” Obering, a senior vice president at Booz Allen Hamilton who retired as MDA director in 2009. (The MDA declined to provide interviews because of ongoing budget hearings on Capitol Hill.)

Even though the focus has shifted toward shorter-range missiles—which burn for less time in flight and emit dimmer IR signatures—missile defense experts say Sbirs is designed to detect and respond to those threats.

GEO-1 and its successors will carry two IR payloads, one scanner and one starer. Both can detect threats in three different spectral bands—shortwave and mid-wave, and one that is only described as a “wider, more open shortwave band” able to “see through to the ground,” according to Jeff Smith, Lockheed Martin Sbirs program manager. “It is opened up more, which means more clutter and more targets at the same time. That is why you can see deeper into the atmosphere,” he says, noting that the performance is secret.

DSP satellites, now in orbit, rely on a spinning mechanism to search for ballistic missile launches. The Sbirs GEO scanner is a step up. It can scan in a box fashion, up one side of its field of view and then down the other, or operate in a “windshield wiper” pattern, Smith adds. With both methods, the revisit rate with Sbirs is improved over DSP, officials say.

The staring sensor, however, is entirely new with Sbirs. It can focus on an area separate from the scanner; the staring sensor does not rely on a scanning mechanism to refresh its IR returns. Typically, the scanner will be the first to catch a launch unless the starer is directed in the right spot, Teague says.

Two Sbirs scanners are already operating; they are hosted on classified satellites in highly elliptical orbit. HEO-1 and -2 are designed to provide coverage of extreme northern regions and linger over the North Pole.

Though not originally designed to focus on the lower latitudes, algorithms are being crafted to extend the reach of the HEO scanning sensors, says Ray Yelle, Air Force Space Command's missile defense and missile warning lead. The algorithms could be ready in as early as a year. This is one of several side projects under way to incrementally boost Sbirs capability as the system is fielded. “It is helping us to deliver capability sooner. . . . In the end, it is about being able to provide that tactical level commander the capability he needs much sooner,” Teague says. “It is hard to make Sbirs—a satellite—real to the warfighter.” GEO-1 was originally expected to launch by September 2004.

In another project, the Air Force is trying to relay an “IR view of the battlefield literally within seconds” to theater commanders. This helps with other Sbirs missions, including battlespace awareness. Teague says he views Sbirs as one of many assets, including aircraft, that provide persistent IR data, and he hopes to eventually find a way to pump all of that information into a cohesive picture.

In both missile warning and general IR data, Teague emphasizes that Sbirs will supply products to users much more quickly than DSP. “The lines start blurring between technical intelligence—which is largely offline and done after the fact—and in real time. The real difference is the time domain . . . being able get all of the threat information into a decision-maker's hands who is deployed—that is power,” Teague says. Air Force officials plan to begin providing theater commanders with IR data as soon as possible. Though the sensor will not be certified, it can be used as an adjunct for some missions. During the 1991 Persian Gulf war, DSP was employed to detect launch points of short-range missiles as they flew from their mobile launchers in Iraq; this information cued Air Force fighters to engage the launchers before they could relocate. In such a scenario, Yelle says, Sbirs would be able to halve the ellipse—the area of ground—in which to look for a launcher.

Both HEO-1 and -2 are certified to relay Integrated Tactical Warning and Attack Assessment messages, meaning they can provide attack warnings to the ballistic missile defense architecture. However, commanders waived some of the operational requirements for HEO-1 and -2 owing to unresolved electromagnetic interference issues with the host satellite. Though design changes are in place for later HEOs, these first two sensors “did not meet all of the program's specifications,” according to government auditors.

Yelle says he expects GEO-1 to be declared operational within 15 months of launch; a first-of-a-kind satellite, it must undergo extensive sensor calibration work and increasingly complex IR scene tests prior to certification.

GEO-1 will be parked initially over North America to conduct tests; U.S. Strategic Command (Stratcom) has not yet announced where its operational slot will be.

The ultimate plan is to have four GEOs with overlapping sensor coverage areas in orbit; they would be augmented by two HEOs. Lockheed Martin is already crafting HEO-3 and -4 for potential launch in October 2014 and August 2016, respectively.

Based on the MDA operational concept, Sbirs will likely be the first to detect a launch. The data would then be relayed simultaneously to ground terminals around the globe. Operators at the Sbirs mission control station would generate a launch report containing the missile type, launch point, precise time and azimuth as well as a predicted impact point. Smith notes that data from multiple satellites provide a more accurate, “stereo” view of the threat than a single spacecraft. Stratcom will likely place the satellites in orbits to optimize this capability.

Those data would then be released to the missile defense battle management system. At this point, the appropriate missile defense sensors, probably radars—the Sea-Based X-Band, regional early warning radars (such as those in California, Greenland or the U.K.), Aegis SPY-1 or Terminal High-Altitude Area Defense (Thaad) TYP-2—are given a search area to acquire the target and develop a firing solution. Launch of interceptors would follow shortly thereafter. Aegis ship-based SM-3 Block IAs, Ground-Based Interceptors in the U.S. and Patriot systems around the globe are on alert, although the interceptor magazine is not as robust Pentagon officials would like.

When Sbirs was originally conceived, the Pentagon envisioned two spaceborne constellations to both detect and track ballistic missiles in flight. While Sbirs—with its short- and medium-wave detectors—is designed to follow the hot motor plume, the Sbirs Low constellation (now called the Space Tracking and Surveillance System, or STSS) flies both an acquisition and a tracking sensor payload. Two STSS concept demonstration satellites are flying now and have proven the elusive “birth-to-death” missile tracking capability, says O'Reilly. Though proven in concept, MDA must still plan for a follow-on to STSS, called the Persistent Tracking and Surveillance System (PTSS), to serve operationally; and PTSS will likely have to connect tightly with Sbirs to provide timely data on missile attacks.

Collectively, STSS's payloads are capable of monitoring activity in the short-, mid- and long-wave bands for this mission. Yelle says PTSS could rely on Sbirs for the acquisition mission and feature only one tracking sensor. This could reduce the complexity of the design and, possibly, cost.

Teague plans to brief Pentagon acquisition chief Ashton Carter this spring on a strategy for acquiring Sbirs GEO-5 and -6. Air Force officials are examining whether to buy them through the Evolutionary Acquisition for Space Efficiency concept, which calls for a block buy of two satellites at once to reduce cost (see p. 53).

Although Sbirs is only now readying for launch, IR detector technology has advanced significantly since the program was conceived. In what may have been the darkest days for Sbirs in 2005, the Pentagon established an alternate program to develop staring focal plane arrays suitable for the missile warning mission. These systems would not require the complex moving parts of a scanner. The Pentagon plans to fly one of the sensors, which is built by SAIC, on an SES American satellite this summer. If this Commercially Hosted IR Payload program proves successful through a series of demonstrations, Pentagon officials could eventually abandon the Sbirs design.

While Lockheed Martin probably has not lost money on this cost-plus contract, the company has not gained as much as it had hoped. Teague says the award fees are “private.” According to Smith, “The award fees have been appropriate. We have seen notable improvement in some areas.”

Many of the problems with Sbirs originated at the inception of the contract in 1996. Both the company and the Air Force over-promised and under-delivered. The government also belatedly added 16 requirements onto the program.

The original Sbirs award was a Total System Performance Responsibility (TSPR) deal, popular in the late 1990s. This called for the contractor to perform many of the management functions typically handled by the government. Since then, Sbirs has been used as an example of why the government will not use TSPR in the future.

For example, HEO-1 experienced major problems because of unanticipated electromagnetic interference with its host payload; the result was a massive overrun to pay for a so-called standing army while awaiting a fix. Also, designers did not properly foresee glare issues on the sensor payload; that led to the hurried design of the deployable light shade.

Finally, the most recent problem called for a virtual redesign of the satellite's onboard computer architecture. A classified Lockheed Martin satellite that went dark 7 sec. after reaching orbit in 2007 used similar software, and the safe-hold mode (used for basic health and communications in the event of a fault) failed. Then Gary Payton, Air Force deputy undersecretary for space, said this satellite became a useless “ice cube.”

Final fixes being worked on GEO-1 were still under way last year, demonstrating the breadth of the work. It was “really a new development effort to start on the flight software,” Teague says. “The program was forced to start from scratch.” Senior service officials refused to fly the satellite without repairing the architectural problem. Highly precise timing for communications between onboard computers is needed to ensure the safe-hold software can engage. This timing was off for the classified satellite, Payton said.

Lockheed Martin officials recently “exonerated” GEO-1 from potentially having the same foreign-object debris problem that crippled the liquid apogee engine (LAE) for an Air Force communications satellite lofted last year. Both are built on Lockheed Martin's A2100 bus. Even though the Advanced Extremely High Frequency satellite will reach orbit months later than expected, officials were lucky they could rely on other onboard propulsion systems, specifically Hall Current Thrusters, and a bipropellant design, for orbit-raising to GEO, says Kevin Bilger, vice president and general manager of Lockheed Martin's Global Communications Systems unit. The Sbirs design will not feature the bipropellant system; so a failure in the LAE could be catastrophic.

Sbirs GEO Facts
Bus: Lockheed Martin A2100
Payload: Northrop Grumman
Sensors: scanning and staring, short- and mid-wave IR
Satellite weight: 10,000 lb. predicted at launch
Payload weight: 1,000 lb.
Per unit price for GEO: $1.3 billion
Total program cost: $15.1 billion
Source: AW&ST research