A blue-ribbon panel recently concluded that a limited national missile defense system is technically feasible. That's good news in light of a series of intelligence reports that says the threat of a missile attack by a rogue nation such as North Korea, Iran or Iraq is real.

But building a system that can detect, find, track and "kill" potentially dozens of incoming warheads in space is the technological equivalent of guiding the head of a pin directly into another from thousands of miles away.

And not everyone agrees that such a system is feasible. Some believe a limited system can work. Others, however, think the entire proposition is a pipe dream and a leftover from the Cold War and the Reagan administration.

But the nation may soon have a better idea about exactly what is possible. On Friday the Pentagon will conduct the first full-scale trial of the national missile defense (NMD) system.

Hitting the target, say military officials, will be a secondary consideration. Ensuring that all of the command and control systems that detect, find and track the incoming missiles can do their job is more important and remains the toughest challenge, they say. The test will provide the clearest picture yet of whether such a system is possible using existing technology.

"The key to this whole system working is the battle management system that integrates all of the multiple sensors and the discrimination capability," said Jacques Gansler, undersecretary of Defense for acquisition, technology and logistics, at a Pentagon briefing June 20.

"It is truly a system of systems...each of which must communicate with each other," said Gansler, speaking before the House Armed Services Committee. "The technical challenges...are truly challenging."

Under a law passed by Congress and approved by President Clinton, the Pentagon intends to deploy an NMD system by 2005 designed to counter dozens of incoming missiles. In fact, the CIA has confirmed that North Korea currently has the ability to hit the United States with a long-range missile. "It is a major growing threat to the U.S.," said Rep. Curt Weldon (R-Pa.), chairman of the House Armed Services Committee's Military Research and Development Subcommittee, in a recent speech.

The system, estimated to cost a minimum of $36.2 billion, will use ground-based interceptor missiles armed with "kill vehicles" designed to collide with incoming targets in their trajectory outside the Earth's atmosphere. The system also will use ground-based radars and satellite-based infrared sensors to home in on their targets.

The Pentagon plans to deploy the system after at least four design reviews, with improved capabilities being added and tested in each phase. The president is scheduled to decide this fall whether to move ahead with full-scale deployment of the NMD system.

Even if successful, however, the technologies will require further development. And integrating the array of information systems will remain the toughest challenge.

"[The information technology] is a lot more robust than we would normally expect at this time because we were working on it before we even had this program, but we have a lot more work to do on it," said Air Force Lt. Gen. Ronald Kadish, director of the Ballistic Missile Defense Organization (BMDO). He added that although the technology is "pretty good the way it is," integrating it and ensuring that it all works properly together remains a hurdle.

"That's why we've got this fundamental test subject. We're finally at a point in time when all of these systems are mature enough to hook together in a test, and that's what we're doing now," Kadish said.

The project also has to overcome tests in the political and academic arenas, where critics of the project are not hard to find.

"The problem becomes more challenging as the number of objects increases," said John Pike, a defense analyst with the Federation of American Scientists. "There is the risk that the [Battle Management Command, Control, Communications and Intelligence] system could lose track of an object that it thought it had discriminated, and then another sensor acquires it and has to discriminate all over again."

The sheer complexity of the NMD system also means that system integration will be a big challenge, said one expert, who believes the Pentagon already has blown the opportunity to build NMD information systems designed from the beginning to operate together. Instead, they will have to lash individual systems together after the fact.

"I believe one of the critical features is integration and interoperability," said Stanley Orman, president of Orman Associates Inc., a Rockville, Md., consulting company. "Unfortunately, most designers still seem to subcontract the design of subsystems - like the missile or the radars or the sensors - to individual companies and really don't do it from a top-down approach from square one. There isn't an overall specification for the total system before they build the subsystems."

The information systems that are scheduled for testing this week include the space-based early-warning sensor; the ground-based early-warning, tracking and discrimination radars; the Battle Management, Command, Control, Communications and Intelligence system; the in-flight communication system; the interceptor missile; and the kill vehicle. The kill vehicle rides atop booster rockets before being fired at the incoming missiles like a bullet, destroying them through sheer mass and speed.

Orman, the former director of the United Kingdom Strategic Defense Initiative Participation Office, the British equivalent of BMDO, blasted BMDO for abandoning its responsibility as prime integrator for the information systems.

"Under the original concept, BMDO would be the system integrator. Now what happened is that they really acted as a pass-through for the money and haven't done the main task, which is the integration and interoperability of the systems," Orman said.

And without integration of the information systems, the NMD system will not work, he argued. "They are now facing a very critical phase because when you bring these subsystems together, they never interface the way you would like them to, and you have to tweak the interfaces," Orman said.

Given the complexity of the system, any number of things might go wrong in the test, Kadish told the House Armed Services Committee June 22.

Although this month's test will be considered a failure on a technical basis if the interceptor does not knock the warhead out of the sky, the Pentagon tends to find successes where others find failure and therefore will likely move forward with the system regardless of the outcome. "The question is whether it is technically feasible," Gansler said. "And depending upon the types of failures, we may still be able to say it's technically feasible."

The upcoming test is the third in a series. The first was in October to test the ability of the interceptor to discriminate between the target and decoy; the second, in January, was an intercept test. The first was considered a success; the second was not.

In the final five seconds of the second test, the cooling system of the infrared sensor in the kill vehicle failed, and the missile missed the target. Still, Pentagon officials point out that up until the critical last few seconds, everything worked as planned.

Orman said the Pentagon's tendency to tout success where others see failure is sometimes valid. "To the public, success is "smash, you hit it.' I've designed and done some of these sorts of tests, and they are extraordinarily complicated. It could get to within the very final closing stages and then a thruster or something else could go wrong, and you have achieved 98 percent of what you want to achieve, but to the public it's a failure."

The NMD system will be extraordinarily complicated, in part to deal with possible countermeasures that are simple and effective enough to make the system unworkable, according to the Union of Concerned Scientists, an independent nonprofit alliance of scientists. The group released a report in April and testified before Congress in early June that the NMD system will not work.

"The so-called national missile defense could be defeated by relatively simple countermeasures designed to overwhelm or fool the system," said Joseph Lach, a physicist at the Fermi National Accelerator Laboratory in Chicago.

Possible countermeasures to the NMD system, according to several experts, include:

* Arming missiles with such a large number of submunitions that interceptors would be overwhelmed.

* Mixing genuine warheads and decoys that mimic the flight pattern, infrared signatures or other characteristics of warheads.

* Having the warheads mimic characteristics of decoys.

* Launching a missile from a nearby location, such as Canada, Mexico or an offshore ship, which would deny the system the response time it needs.

* Using a directed energy or electromagnetic pulse weapon to confuse the NMD's electronic systems.

Although he would not provide details in a public hearing, Kadish told lawmakers he believes the Pentagon has prepared for the electromagnetic pulse threat.

But Orman, who helped Great Britain develop a countermeasure system against Russian intercontinental ballistic missiles, said the Union of Concerned Scientists missed the mark.

"It's a very complicated situation with countermeasures, made grossly oversimplified with the Union of Concerned Scientists report. They have said anyone who can design a missile can design countermeasures. That's a gross, gross oversimplification," he said. "There are always trade-offs with countermeasures. The more complex the [NMD] system is, the more complex the countermeasures have to be."

What's more, the more countermeasures you add, the less room you have for actual warheads, Orman argued. "It's easy to design decoys that mimic the infrared signature or the radar reflection or the aerodynamic behavior or the microwave reflection of a warhead, but when you try to put them all together into one system, it is no longer simple."

The Union of Concerned Scientists said its analysis was anything but simple. "Anyone who can build a long-range missile and weapons of mass destruction - which is not easy - can build effective countermeasures. Countermeasures are simpler than missiles or weapons of mass destruction," said Tom Collina, director of the organization's arms control and international security program.

The group evaluated three relatively simple countermeasures and concluded that the midcourse NMD system, as it is now planned, would not defeat them. Their conclusion: "The United States should reconsider its options for countering the threats posed by long-range ballistic missiles and shelve the current NMD plans as unworkable and counterproductive."

"We're not against a missile defense system," Collina said. "We're against a national missile defense system that won't work."

The NMD also has come under fire from the General Accounting Office, the investigative arm of Congress, which released a May report detailing the plan's risks. The report said that the Pentagon has underestimated the technical challenge of hitting an incoming target with an interceptor and that only three of 19 planned intercepts would be performed before the critical White House decision this fall.

GAO also said that none of the tests would include "the higher acceleration and vibration loads of the much faster actual system booster," and that the aggressive schedule is vulnerable to program delays and costs of up to $124 million each month the program is delayed.

The Federation of American Scientists' Pike, a longtime NMD watcher, said tests like the one scheduled for July 7 add a little more fidelity to the system architecture concept each time they are conducted. However, many questions remain unanswered, he said. "I remain unconvinced that really high-fidelity simulation is in the cards, and totally unconvinced that anyone would ever bet the country on this contraption," Pike said.