|
|||||
|
|||||
|
|
|
|
|
|
|
|||||
 |
 |
    |
|
|
|||
 |
|
||
|
|||
|
|
|
|
||||||||
|
United States: Missile Defense Software Might Be Flawed, GAO Says By Kerry Boyd According to the report from the General Accounting Office, Boeing subcontractor TRW developed software that would allow the missile interceptor kill vehicles to analyze data from Boeing sensors to distinguish between enemy warheads and decoys during flight. In 1997, the National Missile Defense Joint Program Office tested Boeing’s kill vehicle sensor. Boeing and TRW reported that the test had been mostly successful, although the sensor had often detected false targets. In a later report in 1997, the contractors said that the sensor had several problems, including a low probability of detection and inconsistent calibration. The companies reported two more problems in 1998. According to the GAO, the companies had reported the results and equipment limitations after the test, but they used qualitative terms such as “excellent,” which could have created confusion. The U.S. Investigates Claims of Fraud In 1996, TRW engineer Nira Schwartz reported to TRW that certain technology the company planned to add to its kill vehicle software was unable to properly distinguish between warheads and decoys. TRW fired Schwartz, who then sued TRW and said the company falsely claimed the technology met the Defense Department’s technical requirements. The Justice Department then hired Nichols Research Corp. to evaluate TRW’s technology and reports, but concerns arose over the corporation’s objectivity, so the department also hired Phase One Engineering Team. The second group reported that TRW’s software had weaknesses but was well designed. The team said the software would perform successfully in future tests as long as no unexpected factors occurred during tests. Simple Tests With Few Decoys The GAO, however, said that the team did not definitively prove that TRW’s software successfully distinguishes between decoys and warheads, because the team did not process the data from the 1997 flight test or develop its own data. In an independent evaluation, Massachusetts Institute of Technology professor Theodore Postol said that the contractors tampered with the flight test data to hide the inadequacies of the sensor. The Phase One Engineering Team and Nichols Research Corporation said that “TRW’s software used prior knowledge of warhead and decoy differences, to the maximum extent available, to discriminate one object from the other and cautioned such knowledge may not always be available in the real world,” the GAO report said. Officials for the National Missile Defense program had decided to use only one decoy in early tests because an independent panel — the Welch Panel — said tests to hit and destroy missiles are difficult and should not be complicated with many decoys at first, the GAO report said.
| ||||||||||
Back to top
