The Honorable Philip E. Coyle
Director, Operational Test and Evaluation
House Committee on Government Reform
Subcommittee on National Security, Veterans Affairs,
and International Relations
National Missile Defense
September 8, 2000
For Official Use Only
Until Release by the
Committee on Government Reform
U.S. House of Representatives
September 8, 2000
Mr. Chairman, members of the Committee, thank you for the opportunity to discuss the testing of the National Missile Defense (NMD) system. I have not had the opportunity to address this Committee before and am pleased to do so.
You requested that todays testimony focus on the impact of the test results to date on technology maturity and deployment schedules. You also indicated we address the relationship between the Anti-Ballistic Missile (ABM) Treaty and the current proposals to design, test, and deploy an effective missile defense system. First, I would like to briefly discuss the progress so far.
PROGRESS SO FAR
The NMD program has demonstrated considerable progress towards its defined goals in the last two years. The Battle Management Command, Control, and Communications (BMC3) system has progressed well. Potential X-Band Radar (XBR) performance looks promising, as reflected in the performance of the Ground Based Radar-Prototype (GBR-P). A beginning systems integration capability has been demonstrated, although achieving full system-of-systems interoperability will be challenging.
The ability to hit a target reentry vehicle (RV) in a direct hit-to-kill collision was demonstrated in the first flight intercept test last October. However, in this test, operationally representative sensors did not provide initial interceptor targeting instructions, as would be the case in an operational system. Instead, for test purposes, a Global Positioning System (GPS) signal from the target RV served to first aim the interceptor. We were not able to repeat such a successful intercept in the two subsequent flight intercept tests. Preliminary analysis has been completed on the root cause of the failure in the most recent flight intercept test, but has not been fully determined.
Because of the nature of strategic ballistic missile defense, it is impractical to conduct fully operationally realistic intercept flight testing across the wide spectrum of possible scenarios. The program must therefore complement its flight testing with various types of simulations. Overall NMD testing is comprised of interrelated ground hardware and software-in-the-loop testing, intercept and non-intercept flight-testing, computer and laboratory simulations, and man-in-the-loop command and control exercises. Unfortunately, these simulations have failed to develop as expected. This, coupled with flight test delays, has placed a significant limitation on our ability to assess the technological feasibility of NMD.
The testing program has been designed to learn as much as possible from each test. Accordingly, the tests so far have all been planned with backup systems so that if one portion of a test fails, the rest of the test objectives might still be met. Developmental tests in a complex program, especially those conducted very early, contain many limitations and artificialities, some driven by the need for specific early design data and some driven by test range safety considerations. Additionally, the tests are designed so that they will not produce debris in orbit that will harm satellites. Also, the program was never structured to produce operationally realistic test results this early. Accordingly, it was not realistic to expect these test results could support a full deployment decision now, even if all of the tests had been unambiguously successful, which they have not been. Notwithstanding the limitations in the testing program and failures of important components in all three of the flight intercept tests, the program has demonstrated considerable progress.
Compliance with the ABM Treaty has not had an adverse impact to date on the developmental testing of the NMD system. In the future, we desire additional Ground Based Interceptor test launches from more operationally representative locations than the existing Kwajalein Missile Range. Additional target launch sites which are not restricted by the Treaty would expand the test envelope beyond that currently available, as recommended by the Welch panel, to validate system simulations over the rest of the operating regimes. Furthermore, we need a radar to skin track the incoming RV (rather than tracking a beacon transponder as has been done with the FPQ-14 radar on Oahu) during early mid-course flight in order to support creation of the Weapon Task Plan which first aims the interceptor. Some of the options for these improvements could raise ABM Treaty issues. Any NMD test activity must be sufficiently well defined in order to properly assess the ABM Treaty implication and determine whether the activity can be conducted under the existing Treaty.
Since the program was restructured in January 1999, the NMD program has experienced numerous program development delays, while the construction and production schedules have not slipped. To the programs credit, the flight test program has been event driven, with tests conducted only when the Program Office felt ready. As a result, Integrated Flight Test (IFT) IFT-3 was conducted 18 months behind the original 1996 schedule and four months behind the 1999 schedule. More recently, as illustrated in Figure 1, additional significant test slips have occurred since the January 1999 program restructure. In particular, IFT-5 was to be conducted about six months before a June 2000 Deployment Readiness Review (DRR) but was actually executed on July 8. This forced the DRR to be moved to August 2000. IFT-6, which had also been planned to precede the DRR, is expected to occur in January or February 2001.
Figure 1. Schedule Slips in the NMD Test Program
Development delays have already caused schedule slips of flight tests of the tactical booster to beyond the DRR. Boost Vehicle (BV) test #1 was originally scheduled for February 2000, then July 2000, and now second quarter of FY01. BV2 has slipped about a year. BV3, the first test to integrate the Exoatmospheric Kill Vehicle (EKV) with the booster, is behind about a year and a half. Additionally, the first use of the operational booster stack in an intercept test will now occur in IFT-8, vice IFT-7 as originally planned. As a result, the authorization of long lead acquisition for the Capability 1 (C1) interceptor system will have to be delayed commensurate with that testing.
Delays in the flight test program are the most visible, but developmental problems in simulation and ground test facilities may have an even greater impact. Since the flight test scenarios are severely constrained, ground testing and simulation are critical to evaluating system performance and the fulfillment of Operational Requirements Document (ORD) requirements.
Integrated Ground Tests (IGTs), using the computer processor-in-the-loop Integrated System Test Capability (ISTC) simulation, were to provide operationally realistic data on 13 "design-to" scenarios. A high fidelity digital simulation, the Lead Systems Integrator (LSI) Integration Distributed Simulation (LIDS), was to have been used by the contractor and Operational Test Agency (OTA) team to perform analysis of an even broader set of scenarios to demonstrate that the entire United States would be adequately defended. The ISTC proved to be too immature to provide reliable estimates of performance, and the development of the digital simulation, LIDS, is behind schedule and was not available to support analyses of overall system performance as originally intended.
Figure 2. Accumulation of Slips in Test and Development Schedule
Unless these trends are reversed, an Initial Operational Capability (IOC) in FY05 appears unlikely. Figure 2 illustrates the trend of development schedule slips and estimates schedules slipping at a rate of 20 months every three years. If these trends persist and efforts by the NMD Joint Program Office (JPO) to "buy back" schedule are unsuccessful, the first flight test with a production representative interceptor (IFT-13), scheduled for the first quarter of FY03, would slip about two years.
The NMD Test and Evaluation Program is being planned and executed by the NMD Lead System Integrator, Boeing, under the direction of the NMD Joint Program Office. The test program is derived from the current NMD Test and Evaluation Master Plan (TEMP) and aims to demonstrate, incrementally, progress toward C1 capability by fulfilling the following objectives:
In the first three years of the NMD program the Initial Development Phase test events consisted of Integrated Ground Tests (IGTs) 3, 4, and 5; IFTs 1A, 2, 3, 4, and 5; Modeling and Simulation activities; Risk Reduction Flights (RRFs); and User Exercises. This phase culminates with the DRR. Near-term test and evaluation focuses on the ability to provide accurate test information and data in support of the DRR. Test and evaluation activities are also essential for the development and maturation of system elements.
The NMD program activities following the DRR will focus on completing the development of the NMD C1 expanded system. The test and evaluation activities during this period consist of Integrated Ground Tests, Integrated Flight Tests, Modeling and Simulation, Risk Reduction Flights, and User Exercises as for the initial development phase and are intended to support developmental activities and future DAB decisions if the next President decides to authorize deployment. The next DAB will decide whether to proceed with the Upgraded Early Warning Radar (UEWR) Upgrade, XBR build, and BMC3 integration into the Cheyenne Mountain Operations Center, and two years later, the DAB will decide if the weapon system is ready for production.
Limitations on Integrated Flight Tests
The flight test program has demonstrated basic functionality of the NMD
system elements. The most notable achievements have been the hit-to-kill
intercept of IFT-3 and significant
"in-line" participation in IFT-4 and IFT-5 by system elements. However, the configuration of the NMD system during both IFT-4 and IFT-5 remains a limited functional representation of the objective system, as discussed below.
Early integrated flight tests, like IFT-4 and IFT-5, make use of surrogate and prototype elements, because the NMD program is still in its developmental phase. As such, element maturity in near-term flight testing is limited:
In part, the operational realism of integrated flight testing has been limited by having located the GBR-P at KMR. As illustrated in Figure 3, the GBR-P is not sufficiently forward in the test geometry, as it would be in many operational scenarios, requiring that other sensors provide data to the BMC3 for weapon task planning. In the integrated flight tests conducted to date and for the foreseeable future, these "other sensors" are either GPS data sent from the RV and/or the FPQ-14 radar receiving data from a C-band transponder on the target RV. The FPQ-14 radar located on Oahu, Hawaii, picks up the C-Band signal radiating from the target RV and provides the BMC3 with target track information as though it were from a UEWR. Similarly, as in IFT-3 and IFT-4, the GPS can provide the BMC3 with target track information as though it were from an X-Band Radar. In tests to date, the BMC3 was required by the concept of operations to generate a Weapon Task Plan only after the threat object the RV had been resolved by ground based radars. Although the GBR-P acting as the XBR surrogate can acquire the target cluster soon after radar horizon break, the GBR-P alone is not capable of supporting the Weapon Task Plan generation because, in the test geometry, the target RV cannot be discriminated early enough.
Figure 3. Integrated Flight Test Geometry
Another critical function performed by the BMC3 is the generation and uplink
of In-Flight Target Updates (IFTUs) target data sent to the EKV while in
flight to correct for any targeting errors. In the "on-line"
portion of IFT-3, the GBR-P acting as the XBR surrogate was not required nor
planned to be the sole provider of track data to the BMC3 for IFTU generation.
GBR-P track data was augmented by FPQ-14 data for IFTU generation. GBR-P participation in IFTU generation especially of IFTUs sent late in the engagement timeline has increased in recent flight tests. In particular, the BMC3 generated all three IFTUs exclusively from GBR-P data in IFT-5.
Characteristic of ballistic missile defense flight tests, limitations associated with developmental testing impact the operational realism of integrated flight tests. Safety concerns about intercept debris and range constraints impose limitations on engagement scenarios. While a successful intercept during any future flight test will be a significant achievement in the development of the NMD system, it should be seen in context of the caveats enumerated above as well as the following limitations:
Operational engagements for the NMD C1 System are expected to cover a much larger engagement space than what can be achieved during integrated flight tests. Figures 4, 5, and 6 illustrate the differences. Figure 4 shows that targets launched from VAFB in California toward KMR in the Western Pacific occupy one point of the target-apogee vs. target-range parameter space. Figure 5 underscores the fact that interceptor flyout in the VAFB-KMR engagement is on the very low end of the engagement space a flyout range of roughly 700 kilometers and at a fixed intercept altitude of 230 kilometers. And, Figure 6 compares the flight envelope closing velocity vs. interceptor ground range of the test program to that of the C1 engagement space. The engagement space of the test program occupies nearly a single point.
Integrated ground testing using simulated environments and full threat scenarios will be used to evaluate the performance and effectiveness of the NMD C1 system throughout the engagement envelope. These ground activities, along with modeling and simulation, are planned to mitigate flight test limitations described above. Unless additional points in the flight envelope of Figure 4 are flown in integrated flight tests, the scope and validity of system performance estimated in ground testing would remain limited.
Figure 4. Target Apogee vs. Target Range Parameter Space
Figure 5. Interceptor Flyout Comparisons
Figure 6. Closing Velocity vs. Interceptor Ground Range Parameter Space
Flight Test Results
Integrated Flight Test 1A Boeing EKV Flyby
Integrated Flight Test 1A (IFT-1A), conducted on June 24, 1997, was the first flight test of the NMD Test Program. A test was attempted in January 1997 (IFT-1) but was aborted because the surrogate for the ground based interceptor booster failed to launch. The primary objective of IFT-1A and the subsequent test, IFT-2, was to provide a basis for down-selecting candidate EKVs built by competing contractors, Boeing and Raytheon.
IFT-1A assessed the performance of the Boeing EKV sensor, collected phenomenological data used for post-test analysis of the onboard discrimination algorithms, and collected functional data on the dynamic flight-test environment and its effects on the EKV. Range assets and surrogate hardware GPS and the FPQ-14 radar tracking a C-band transponder were used to guide and deliver the EKV to a point in space where it began executing sensor functions; the BMC3 element played no role in the execution of IFT-1A. Since the EKV did not have propulsion capabilities, it was incapable of intercept but came to within 5,200 feet of the target reentry vehicle.
The principal component of the Boeing EKV design is a multiple-waveband IR sensor that allows the EKV to acquire, track, and collect data on objects of the representative threat target suite. The sensor payload consists of a focal plane array of highly sensitive silicon-based sensors and a cryogenic cooling assembly at the end of an optical telescope.
The EKV sensor payload was launched from Meck Island in the Kwajalein Atoll and set on a trajectory that permitted it to view a pre-planned target scene. The target suite was launched from VAFB using a specially configured Minuteman II booster and consisted of nine objects: one medium reentry vehicle, two medium rigid light replicas, one small canisterized light replica, two canisterized small balloons, two medium balloons, and a large balloon. Viewing objects of the target suite, the EKV seeker successfully gathered signature and phenomenology data which, in turn, were used to verify predictions made by corresponding models and simulations. One of the medium balloons did not fully inflate.
Integrated Flight Test 2 Raytheon EKV Flyby
Integrated Flight Test 2 (IFT-2) conducted on January 16, 1998, was the second flight test of the NMD Test Program. The objectives of IFT-2 were the same as that for IFT-1A, namely, to assess the performance of the EKV sensor built by the second EKV contractor, Raytheon Missile System Company. The same target suite of nine objects was flown.
EKV seeker data was downlinked and used for evaluating sensor performance and for performing post-test discrimination and signature analyses of the target suite. Range assets and surrogate hardware GPS and the FPQ-14 radar tracking a C-band transponder guided the EKV to a point in space where it began executing sensor functions; the BMC3 element played no role in the execution of IFT-2. As in IFT-1A, the Raytheon EKV did not attempt to intercept the medium reentry vehicle since it had no propulsion capabilities.
The principal component of the Raytheon EKV design is a multiple-waveband, Visible/IR sensor payload that allows the EKV to acquire, track, and collect data on objects of the representative threat target suite. The sensor payload consists of an HgCdTe focal plane array and a cryogenic cooling assembly at the end of an optical telescope. As in the launch of the Boeing EKV, the Raytheon EKV sensor payload was launched from Meck Island at KMR and set on a trajectory that permitted it to view a similar target scene of ten objects (nine objects of the target suite plus the deployment bus). And, as in IFT-1A, one of the medium balloons did not fully inflate.
IFT-2 was successful in collecting target object data, and post-test analyses demonstrated that the MRV could be discriminated from the other objects of the target suite. Because the discrimination algorithms were not executed in real time and relied on simulations that were anchored by IFT-2 test data, the successful discrimination of the medium reentry vehicle should not be viewed as a verification of the discrimination algorithms in an operational engagement, but rather, as a successful experiment.
At the recommendation of the Lead System Integrator (Boeing North American), the NMD Joint Program Office opted to down-select to a single EKV design prior to IFT-3, which afforded more intercept test opportunities before the DRR. The Joint Program Office selected Raytheon as the EKV contractor over Boeing.
Integrated Flight Test 3 Intercept Achieved
The first NMD intercept attempt of a target reentry vehicle by the Raytheon-built EKV was successful, albeit with significant limitations to operational realism, on October 2, 1999. IFT-3 began with the launch of a Minuteman-based booster from VAFB and the subsequent deployment of its target payload MRV and Large Balloon for reentry near KMR. An interceptor was launched from Meck Island to engage the MRV, and EKV intercept of the MRV occurred at an altitude of 230 km, 1,782 seconds after target liftoff. IFT-3 was planned and jointly executed by the NMD Joint Program Office and Boeing, the LSI. Future flight tests are being planned and executed by Boeing.
IFT-3 was an element test of the Raytheon-built EKV, not an Integrated System Test. IFT-3 was comprised of two concurrent test activities: an "in-line" test that focused on the performance of the EKV, and a simultaneous "on-line" or shadow test that focused on assessing NMD functionality as an integrated system using prototype elements that approximate the objective system. The principal objective of the on-line test was to demonstrate integration and operation of system elements as a risk reduction effort for future flight tests, IFT-4 and IFT-5.
IFT-3 In-Line Test (EKV Flight Test)
The in-line or flight test part of IFT-3 was a test of the Raytheon-built EKV.
GPS track information of the target RV was used to guide and deliver the EKV to
a point in space where it began executing mission-critical functions: midcourse
guidance, target-complex acquisition,
real-time discrimination, target selection, active homing, and intercept. Although the EKV successfully intercepted the MRV, acquisition of the target complex by the EKV was accomplished in an off-nominal manner because of a malfunctioning Inertial Measurement Unit (IMU) onboard the EKV. The IMU problem was caused by a vendor calibration procedure error, which was corrected for IFT-4.
Because of the problem with IMU operation, the EKV was forced to utilize its "step-stare" capability that is activated only during off-nominal situations.
IFT-3 On-Line Test (Shadow Test)
The on-line portion of IFT-3 ran in parallel with the in-line test to assess the performance of NMD functionality as an integrated system using prototype and surrogate elements. Elements operating on-line did not affect the operation of the in-line test but did demonstrate NMD functionality in a configuration more representative of the integrated system that might be deployed. The most notable results of the IFT-3 on-line test pertained to BMC3 and GBR-P performance.
The BMC3 successfully demonstrated integrated system performance through the coordination of system elements operating in shadow mode. It performed engagement planning that ultimately led to a successful (simulated) mission. GBR-P performance was generally poor and unsuitable for anchoring associated radar simulations. GBR-P track quality was adversely affected by a software error in the antenna mount motion equation. A software fix was implemented and later verified in the target of opportunity flight, RRF-7, which was conducted in November 1999, and in IFT-4 and IFT-5.
Integrated Flight Test 4 Intercept Not Achieved
Integrated Flight Test 4, which was conducted on January 18, 2000, was the first end-to-end NMD flight test attempting a hit-to-kill intercept of a target reentry vehicle. Whereas IFT-3 was an element test of the Raytheon-built EKV, IFT-4, using surrogate and prototype elements, strived to demonstrate NMD system integration in a configuration more representative of the system that might be deployed. In particular, both the BMC3 and the GBR-P participated in the flight test "in-line." The FPQ-14 radar located in Oahu, Hawaii, was to have used the C-Band transponder data from the MRV to provide the BMC3 with target track information as though it were from a UEWR. The FPQ-14 data, however, was (erroneously) judged in real time to be of poor quality. Instead, GPS track data of the MRV was used in IFT-4 after being translated into XBR format. The geometry of the test scenario of IFT-4 was identical to that of IFT-3.
The EKV failed to intercept the MRV, a failure directly traceable to the cryogenic cooling system of the EKV. The primary cooling line that delivers krypton to the IR focal plane arrays was restricted with either frozen moisture or other contamination, and the IR sensors were prevented from cooling down to their operating temperatures. Consequently, the IR sensors did not acquire or track target objects for terminal homing and intercept.
IFT-4 demonstrated the successful operation and integration of NMD elements. Data analysis of IFT-4 has been completed, and the following assessment of test results can be made:
Integrated Flight Test 5 Intercept Not Achieved
Integrated Flight Test 5 was conducted on July 8, 2000. It was to be an end-to-end NMD intercept flight test nearly identical to IFT-4 and aimed to demonstrate NMD system integration with surrogate and prototype elements in a configuration representative of the system that might be deployed. The most prominent new feature of the test was the participation of the In Flight Interceptor Communications System as the communication link between the BMC3 and EKV. As in all previous intercept tests, a Minuteman-based target system was launched from VAFB, and its target payload consisting of an MRV was deployed for reentry near KMR. The target payload also included a Large Balloon, but it was never deployed because of some unknown failure of the deployment mechanism. Then, at 1,294 seconds after target liftoff, an interceptor was launched from Meck Island to engage the MRV. The planned intercept, which did not occur, was to have been at an altitude of 230 km, 1,782 seconds after target liftoff, identical to the planned intercepts on IFT-3 and IFT-4.
The failure to intercept the MRV is the direct result of the EKV not separating from the upper stage assembly of the Payload Launch Vehicle, the surrogate for the interceptor booster. Preliminary failure analysis of the telemetry data indicates that the EKV did not receive a second-stage burnout message, a prerequisite for initiating the separation sequence. The cause of this failure has not yet been determined but appears to be isolated to the Payload Launch Vehicle. A notable consequence of the failure is that all EKV events subsequent to separation, e.g., sensor operation and divert and attitude activities, did not occur. Therefore, none of the EKV primary objectives were met.
The FPQ-14 radar located at the Kaena Point Satellite Tracking Station in Oahu, Hawaii, which tracked the C-Band transponder on the MRV, played an important role in IFT-5. Unlike IFT-4 in which GPS track data was the source for Weapon Task Plan generation, the BMC3 generated the Weapon Task Plan using FPQ-14 transponder data. GPS was still used, however. The FPQ-14 data, prior to being used to generate the Weapon Task Plan, was checked against the GPS track for accuracy; GPS data could have been used in the event that FPQ-14 data was of poor quality. The Weapon Task Plan directed the launch of the interceptor at 1,294 seconds time after liftoff (TALO).
The GBR-P, the prototype X-Band Radar, successfully participated in IFT-5 as an integrated element of the system. It received target cluster cues from the BMC3, tracked all objects of interest, and correctly performed real-time discrimination on all target objects. The GBR-P tracking and discrimination timeline of IFT-5 closely matched the timeline predicted by pre-mission simulations, except that MRV acquisition occurred earlier than predicted. GBR-P participation in integrated flight tests is increasing. In IFT-5, all In Flight Target Updates (IFTUs) including the backup IFTU were generated solely from GBR-P track data. However, GBR-P track data was prevented from entering the BMC3 element until after the Weapon Task Plan had been sent to the Weapon System and, therefore, did not contribute to Weapon Task Plan generation.
IFT-5 demonstrated integrated system performance through the operation of the non-tactical, flight-test version of the BMC3. The BMC3 provided end-to-end tracking of the target complex utilizing multiple sensor sources and demonstrated all operations of engagement planning and real-time communications. It successfully generated the Weapon Task Plan, Sensor Task Plans, Communication Task Plans, and IFTUs. Failure of EKV operation precluded the successful in-line operation of the In-Flight Interceptor Communications System (IFICS) closure of the BMC3-EKV communication link and, thus, associated objectives were not fully achieved, e.g., the receipt of In Flight Status Reports from the EKV were not evaluated. System integration of early warning elements with the BMC3 was achieved: DSP satellites successfully acquired the boosting Minuteman II target vehicle and sent Quick Alert and Boost Event Reports to the BMC3. The EWR also acquired and tracked the target complex, including spent fuel tanks, early in the mission timeline.
Integrated Ground Tests
Boeing is performing ground testing to mitigate the risks associated with the limited flight test program. Ground testing can exercise the system through variation of threat characteristics such as launch point, aimpoint, trajectory, apogee, number of RVs, target type, and environmental effects. This ground testing is done in month-long phases called Integrated Ground Tests. IGT-4 and IGT-5 occurred in 1999; IGT-6 will not occur until after the DRR.
These ground tests use the ISTC at the U.S. Army Space and Missile Defense
Commands Advanced Research Center in Huntsville, Alabama. ISTC provides test
execution and control, threat and environment data, and test drivers for some
NMD elements. Each NMD element is represented at a standalone computer station
called a node. Each node incorporates system element mission and communications
processors, which run prototype element software. ISTC supplies the nodes with
simulated inputs threats and associated environments, natural and
man-made which are nominally consistent for each NMD element in the scenario.
IGTs use a combination of models, software-in-the-loop, and hardware-in-the-loop to test the NMD engagement space and threat in an operational environment. They are supposed to validate the functionality and functional interfaces between the elements, subject the system to stressing environments and tactical scenarios, and evaluate target-intercept boundary conditions. IGTs can help to identify "unknowns" in an interactive system context and verify interoperability of NMD system elements.
There was very little operational hardware or software used in IGT-4 or IGT-5. The BMC3 was a prototype, flight-test version of the operational BMC3; it included some real communications hardware (T1 links). It is possible that some of the software in the UEWR representation could eventually be used in the operational UEWR. Also, some of the EKV digital signal processing software and data processing software might be used in the operational EKV.
The element hardware components are represented digitally in the Processor Test Environment. It duplicates the real-time tactical interfaces in order to inject the perceived data into the test article. For example, the Processor Test Environment for the GBR-P element contains simulation software that represents the transmitter, receiver, antenna, signal processor, measurement generation, beam volume, detection response, and radar status.
IGT-4 and IGT-5 had a number of limitations. For example, the threat apogees were unrealistically high in IGT-4, which provided optimistic assessments of timelines and radar detections. Because the simulation had limited processing capability, Boeing (LSI) eliminated most of the threat objects in many of the scenarios, which was unrealistic for testing discrimination, radar resource management, and BMC3 processing capabilities. In addition, all of the element representations suffered from limitations that produced significantly different performance than is expected from the NMD C1 system. These limitations included, but were not limited to:
The primary goal of IGT-4 and IGT-5 was to demonstrate the integration of BMC3 with the UEWR and XBR. Boeing successfully demonstrated integration between these three NMD elements in the two IGTs. The secondary goal of the IGTs was to assess the C1 architecture and performance against a limited set of C1 scenarios. This goal was less successful, in part because of the immaturity of the element representations in IGT-4 and IGT-5. The exact amount attributable to element model immaturity is currently undefined and will remain so until truly element-representative models are installed in ISTC.
Boeing demonstrated integration between the BMC3 and radars by generating and recording messages between the elements. They confirmed that the planned messages had been exchanged between the BMC3 and the GBR-P and UEWR, and measured the time delays between the messages.
The radar performance in IGT-4 and IGT-5 was generally poor. In IGT-4 the XBR had reasonable position track performance but the velocity track performance was much worse than specifications. The XBR improved in IGT-5 and usually met the track accuracy performance. The UEWR failed to detect a significant number of RVs in IGT-4 and IGT-5. Once an RV was acquired, the performance of the UEWR representation at a given time was generally much better than specifications in both position and velocity tracking. However, the UEWR rarely succeeded in maintaining the specified track accuracies against RVs throughout an engagement. The probability of track maintenance was well below the NMD system specification requirements for both the XBR and UEWR. The XBR discrimination results were also well below the NMD system specification requirements.
The ISTC hardware and software used to date in the IGTs are immature and do not provide an adequate representation of the NMD C1 architecture. None of the major NMD elements BMC3, XBR, UEWR, Weapon System, and DSP/SBIRS are mature enough to provide a good assessment of the C1 system. The 1997 TEMP discussed the consequences if the representations were not mature before the DRR: "The validity and credibility of the surrogates and the representations must be fully characterized with respect to the NMD system and element requirements prior to making any decisions based on data drawn from tests using these systems. Without this information, the results of the tests will be inconclusive at best and misleading at worst." IGT-4 and IGT-5 did demonstrate the integration of the BMC3 with the UEWR and XBR (not with the weapon system, however), but these tests provided only limited data to support an evaluation of the effectiveness of the initial, proposed NMD C1 system at the DRR.
Battle Planning Exercise 99-5 and BMC3 Assessment
Battle Planning Exercise 99-5 (BPEx 99-5) was conducted in the BMC3 Element Laboratory at the Joint National Test Facility on September 28-30, 1999. Conceived in 1998 by U.S. Space Command (USSPACECOM/J35), BPEx events enable the User to examine and assess as-built BMC3 operational functionality for the purpose of influencing future development of the BMC3 element. The OTA Team was invited by USSPACECOM to co-lead BPEx 99-5 to benchmark BMC3 behavior in support of the Deployment Readiness Review.
The primary objective of BPEx events is to identify operational defects of the BMC3 element to be corrected in future builds. BPEx 99-5 was performed, in particular, to evaluate BMC3 element behavior in support of the OTA Teams early operational assessment of Key Performance Parameters #2 and #3 human in control (HIC) and automated battle management for the DRR. The evaluation of Key Performance Parameter #1, effectiveness of the NMD system to defend the United States against ballistic missile attacks, was not an objective of BPEx 99-5. The test environment representing the NMD system consisted of the following components:
Notable BMC3 Behavior
The following BMC3 behavior was observed during BPEx 99-5 execution:
The BMC3 element is currently at an early stage of development and noted shortcomings are likely to be addressed before the initial operational capability. NMD operators had difficulty with resource management, engagement control, and situation awareness.
The LSI is developing the BMC3 with maximum automation. Inherently, the BMC3 is designed to preclude direct launch control by the operator. Rather, positive control is exercised through Rules-of-Engagement development, battle-planning development, and management by exception. The BPEx, therefore, reflects the outcome of these efforts and can be frustrating to an operator attempting real time control.
Modeling and Simulation
Restrictions on realistic operational flight testing, and the complexity of the operational engagements, require the T&E program to rely heavily on integrated ground testing and the execution of digital simulations for assessing the operational suitability and effectiveness of the NMD system concept. Integrated ground testing was of limited utility in assessing the potential performance of the NMD system. Late delivery of LIDS a high fidelity, system-level digital simulation of the NMD system precluded its use for making a credible assessment of potential NMD system performance.
LIDS model development is taking much longer than expected. It was to be the principal digital simulation tool providing DRR support. Modeling and simulation in general and LIDS in particular were supposed to be employed to repeat hypothetical experiments in order to improve the statistical sample and to determine the values of key technical parameters unable to be measured by testing. Boeing released a beta version LIDS Build 4 at the end of April 2000. There was not enough time before the DRR to accredit LIDS and perform the required system analyses. As a result, the Service Operational Test Agencies do not have a simulation that they can use to assess the potential system effectiveness.
LIDS build 4 has serious limitations, so even if it had been released on time there would still be major issues in using LIDS to assess the potential performance of the NMD system. One problem is that LIDS users will not be able to generate their own scenarios. Boeing will provide users with canned scenarios, including fixed launch points, aim points, Inter-Continental Ballistic Missile (ICBMs), debris, and apogees. The Operational Test Agencies had been planning to run hundreds of digital simulation scenarios, varying such parameters as raid size, trajectories, atmospherics, debris, nuclear effects, threat launch and impact points, threat types, and Penetration Aids. LIDS will not have the flexibility to support such studies.
LIDS will allow users some flexibility. They will be able to change the location and number of the various NMD elements. Users will also be able specify such parameters as the reliability of GBI boost phase completion, the probability of target acquisition by the EKV sensor, the probability of the EKV correctly identifying the RV, the probability of hitting the RV given correct discrimination, and the probability of killing the target given a hit. Such analyses will be useful but not sufficient to adequately assess the potential performance of the C1 system.
LIDS does not simulate any of the element prototypes or surrogates currently used in flight testing. Consequently, use of the IFTs to provide traditional model validation data will not be possible until the actual system elements finally work their way into the intercept flight test program. This limits the confidence that can be placed on LIDS predictions in the foreseeable future.
Boeing is using a number of low-fidelity simulations in their development of the NMD system. One is NMDSim, which estimates the interceptor launch windows for different scenarios. The NMDSim does not simulate discrimination functionality, does not generate weapon task plans, has no interceptor flyout representation, and does not perform kill assessment. It can be a useful tool for planning engagements in higher-fidelity models or simulations, but it is too limited to credibly assess the potential performance of the NMD system.
NMD lethality testing and analysis activities before the DRR have focused on the development and accreditation of version 8.1 of the Parametric Endo-Exoatmospheric Lethality Simulation (PEELS). PEELS is the only lethality simulation to be accredited for endgame evaluation of NMD intercepts. In effect, it is the simulation used in both lethality and effectiveness analyses to assess whether an NMD hit on a threat target results in a target kill. To develop an NMD-capable version of PEELS, the database of empirical results that anchors the simulation for theater ballistic missiles had to be expanded to include lethality information for intercepts of NMD-type targets by the EKV in the velocity regime expected for NMD engagements. Because there is no capability to run ground tests at the upper end of NMD intercept velocities, a series of hydrocode analyses were used to generate the bulk of the "empirical data" for NMD EKV intercepts.
A total of 490 hydrocode simulations are planned, covering the quarter-scale Light Gas Gun test projectile, warhead and aeroshell damage, and different threat targets and intercept parameters. Of these, 218 have been completed to date, namely, 178 for the Attitude Control Reentry Vehicle target and 20 for Medium Lethality Reentry Vehicle target. The main purpose of the quarter scale Light Gas Gun series was to generate instrumentation data and damage data, which are used to anchor the hydrocode prediction methodology for varying hit points, velocities, and impact angles.
A series of 20 quarter-scale light-gas-gun impact tests were conducted at the Arnold Engineering Development Center in Tennessee in 1999 against Attitude Control Reentry Vehicle targets, and a second series of 20 shots have begun testing in FY00 against the Medium Size Reentry Vehicle, Long Range Nuclear Threat, and Attitude Control Reentry Vehicle targets. These tests employ a quarter-scale surrogate of the EKV launched against a quarter-scale replica of the target at a nominal velocity of 7 km/s. FY99 test results are described in the U.S. Army Space and Missile Defense Command Test Report. A report comparing test results to hydrocode predictions, originally scheduled for publication in April 2000, is still pending.
Besides providing a backup for the hydrocode prediction methodology, the 1999 tests provided the following information:
Additional testing is being done to improve and validate the hydrocode simulations. Sandia National Laboratory is conducting a set of high-speed impact tests using a three-stage Light Gas Gun to develop the equations of state the characterization of the physical phenomena that occur during impact of several aerospace materials present in the test targets and EKV at impact velocities of 6 km/s and 12 km/s. The materials studied are silica phenolic, E-glass, and graphite epoxy. Testing is expected to be completed later this year. If significant differences between the new empirically-derived equations of state and inputs used for the hydrocode runs are found, the hydrocode analysis will be corrected and PEELS modified accordingly. Results to date suggest that such modifications will not be necessary.
Sandia is also performing a series of hydrocode analyses for the Attitude Control Reentry Vehicle and Medium Target Reentry Vehicle targets. Their objective is to characterize the lethal volume for aerothermal structural kills. Aerothermal structural kills could occur if the target incurs sufficient damage from an EKV impact and suffers aerothermal demise during atmospheric reentry. As of March 2000, 93 hydrocode runs had been made. The analyses are expected to continue through 2000.
Based on the accumulated data from lethality tests and analyses, PEELS 8.1 was accredited by the Accreditation Working Group (AWG) on April 4, 2000. In the accreditation report dated April 28, 2000, the AWG recommends accreditation of PEELS 8.1 for the following experiments:
The accreditation report has specified the following caveats under the recommendation for accreditation approval.
The quarter-scale Light Gas Gun testing conducted to date utilized a low fidelity surrogate of the EKV that matched the average mass properties of both the Raytheon and Boeing EKV concepts but not their precise structure or materials. The results obtained could be representative of the grosser aspects of NMDs direct hit lethality against the Attitude Control Reentry Vehicle target. The tests showed that damage to NMD targets from direct hit by the EKV will depend on the location of the impact within the payload. Not every hit would necessarily result in a kill.
The hydrocode analyses provided predictions of expected NMD lethality against threat targets in the hypervelocity regime and supported the development of the lethal volume in PEELS version 8.1 and enabled its use as a tool for DRR analysis.
After DRR, the development of the Live Fire Test and Evaluation (LFT&E) program will be addressed in the NMD Lethality IPT under the joint leadership of the JPO and the LSI. Although the LFT&E strategy is yet to be finalized, it is expected to include three flight tests: reduced-scale light gas gun tests, hydrocode analyses, and PEELS analyses.
FUTURE TEST PLANNING
Under the program-of-record, test results are not likely to be available in 2003 to support a recommendation then to deploy a C1 system in 2005. This is because the currently planned testing program is behind, because the test content does not yet address important operational questions, and because ground test facilities for assessment are considerably behind schedule.
NMD developmental testing needs to be augmented to prepare for realistic operational situations in the Initial Operational Test and Evaluation (IOT&E) phase, and is not yet aggressive enough to keep pace with the currently proposed schedules for silo and radar construction and missile production. The testing schedule, including supporting modeling and simulation, continues to slip while the construction and production schedules have not. Important parts of the test program have slipped a year in the 19 months since the NMD program was restructured in January 1999. Thus, the program is behind in both the demonstrated level of technical accomplishment and in schedule. Additionally, the content of individual tests has been diminished and is providing less information than originally planned.
I am especially concerned that the NMD program has not planned nor funded any
intercept tests until IOT&E with realistic operational features such as
multiple simultaneous engagements, long-range intercepts, realistic engagement
geometries, and countermeasures other than simple balloons. While it may not be
practical or affordable to do all these things in developmental testing,
selected stressing operational requirements should be included in developmental
tests that precede IOT&E to help ensure sufficient capability for
deployment. For example, the current
C-band transponder tracking and identification system, justified by gaps in radar coverage and range safety considerations, is being used to provide target track information to the system in current tests. This practice should be phased out prior to IOT&E. This will ensure that the
end-to-end system will support early target tracking and interceptor launch.
There is nothing wrong with the limited testing program the Department has been pursuing so long as the achieved results match the desired pace of acquisition decisions to support deployment. However, a more aggressive testing program, with parallel paths and activities, will be necessary to achieve an effective IOC by the latter half of this decade. This means a test program that is structured to anticipate and absorb setbacks that inevitably occur. The NMD program is developing test plans that move in this direction.
The time and resource demands that would be required for a program of this type would be substantial. As documented in the Congressional Budget Office (CBO) report on the budgetary and technical implications of the NMD program, the Safeguard missile program conducted 165 flight tests. The Safeguard program was an early version of NMD. The SPRINT program conducted 42 test firings in a five-year period between 1965 and 1970, more than 8 per year, before its first intercept-like test. Over the next three years, SPRINT flew 23 intercept-type tests before production. The Spartan program fired 15 missile tests between 1968 and 1969 before conducting 24 intercept-type tests over the next five years. Similarly, the Polaris program conducted 125 flight tests, and the Minuteman program conducted 101 flight tests. Rocket science has progressed in the past 35 years, and I am not suggesting that a hundred or more NMD flight tests will be necessary. However, the technology in the current NMD program is more sophisticated than in those early missile programs, and we should be prepared for inevitable setbacks. More recently, in the 1980s, the Peacekeeper (MX) program launched 15 missiles in the four years before its IOC, ramping up from three flight tests per year to five flight tests per year between 1983 and 1986. It is apparent from these test schedules that an extensive amount of work was done in parallel from one flight test to another. Failures that occurred were accepted, and the programs moved forward with parallel activities as flight testing continued.
As in any weapons development program, the NMD acquisition and construction schedules need to be linked to capability achievements demonstrated in a robust test program, not to schedule per se. This approach supports an aggressive acquisition schedule if the test program has the capacity to deal with setbacks. On three separate occasions, independent panels chaired by Larry Welch (General, USAF Retired) have recommended an event driven, not schedule driven, program. In the long run, an event driven program may take less time and cost less money than a program that must regularly be re-baselined due to the realities of very challenging technical and operational goals.
OBSERVATIONS AND CONCLUSIONS
Aggressive flight testing, coupled with comprehensive hardware-in-the-loop and simulation programs, will be essential for NMD. Additionally, the program will have to adopt a parallel, "fly through failure," approach that can absorb tests that do not achieve their objectives in order to have any chance of achieving an FY05 deployment of an operationally effective system. As noted by CBO, the Navys Polaris program successfully took such an approach 30 years ago.
Deployment means the fielding of an operational system with some military utility which is effective under realistic combat conditions, against realistic threats and countermeasures, possibly without adequate prior knowledge of the target cluster composition, timing, trajectory or direction, and when operated by military personnel at all times of the day or night and in all weather. Such a capability is yet to be shown to be practicable for NMD. These operational considerations will become an increasingly important part of test and simulation plans over the coming years.
In particular, more work is needed in the following areas:
The target sets for the three intercept flight tests conducted so far have only included a single target RV with a single large balloon that did not resemble the target RV in those features which the NMD system might use for discrimination and which an enemy might try to employ. The large balloon is an unrealistic representation of the threat, and operational NMD capability has not yet been demonstrated against the simplest of realistic, unsophisticated countermeasures. No tests against such decoys are planned until IFT-10, now scheduled for the first quarter of FY03, at the earliest, when balloons alone may be flown that may have signatures but not shape, or motion, similar to the target RV.
The NMD Program is planning flight intercept tests with different balloon types and sizes which become more difficult to discriminate as the testing program moves forward. In addition, the NMD Program is planning tests with other types of decoys in non-intercept "risk reduction" flights. Eventually, intercept flight tests with such decoys will be needed as well. For example, no flight intercept tests have been conducted or are scheduled with tumbling target reentry vehicles or decoys designed to resemble tumbling RVs, perhaps the easiest RV for an enemy to deploy.
Intercept tests so far have used essentially identical trajectories, where the intercept points were known and planned in advance, as required for range safety. More operationally realistic scenarios will need to be developed, including long range intercepts and multiple simultaneous engagements.
Like the kill vehicles, X-Band Radars should be able to deal with unsophisticated decoys that resemble the target RV in signature, shape, and/or motion. We have not yet determined the operational ability of X-Band Radars to discriminate target RVs from such decoys in an intercept flight test. Also, new sensors may be required on the ground or in space on satellites. Again, these sensors have not been tested as part of the NMD architecture.
In the flight intercept tests so far, GPS or C-band beacon transponders have been used by the BMC3 to create the Weapon and Sensor Task plans which first aim the interceptor and the GBR-P. These sensors will need to be separated from the operational system in future tests prior to IOT&E.
Much of the operational context for assessing NMD is to be provided by end-to-end simulation tools which have not progressed as planned. The Lead System Integrator Integration Distributed Simulation (LIDS) has not achieved the planned level of operability or realism which was to have been available to support the DRR. Other alternative simulations have been pieced together to assess the potential of the NMD system. The Integrated System Test Capability processor-in-the- loop facility is not yet adequate to produce valid Integrated Ground Test results for system effectiveness assessment. Hardware-in-the-loop facilities need to be developed in time to support meaningful testing against countermeasures. Overall, modeling and simulation efforts are considerably behind schedule and also have not yet produced results that would support a recommendation to deploy.
We have no flight intercept test results yet to demonstrate the residual capability of a C1 system to handle the unsophisticated countermeasures that would be expected to be contained in accidental or unauthorized launches.
The test results to date also do not support a recommendation to deploy an expanded C1 capability by 2007 with additional interceptors and radars. Initial capability C1 interceptors may need to be upgraded for the expanded capability, as new test results emerge and as new information becomes available about the threat.
Testing is currently designed to accommodate an aggressive pace of development. Flight testing, however, needs to aggressively increase in complexity to keep pace with NMD C1 development and to adequately stress design limits, particularly for the missile system.
? How will an EKV respond to another EKV in its field of view, or multiple RVs in its field of view?
How is the performance of an EKV seeker affected by a thrusting EKV or another EKV intercepting an object in its field of view?
? Can the X-Band radar simultaneously track multiple RVs that require different antenna orientations?
Can the IFICS communicate with multiple KVs?
Current test range limitations need to be removed to adequately test the NMD system.
Avoidable limitations to operational realism must be removed before conduct of IOT&E.
Plans for providing adequate spares should be developed, especially for targets where current target components can be as much as 30 years old.
Ground Testing and Simulation
An innovative new approach needs to be taken towards HWIL testing of the EKV, so that potential design problems or discrimination challenges can be wrung out on the ground in lieu of expensive flight tests.
Current analysis of exoatmospheric lethality is limited to computer simulations and light gas gun tests.
LIDS development has taken much longer than originally promised. Additionally, it is practically a hard-wired simulation that only the Boeing developers can modify. This precludes independent, government sensitivity analysis and assessment.
Discrimination by the radar and weapon system (EKV) should be given more weight in performance criteria. All other aspects of the NMD performance requirements appear to be within the state of the art of technology. Discrimination by the EKV on the other hand will be the biggest challenge to achieving a hit-to-kill intercept. Decoys that provide a close representation of the RV or modify the RV signature have only been minimally investigated.
ORD Reliability Requirements
The NMD requirements for reliability, availability, and effectiveness are specified in the NMD ORD. When these requirements are allocated to the individual elements of the NMD system, the resulting reliability performance standards are unrealistically high as well as difficult to test. As the program develops, it may be necessary to re-examine the overall requirements for NMD reliability and availability.
Risk Reduction Efforts
The following programs can make significant contributions to risk reduction efforts if properly utilized.
Countermeasures Hands-On Program (CHOP)
The Ballistic Missile Defense Organization sponsors a red team approach to the possible development of countermeasures. Operated at very modest funding levels, CHOP develops and demonstrates Rest-of-World (ROW) countermeasures that could be challenging for U.S. missile defense systems. By charter, CHOP does not try to develop "sophisticated" countermeasures. However, the unsophisticated, ROW countermeasures they do develop are realistic and challenging and should be included as an integral part of the NMD flight testing and ground test HWIL simulation programs.
Operations in a Nuclear Environment (OPINE)
The NMD Program Office chartered a red team to look at OPINE testing and facility requirements for the EKV. The red team found the Raytheon-proposed test and parts screening program to be inadequate.
Hit to Kill
The NMD Program Office should investigate lethality enhancement options for dealing with potential countermeasures, using relatively simple techniques, that try to alter the effective RV size or shape in an attempt to foil discrimination and aimpoint selection.
Mr. Chairman, I want to thank you again for the opportunity to discuss these matters today. There are many important issues which justify the oversight of this Committee.
Much progress has been made, and much remains to be learned and accomplished.
A key to success will be a vigorous and robust testing program.
ABM Anti-Ballistic Missile
AWG Accreditation Working Group
BMC3 Battle Management Command, Control, and Communications
BMD Ballistic Missile Defense
BPEx Battle Planning Exercise
C1 Capability 1
C2Sim Command and Control Simulation
CBO Congressional Budget Office
CHOP Countermeasures Hands-On Program
COTS Commercial Off The Self
DAB Defense Acquisition Board
DIA Defense Intelligence Agency
DoD Department of Defense
DOT&E Director, Operational Test and Evaluation
DRR Deployment Readiness Review
DSP Defense Support Program
EKV Exoatmospheric Kill Vehicle
FY Fiscal Year
GBI Ground Based Interceptor
GBR-P Ground Based Radar-Prototype
GPS Global Positioning System
HWIL Hardware in the Loop
ICBM Inter-Continental Ballistic Missile
IFICS In-Flight Interceptor Communications System
IFT Integrated Flight Test
IFTU In Flight Target Update
IGT Integrated Ground Test
IMU Inertial Measurement Unit
IOC Initial Operational Capability
IOT&E Initial Operational Test and Evaluation
IPT Integrated Product Team
ISTC Integrated System Test Capability
ITW/AA Integrated Tactical Warning / Attack Assessment
JPO Joint Program Office
KMR Kwajalein Missile Range
LFT&E Live Fire Test and Evaluation
LIDS LSI Integration Distributed Simulation
LSI Lead System Integrator
MRV Medium Reentry Vehicle
MSE Multiple Simultaneous Engagement
NCA National Command Authority
NMD National Missile Defense
NMDSim National Missile Defense Simulation
NORAD North American Aerospace Defense Command
OPINE Operations in a Nuclear Environment
ORD Operational Requirements Document
OTA Operational Test Agency
PEELS Parametric Endo-Exoatmospheric Lethality Simulation
PLV Payload Launch Vehicle
RRF Risk Reduction Flight
RV Reentry Vehicle
SBIRS Space Based Infrared System
TEMP Test and Evaluation Master Plan
TPM Technical Performance Measure
UEWR Upgraded Early Warning Radar
USSPACECOM U.S. Space Command
VAFB Vandenberg Air Force Base
XBR X-Band Radar