PHOTO F/X2

he Johns Hopkins University Applied Physics Laboratory (JHU/APL) has served as an independent evaluator of the U.S. Navys Fleet Ballistic Missile Strategic Weapons System since 1957. In that role, the lab has developed a comprehensive test and evaluation program to validate the integrated weapons system design for three generations of nuclear-powered submarines (Polaris, Poseidon, and Trident) and six variants of the submarinelaunched ballistic missile (SLBM): Polaris A1, A2, and A3; Poseidon C3; Trident I (C4); and Trident II (D5). The increasingly stringent targeting requirements imposed on the Trident I and II weapons system exceeded the capabilities of the systems designed to test it and required innovation in the current test and evaluation technology.

T
June 2003

Marc Camacho and Sung Lim

APL developed a guidance system evaluation concept based on tracking signals received at the missile from a satellite-based navigation system. APL selected the global positioning system (GPS), which in 1974 was still in its early design stages, over another custom-designed constellation. One unusual feature of the evaluation system is the use of a GPS signal translator rather than a GPS receiver. The designers named this program SATRACK to emphasize its capability to receive, rebroadcast, record, and track the satellite signals. This article presents the SATRACK concept, along with the descriptions of various GPS translators, signal recording equipment, and the postflight processing facility.

IEEE Instrumentation & Measurement Magazine

1094-6969/03/$17.002003IEEE

37

SATRACK Concept

concluded that only a satellite-based measureGuidance system evalment system could uation of the earlier meet future requireweapon systems (Poments. By 1974, the laris and Poseidon) reSATRACK system, inlied on impact scoring cluding the evaluation techniques. This shoot methodology, hardware and score approach development, and modcompared observed test impact statistics with results com- ification requirements for the missile and the telemetry stations, puted from models used for development. This technique began moving forward. was unacceptable for evaluating the more precise accuracy Figure 1 outlines the basic SATRACK concept. Signals for requirements of the Trident system, and the U.S. Navy GPS satellite navigation are two L-band frequencies, designeeded a new methodology that provided insights into ma- nated L1 and L2. The primary frequency, L1 at 1575.42 MHz, is jor error contributors within the flight test environment so modulated with two different ranging codes: a narrowband that accuracy projections could be based on a high-confi- clear/acquisition (C/A) code with a 2-MHz bandwidth and a dence understanding of the underlying system models. As- wideband encrypted protected (P) code with a 20-MHz bandsessing performance models in the flight test environment width. The secondary frequency, L2 at 1227.60 MHz, carries requires guidance-independent measurements with suffi- only the P-code modulation. These modulations applied to cient precision to separate out the important contributors to each frequency provide the basis for epoch measurements system inaccuracy. used to determine the distance to each satellite (range meaThe existing range instrumentation that was used for mis- surements). Tracking of dual-frequency GPS signals (L1 and sile tracking and trajectory estimation was largely provided L2) provides a way to correct measurements for the effects of by radar systems. It was not clear that the radar systems could refraction through the ionosphere. This is a significant effect at provide the needed measurement accuracy or coverage in the the GPS prime frequency, L1. Because the Trident II translator broad ocean test areas. Therefore, APL initiated a study that was not able to capture the full 20-MHz spectrum of the L2 signal due to system constraints on the signal bandwidth, the U.S. Navy and U.S. Air Force agreed to use an alternate frequency, L3 All-in-View GPS Satellites (L-Band Signals) at 1381.05 MHz, to compensate for ionospheric effects. The additional frequency transmits for the duration of a Trident test flight and is modulated with the narrowband ranging code, First IF and S-Band L-Band which allows the D5 translator to GPS Power Low Noise M1 M2 Signal Filter Amplifier Amplifier use dual-frequency signals (L1 and L3) even with the bandwidth limitations. Oscillator and Missle Satellite signals received at Frequency Translator Synthesizer the missile are amplified, shifted to an intermediate frequency, fil(S-Band Signals) tered to cover the satellite signal modulation bandwidth, shifted to the desired output frequency, and amplified for transmission APL to one or more ground stations. (Signal Data) Recording Telemetry Postprocessing Equipment Station The device that accomplishes Facility this function was called a translator to emphasize that the missile hardware: received the satellite signal Fig. 1. SATRACK measurement concept. Signals transmitted from the Global Positioning System satellites are translated it to a missile tereceived at the missile, translated to another frequency, and relayed to the telemetry station, where they are recorded for later playback and postprocessing at APL. lemetry frequency (S-band)

The designers named this program SATRACK to emphasize its capability to receive, rebroadcast, record, and track the satellite signals.

38

IEEE Instrumentation & Measurement Magazine

June 2003

 then rebroadcast the received signal (i.e., no signal pro-

cessing or tracking functions were provided by the missile hardware). The translator name also indicated that this signal relay device was for missile tracking and not for communications. From the translator perspective, only signal frequency and bandwidth matter. On the ground, signal recording equipment is a simple extension of the normal telemetry support function required for all flight tests. The translator output signal uses the missile telemetry band (S-band) and the same station-tracking antenna used for all the missile telemetry signals. The station: receives the translated GPS signals shifts the carrier frequency to near zero coherently samples the signal to capture the desired spectrum then digitally records the full signal data. The receiving equipment does not provide a SATRACK signal tracking function, only recording. Although not part of the accuracy function, some ground sites use the L1 C/A GPS signals to provide a real-time tracking solution for range safety.

Abbreviations and Acronyms

APL BMT C/A code DBT DGT EE ENTB ERIS FAR FST FTSS GPS GTP IEC IR&D LCTMK NMD-IFT PC P code PGE RAID Applied Physics Laboratory ballistic missile translator clear/acquisition code dual-band translator digital GPS translator enhanced effectiveness extended Navy test bed exoatmospheric reentry intercept subsystem flexible architecture receiver full-signal translator flight test support system global positioning system GPS translator processor Interstate Electronics Corporation independent research and development low-cost test missile kit national missile defense integrated flight test pilot carrier precision code portable ground equipment redundant array of inexpensive disks submarine launched ballistic missile translated GPS range system translator processing system wideband translator

Postflight Tracking and Data Processing

Figure 2 illustrates how the APL postflight tracking facility accomplishes precision tracking of the GPS signals through playback of the recorded translator signals. High-accuracy satellite ephemerides and clock estimates covering the span of SLBM the test flight are collected from the National Imagery and Mapping Agency. These data, along with processed telemetry TGRS data, help provide tracking aids for the postflight receiver and TPS measurement estimates for the missile processor. During playWBT back, satellite signals are tracked through delay-locked loops (DLLs) for range code modulation tracking and phase-locked Real Time loops (PLLs) for carrier-phase GPS GPS tracking. The translator passes Tracking Satellites Network signals for all satellites in view of the missile antenna, and the Test Missile postflight receiver provides all-in-view satellite-signal trackTracking Antenna ing. Currently it is common for a Trident missile to see 15 to 18 Preamplifier satellites during a test flight. After the signal-tracking data are recovered and several systematic corrections are applied, a large Kalman filter uses the derived satellite-to-missile (link) range and integrated Doppler data to provide estimates of parameters for the trajectory-observable model. APL used this terminology because not all model parameters produce observable signatures in
Range Safety System Translator Receiver SATRACK Recorder Telemetry Receiver Translator Recorder

Postflight
Orbit/Clock Determination Facility

Ephemerides and Clock Estimates

Tracker Corrections

Postflight Receiver

Missle Processor System Models

Telemetry Guidance Data Processor Estimated Initial Conditions

Fig. 2. Basic SATRACK configuration. For a number of days surrounding the missile flight, GPS signals are received, tracked, and recorded at the GPS tracking sites. During the missile flight, GPS signals are received by the missile, translated in frequency, and transmitted to the surface station(s). A tracking antenna at the station receives the missile signals, separates the various components, and records the data. The postflight process uses the recorded data to give satellite ephemerides and clock estimates, tracked signal data from the postflight receiver, and missile guidance sensor data. After the signal tracking data are corrected, all the data elements and the system models are used by the missile processor to produce the flight test data products.

June 2003

IEEE Instrumentation & Measurement Magazine

39

butions to impact miss. The method is based on projecting Uncertainty in each error contributor and its unInitial Conditions Estimate certainty into the impact domain. tation n First-level allocation is at the ie r O subsystem level: initial conVelocity ditions, guidance, and deInitial s ter Conditions e ployment and reentry. rom Second-level allocation protion ele i c s Crossrange Miss Po Ac vides data for major error Guidance groups within each subsysK C "Observed" Impact RA tem, e.g. accelerometers. T Gy A ro lS a Third-level allocation (not s t To Impact shown in Figure 3) produces Uncertainties estimates of fundamental Deploy Uncertainty in and Re ment error terms of the guidance e n try Guidance model, e.g., an acceleromeEstimate Estimated Total SATRACK ter scale factor error. SATRACK Impact Uncertainty In addition, SATRACK provides a point estimate of the initial conditions and a predeDownrange Miss ployment estimate; however, its principal purpose is to evaluate guidance subsystem error. The reFig. 3. Hypothetical example of SATRACK impact evaluation. Each error estimate provided by the SATRACK sults from each flight are checked process is projected into the impact domain, showing its downrange and crossrange contribution to impact miss for statistical consistency with in(the center of the coordinate system is the aim point). Uncertainties for each estimate are also projected. The estimated errors and their uncertainties are tested for statistical consistency with system models and other dependent measures of impact, independently measured results (e.g., the initial conditions derived from launch area instrumentation). initial conditions, and several other factors. any specific trajectory geometry. The processing methodolThe most significant development activity of SATRACK II ogy developed at APL properly combines a priori model data was the implementation of an evaluation capability for cumulative flight test accuracy. Limitations of the test geometry proand trajectory-observable model data for each flight test. Significantly, carrier-phase tracking of the GPS-to-missile hibit observations of all errors in any single flight test. signals provides the critical measurements. The measure- However, since each flight provides observations of the unments for GPS signal phase, i.e., integrated Doppler, sense derlying missile guidance error models, the data can be comrange changes along the line of sight for each signal to a small bined from many flight tests. The final cumulative analysis of fraction of a wavelength, usually a few millimeters. These flight test data produces a guidance error model for the Trimeasurements, which are compared with their values com- dent weapon system. It combines observations from each puted from guidance sensor data and satellite position and ve- flight to derive a missile guidance model that is both tactically locity estimates, provide most of the information. Noise in the representative and based completely on flight test data. This range measurement of the recovered GPS range-code signals model, combined with other similarly derived subsystem is of secondary importance. In essence, the inertial sensors models, helps develop planning factors used to assign provide high-frequency motion information better than the weapon system targets. signal processes, the Doppler information senses the systematic errors associated with the inertial sensors, and the range data provide an initial condition for all the dynamic measurements. Therefore, range noise, i.e., noise in the range tracking loops, can be smoothed over the full flight interval. The range noise remaining after this process is smaller than other bias-like uncertainties that set the limit on absolute position accuracy, e.g., the satellite position. Assuming that the data processing has not identified a system fault, which is an error component well outside its expected performance, the processed data from each flight test are used to provide estimates of major contributors to impact miss. Figure 3 shows a hypothetical diagram used to allocate contriThe requirements for missile position and velocity were initially set for the SATRACK II system at 20 ft (6 m) and 0.01 ft/s (3 mm/s), respectively. Currently, the system is providing velocity accuracy below 0.01 ft/s (3 mm/s) in all flight regions; the position accuracy is now less than 3 ft (1 m). GPS improvements with more satellites and more accurate ephemerides provided this performance gain.

Instrumentation
The instrumentation for the SATRACK system can be segmented into flight hardware (the translator and antenna assembly), field-support equipment (the data recorder), and the postflight data-processing facility (the SATRACK facility).

40

IEEE Instrumentation & Measurement Magazine

June 2003

The evolution of each instrumentation segment is interrelated and has evolved to support requirement changes. For example, moving from C4 operations to D5 operations required tighter accuracy requirements, thus necessitating the use of dual frequencies. Figure 4 describes the general categorization of GPS translators and corresponding generation of SATRACK postprocessing system. Detailed discussion of each segment follows, with emphasis on the current instrumentation at APL. A much greater depth of historical material is available in [1]-[6].

Flight Hardware (Translator)

GPS translators can be classified into two fundamental types, analog and digital. Reference [4] very elegantly discusses the pros and cons of analog versus digital translators. Analog translators heterodyne the L-band GPS signal to another frequency band, typically S-band, and add a pilot carrier to allow monitoring of the reference oscillator variations as well as the missile-to-ground dynamics. Narrowband analog translators limit the signal bandwidth to 2 MHz, which allows full tracking of the C/A code. Wideband analog translators have up to 20-MHz bandwidth for dual-frequency P-code tracking. Wideband translators typically overlay the two L-band frequencies to reduce total spectral bandwidth, but at the cost of increased noise in each signal. Digital translators down-convert the received L-band GPS signal to near base band and digitize it. The digitized GPS data is modulated onto a carrier, again typically at S-band, and transmitted to the ground recorder. Digital translators are more complex, but they allow the data to be encrypted as necessary. Narrowband Many types of translators have been in service since the inception of GPS-based missile guidance system evaluation. From the perspective of a postflight analyst, the key parameters for the translators are the spectral arrangement, bandwidth, and the L-band antenna design (although not directly related to the translator itself). All of these parameters have profound effects on the recorded GPS signal characteristics. Table 1 summarizes the general characteristics of the GPS translators and the low-cost test missile kit (LCTMK) translator that is currently in development.
Single Frequency

opment program. Because of C4 telemetry bandwidth restrictions and limitations in the recording technology, which precluded the use of wideband L2 P code, both Transat and C4 translators were designed to receive only the L1 C/A transmission from GPS satellites. Ground-based transmitters provided dual-frequency (L1 C/A and L1/4) signals for ionospheric corrections. Although this does not provide direct measurement of ionospheric refraction on the satellite links, it provided model information for corrections sufficient to satisfy the C4 accuracy requirements. The C4 translators were designed to support the GPS and ground-based transmitters. The missile employed four-patch GPS L1 antennas that summed in opposing pairs to provide 360 of coverage. The two summed pairs were multiplexed at 434 Hz, whose switching signal was amplitude-modulated onto the pilot signal to allow the receiver to demultiplex the antenna pairs. Trident II (D5) requirements could not be met without a dual-frequency GPS tracking capability. Since the recording technology at the time could not support wideband data for the L2 P code, the D5 translators utilized L1 and L3 C/A codes. The D5 translators employed wraparound array antenna for both L-band and S-band downlink that minimized phase variation through missile roll and eliminated the need for antenna multiplex switching. The ballistic missile translator (BMT) was similar in design to the D5 translator except for the lack of L3 as the second frequency. In addition to the C4, D5, and BMT translators, which have flown over 280 flight

GPS Translators Analog Wideband Digital

Dual Frequency

Transat/C4 (Trident I) BMT (NMD-IFT) D5 (Trident II) DBT (Peacekeeper)

SATRACK I

SATRACK II WBT (ENTB) FST (GT, EE) SATRACK III LCTMK* SATRACK IV* DGT (NMD-IFT)

The Transat translators were Trident I (C4) prototypes, and two * Under Development. units were flown together to proIn parenthesis are the primary programs supported by a given translation. vide redundancy for this critical element of the SATRACK devel- Fig. 4. GPS translator tree.

June 2003

IEEE Instrumentation & Measurement Magazine

41

Table 1. Evolution of GPS translators, as of December 2002.

Translator Transat Trident I (C4) Trident II (D5) BMT DBT WBT DGT FST LCTMK
a b a

Developer APL (1977) RCA/Cubic (1978) RCA (1985) IEC (1985) IEC (1989) APL (1994) IEC (1999) APL (2000) Lockheed/Hurley

Sponsor Navy Navy Navy RAJPO USAF Navy RAJPO Navy Navy

Frequency L1 and L1/4 L1 and L1/4 L1 and L3 L1 L1 and L2 L1 and L2 L1 and L2 L1 and L2 L1 and L2

Bandwidth (MHz) 2 and 0.2 2 and 0.2 2 and 2 2 2 and 2 20 and 20 Various 8 and 8
b c b b

Modulation C/A C/A C/A C/A C/A and P C/A and P C/A and P C/A and P C/A and P

Weight (lb) 13 13-16 20 3 4 2 1 1 n/a

No. Flown 2 154 100+ >30 2 4 5+ 8+ N/a

d

20 and 20

Digital unit with encryption capability for translated signal and telemetry data channel (to 10 Mb/s). Two signals are overlaid in a common downlink bandwidth. c Use of L2 signal optional; when used, it overlies L1. Three translator bandwidth choices with telemetry data channel; fourth choice is a 16-MHz translated signal only; encryption is optional for all choices. d Still in development. tests among them, several other narrowband analog translators have seen limited service. The translators that represent the latest technology are the digital GPS translator (DGT), full-signal translator (FST), and LCTMK translator that will replace the aging D5 translator in the near future. The DGT is a digital translator that has flown in several recent NMD-IFT missions. It is capable of a variety of configurations, including wideband L1/L2 dual frequency. The FST, shown in Figure 5, is functionally equivalent to the wideband translator (WBT), which had flown in reentry bodies for two earlier Trident missions. The FST is a wideband, dual-frequency (L1 and L2) translator with overlaid L1 and L2 spectrum with total spectral bandwidth of 20 MHz. It has flown in several GT (Glory Trip, Air Force missile flight tests) flights and Trident reentry bodies. For future GT flights, the FST will be used for range safety as well as postflight evaluation. Currently in development for the Navy is the LCTMK translator, whose specification defines it as an analog translator with L1/L2 overlaid channel with bandwidth of 8 MHz for accuracy evaluation and a separate 2 MHz L1 spectrum for the range safety function plus a pilot carrier. Like Trident I, a four-patch antenna will be used for the LCTMK. The wideband channel will be four-way multiplexed, and the narrowband range-safety channel will be two-way multiplexed for combined pairs. The switching signal will be amplitude-modulated on the pilot carrier.

Field Support Equipment (Recorders)

The ground recording equipment consists of the receiving antenna(s), data recorder, and auxiliary reference timing systems. The equipment receives the translated GPS signal along with other telemetry signals and distributes it to the data recorder. The GPS signal is down-converted to near base band, sampled, and recorded by the data recorders. Most telemetry stations are capable of generating a precise atomic timing standard. Table 2 summarizes the general characteristics of the recorders. The earlier equipment, flight test support system (FTSS) 1, 2 and 3, were narrowband recorders that relied on high-rate tape recorders, which provided up to 14 tracks of recording channels at four mega samples per second. The translator processing system (TPS) was developed for the national missile defense exoatmospheric reentry intercept subsystem (NMD-ERIS) program, where it served as a real-time GPS processor for range safety as well as a data recorder. The GPS translator processor (GTP), developed by IEC as a part of the Translated GPS range system (TGRS), also combined the data recorder with the real-time signal processing for the range safety function. It is capable of processing signals from both analog and digital translators. APL developed the portable

Fig. 5. Full-signal translator (FST) with LNA and prefilter.

42

IEEE Instrumentation & Measurement Magazine

June 2003

Table 2. Evolution of ground recording equipment, as of December 2002.

Recorder FTSS 1 FTSS 2 FTSS 3 TPS PGE GTP
a b b b

Developer IEC IEC IEC IEC APL IEC C4

Target Translator C4,D5 All analog BMT, DBT All analog All analog and digital

Media HRR HRR HRR RAID RAID RAID

Frequency Pilot, L1, L3 Pilot, L1, L3 Pilot, L1, L3 Pilot, L1, L2 Pilot, L1, L2, L3 Pilot, L1, L2, L3

Sample Rate 2 ms/s 4 ms/s 2 ms/s

a

Quantization 1b 1b 2b 2b

2 msp/s 4/20 ms/s Up to 16 d ms/s

c

1b Up to 4 b

FTSS3 GPS rate. Pilot was sampled 0.5 ms/s at 4-b quantization. FTSS3 and GTP are range safety systems. c Sample rate for GPS. Records Pilot, L1/L2 overlay @ 20 ms/s or Pilot, L1 and L3 @ 4 ms/s. d Samples entire spectrum.

ground equipment (PGE), shown in Figure 6, as the wideband recorder for the WBT, although it is compatible with other types of analog translators. GTP/PTS and PGE represent the latest in technology with RAID-based recording. PGE was designed for portability in a suitcase-size package that included a full complement of diagnostic features, such as a set of GPS trackers and a base-band GPS satellite/translator simulator. Although developed specifically to support the Navys ENTB program, it has since supported dozens of other missions for Trident, IFT, and GT programs.

SATRACK Facility
The postflight processing system consists of the SATRACK tracking facility, which processes the raw data into a time series of range and Doppler measurements for each satellite, and the Kalman filter, which incorporates various corrections and generates a navigation solution for the missile. The system has undergone several generations of redesign and development as requirements evolved with the new types of translators and recorders. As shown in the GPS translator tree in Figure 4, each successive generation of processing system was backward compatible with all known types of translators at the time. SATRACK I was developed for the C4 (Trident I) translator and laid the theoretical foundation for its successors. It was a pioneering work in many areas, with GPS still in its infancy and the techniques for postflight processing largely untested. Additionally, a separate simulation facility was developed to provide simulated data for all major elements of the processing system. SATRACK II went into service in 1987 for the D5 (Trident II) dual-frequency translators. It consisted of 56 tracking channels distributed among four VME chassis. This configuration was implemented to simultaneously process up to 12 dual-frequency satellites and a pilot carrier for both rightand left-hand circularly polarized signals. Utilizing the state-of-the-art components of the time, it was based on Motorola 68000 processors with much of the tracker hardware implemented in programmable logic devices (PLDs) and dis-

crete logic. The signal correlators were implemented in custom designed ASIC chips. Each tracker board was able to process one link at a time and, therefore, was called a single-channel tracker (SCT). During the late 1980s and most of the 1990s, SATRACK II was the main processing system for GPS translator-based Trident accuracy evaluation. SATRACK III represents the latest GPS postflight processing system at APL (Figure 7). It went into service in 1998 for processing the wideband L1/L2 dual-frequency P code as required by the wideband translators. The system hardware is based on VME chassis with Analog Devices SHARC DSP processors, and most of the custom GPS processing hardware is implemented in the latest available field-programmable gate array (FPGA). Each tracker board is populated with enough hardware to simultaneously handle up to eight tracking channels. Each VME chassis is typically populated with eight boards for simultaneous processing of 64 channels of data (the number of boards per chassis is limited only by the number of available slots). In a departure from previous user interfaces, it

Fig. 6. Portable ground equipment (PGE) with a controller PC

June 2003

IEEE Instrumentation & Measurement Magazine

43

is implemented in Windows PC workstations over an Ethernet local area network (LAN). This allows more streamlined and automated processing and logging, while improving the maintainability of the system software. SATRACK III represents a significant breakthrough in two areas. First, it is fully digitally implemented and can be expected to produce fully repeatable results, unlike its predecessors that relied on some level of analog circuitry such as analog PLLs for carrier-tracking loop or analog delay-lines for code-phase control. This eliminates the heuristic nature of the previous systems as well as the need for periodic hardware

calibration that accompanies analog circuits. Second, SATRACK III is based on the batch-mode processing concept, which allows the hardware to operate with software-like flexibility. Since the speed of the purely software system was prohibitively slow at the time (although it is continually improving), hardware that is fully configurable under software control implemented the most computing-intensive portions of the process, such as the signal correlation, generation of local code and carrier, and signal mixing. The batch-mode architecture provides the type of processing technique that was unavailable for the previous systems. As an example, it is possible to acquire the signal with virtually no acquisition delay by conducting extensive searches with the initial batch of data until all signals are found. This makes it possible to extract solutions from postplasma data in reentry body flights, where in some cases the data are available for only a fraction of a second. SATRACK III is compatible with all types of translators. It has been in service since 1998 and has processed data from Trident, NMD-IFT, and GT flight tests. The advent of the LCTMK necessitates the next generation of the processing system, which brings a new processing load into the requirement. The number of signals increases by at least four-fold because signals from each of the four antenna patches must be processed. The initial high-level design of the new system is currently under way. It will likely retain the batch-mode processing technique that has proven so successful in postflight processing but reduce reliance on embedded hardware processors and move closer to off-the-shelf, PC-based hardware that has advantages such as better software tools, easier path to upgrade, and lower cost.

Fig. 7. SATRACK III processing facility with a PGE in the foreground.

Flexibe Architecture Receiver (FAR)

Batch-mode processing has been applied to a stand-alone, real-time capable receiver called FAR (Figure 8). Although not directly applicable to SATRACK, it retains the essence of the batch-mode architecture, while maintaining the capability to process the data in real time. FAR is a single-channel L1 C/A-only receiver with a front-end data-storage memory that buffers up to 1 s of data. By running the correlation hardware at 20 times the sample rate of the 2-MHz data, it can track up to 16 satellites in real-time without any loss from channel multiplexing. FAR is capable of generating snapshot solutions from snippets of received data and, in a recent application, was able to reconstruct full GPS ephemerides from frequently interrupted signals.

Receivers Versus Translators

GPS receiver design has evolved over the lifetime of the SATRACK program due to the high-volume demand of the commercial market. This has driven down costs considerably, even on designs specifically geared towards the higher precision requirements and high-dynamic environment of missile guidance evaluation. Receivers could then be a potentially less expensive tool for test and evaluation. This naturally leads to the question, Why continue to use translators?

Fig. 8. Flexible architecture receiver. At left is the digital module with custom GPS hardware. At right is the RF downconverter unit.

44

IEEE Instrumentation & Measurement Magazine

June 2003

There are several limitations that contribute to the risk of using receivers instead of translators. Receivers have more complex hardware. Receivers need preflight initialization. Translator postflight tracking can adapt to unexpected test conditions. The translator configuration has the unique ability to reexamine the received GPS signals in the postflight environment and thereby adjust for the unexpected. The benefit from the translators tracking precision is also achieved through its postflight tracking capability. It is always possible to optimally adapt the postflight tracking bandwidths to the actual flight conditions, simply because the input signals can be replayed as often as needed. One benefit of a receiver configuration is associated with the output bandwidth requirement. A full GPS signal translator requires a 20-MHz bandwidth. In those applications where the capability of the evaluation system is dominated by phase-tracking performance, e.g., guidance evaluation applications, translator output bandwidth can be reduced to about 10 MHz with very little overall degradation. In contrast, a receiver can output tracking-loop data (range and phase) with a data rate of a few kilobytes per second, which is easily accommodated by standard missile telemetry. This is an important issue only if the translator output is within the standard missile telemetry band and if the band is fully occupied. The level of accuracy required for GPS-to-missile signal measurements and the acceptable level of risk relate to the accuracy evaluation of high-value weapon systems. No receiver can provide more precise measurements than a properly configured translator system, but the receivers may be adequate for some applications. The overall benefits of GPS translator systems will continue to make them the instrument of choice for high-value evaluations of precision weapon systems. The better instrumentation configuration will normally provide quantifiable performance assessments with fewer tests. Furthermore, since the national test ranges have made a significant investment in TGRS, there will be program cost benefits associated with using that system. Finally, with regard to precision interceptor testing, there is virtually no other instrumentation candidate that is anywhere close to meeting flight-test evaluation objectives. Reference [5] discusses this in detail.

levels of assurance in those capabilities will require similar skills and facilities. Continued D5 accuracy evaluation support will remain the primary objective for SATRACK with the current effort focused on modernization of the facilities and processing techniques to meet the requirements of the LCTMK. We will continue to support NMD-IFT and Minuteman flights, as well as pursue other precision intercept test opportunities.

References
[1] D.R. Coleman and L.S. Simkins, The fleet ballistic missile accuracy evaluation program, Johns Hopkins APL Tech. Dig., vol. 19, no. 4, pp. 393-397, 1998. [2] J.P. Gibson, Fleet ballistic missile test and evaluation, Johns Hopkins APL Tech. Dig., vol. 19, no. 4, pp. 388-393, 1998. [3] T. Thompson, L.J. Levy, and E. Westerfield, The SATRACK system: Development and applications, Johns Hopkins APL Tech. Dig., vol. 19, no. 4, pp. 436-446, 1998. [4] T. Thompson and E. Westerfield, Global positioning system translators for precision test and evaluation, Johns Hopkins APL Tech. Dig., vol. 19, no. 4, pp. 448-458, 1998. [5] T. Thompson, Demonstration of a precision missile intercept measurement technique, Johns Hopkins APL Tech. Dig., vol. 19, no. 4, pp. 513-523, 1998. [6] J.M. Watson, The origin of the APL Strategic Systems Department, Johns Hopkins APL TTech. Dig., vol. 19, no. 4, pp. 375-387, 1998.

Marc Camacho is an electrical engineer at the Johns Hopkins University Applied Physics Lab in the Strategic Systems Department. He has worked as an analyst for SATRACK program for 11 years, where he develops and maintains software to support the postflight evaluation and processing of SATRACK-derived measurements. He is currently involved in the modernization of the analysis and processing software to support current and future SATRACK efforts.

Accuracy Evaluation
SATRACK has been a significant contributor to the successful development and operational success of the Trident system. It continues to provide a unique monitoring function that is critical to the maintenance of the Navys strategic weapon system. As the United States moves toward a higher reliance on missile defense, the need to establish equivalent

Sung Lim received the B.S. and M.S. degrees in electrical from the University of Virginia. He joined the Strategic Systems Department of the Johns Hopkins University Applied Physics Lab in 1988, where he has been involved in development and operation of SATRACK processing facility. During SATRACK III, he was primarily responsible for the GPS tracking algorithm and hardware designs. He is currently working on the design of next-generation SATRACK processing system.

June 2003

IEEE Instrumentation & Measurement Magazine

45