The Bistatic Coherent Measurement System (BICOMS)
Georgia Tech Research Institute
Georgia Institute of Technology
Atlanta, Georgia 30332
Clay A. Blevins and Capt. Jason R. Girard
Radar Target Scattering Division
46th Test Group/TGR, Holloman AFB, NM 88330-7715
Introduction
The use of low-observable technology has significantly reduced the monostatic radar cross section (RCS) of aircraft, making their detection by conventional surveillance systems extremely challenging. The principal signature reduction technique employed in low-observable design is vehicle shaping. This technique relies on redirecting the incident radar energy in a direction away from the transmitting source, thereby reducing the monostatic signature of the vehicle. The laws of physics maintain that energy must be conserved. Hence, if the monostatic RCS is reduced by shaping, the incident energy must be distributed elsewhere such that the signature is increased at some or all bistatic angles. It is therefore conceivable that a bistatic radar will have better performance than a monostatic radar for detection of low-observable targets where shaping is the principal method used in RCS reduction.1 In addition, a number of radar-directed anti-aircraft missiles in use today operate in a bistatic mode during some or all of their engagement envelope. Thus, there must be concern for the bistatic signatures of even non-low-observable air vehicles. The Department of Defense (DoD) has therefore undertaken a program to upgrade and expand its bistatic RCS measurement capability at the U.S. Air Force 46th Test Group Radar Scattering Division RATSCAT at Holloman AFB, NM. BICOMS will provide the DoD with the capability to conduct monostatic and bistatic exploitation measurements of threat targets and evaluation of the radar signatures of blue targets.
System Description
2.1 Overview
The BICOMS system specifications reflect the requirement for a flexible, robust, highly accurate, and efficient RCS monostatic and bistatic measurement system. BICOMS is designed to operate in the gypsum flats of the outdoor RATSCAT Mainsite RCS ground-plane range, thereby requiring the BICOMS mobile radar unit (MRU) to be mobile in a loose, sandy gypsum surface which overlays a very hard and dense sub-soil. BICOMS is designed to operate over the frequency spectrum from 1-18 GHz and 34-36 GHz. At Mainsite, it will operate at ranges from 300 ft to 8,000 ft in a monostatic, a bistatic, or a simultaneous monostatic-bistatic mode. In the latter mode, the system is capable of collecting four independent sets of data, two monostatic and two bistatic, for both co-polarized and cross-polarized (complex) data during each target rotation. It also can simultaneously collect data from in-situ reference reflectors (or sources) for both the monostatic and bistatic scenario, providing up to 8 independent range gates per radar during data collection. Using polarization agile transmitters, dual channel receivers, and multiple range gates, BICOMS collects full polarization scattering matrix (PSM) data for both monostatic and bistatic modes. In the design and development of the BICOMS, emphasis has been placed on RCS accuracy, resolution, sensitivity, dynamic range, and stability, as well as operability, maintainability, and reliability. All of these elements are required to provide the range user with affordable, state-of-the-art, traceable RCS measurements. The BICOMS salient features are listed in Table 1.
40 ft span at 2,460 ft 80 ft span at 5,600 ft |
|||
1-18 & 34-36 GHz |
(complex) HH, HV, VH, VV |
||
(Note: MRU does not cover 1-2 GHz/L band or 34-36 GHz/Ka band*) |
2-4 GHz (S band) 4-8 GHz (C band) 8-12 GHz (X band) 12-18 GHz (Ku band) 34-36 GHz (Ka band) |
for a 20-MHz Bandwidth |
70 dB over 1-18 GHz 60 dB over 34-36 GHz |
4.5 kW from 34-36 GHz |
Automatic antenna positioning Automated field probing |
||
1-18 GHz coverage |
|||
The BICOMS system configuration is shown in Figure 1. BICOMS consists
of two functionally identical radar subsystems that are located in the
fixed radar site fixed radar unit (FRU) and housed in the mobile unit
(MRU). Both sites include instrumentation radar, antenna and support structure, control, communications, environmental and prime power monitoring and data storage and processing equipment. The MRU can be located up to 9,000 ft from the FRU and contains its own power system and environmentally controlled shelter. The MRU and the FRU are connected via a mobile fiber optic link (FOL), limiting the separation between the two units. During bistatic operation, the FRU provides timing and frequency control, as well as centralized data storage and processing. The FRU is located in Building 7000 at RATSCAT in the Data Processing and Control Center (DPCC).
(Click here for a larger version of figure 1.) |
The target under test is mounted at fixed target pit areas (TPAs) on low-RCS column supports or pylons and is rotated to provide full 360° aspect measurements. The rotation also provides the Doppler offset used to process high-resolution, cross-range data via the inverse synthetic aperture radar (ISAR) process. These data, along with the high-downrange-resolution data generated via the wide band stepped-frequency waveform, are processed to generate two-dimensional images with up to 17 GHz of bandwidth (0.3-inch resolution). Target control and position data are available at either site using the mobile FOLs from the MRU to the FRU and fixed in-ground FOLs from the FRU site to the TPAs. When field probing is conducted, the field probe is located at the TPA in front of the target area. Field probe control and RF signals are provided via the FOLs from either the MRU or the FRU as is the TPA control and feedback link. The field probe is further described in Section 2.4.
A wireframe design drawing of the MRU is shown in Figure 2 and a photo of the MRU at RATSCAT is shown in Figure 3. Figure 4 shows BICOMS set up for initial testing at a 1.5 ° bistatic angle.Referring to Figure 2, the MRU consists of several major structures: the transporter with its antenna support structure (super structure), the antennas and antenna positioners, the RF enclosures (located on the antenna positioners), the emergency generator, the heating, ventilation, and air conditioning (HVAC), and the equipment-personnel shelter.
(Click here for a larger version of figure 2.) |
The shelter houses the MRU radar subsystem, the auxiliary subsystem, intercom, FOL transmitter and receivers, the antenna controllers, the uninterruptible power supply (UPS), and a test crew of 4 to 6 people. Not shown are the primary power generator and mobile fiber optic cable dispenser which are deployed with the MRU.
 |
The MRU provides frequency coverage over 2-18 GHz in 4 RF bands (S, C, X, and Ku) using 7 antennas. Note in Figures 2 and 3 that the mechanical structures are already in place for the addition of the 34-36 GHz band (ie., antenna positioner and RF enclosure). Also, note that RF bands C, X, and Ku are split into sub-bands with separate antennas used to meet system sensitivity and antenna amplitude taper specifications. Each of the lower frequency sub-bands (C1, X1, and Ku1) is on a separate delta height control, allowing independent optimization of the ground bounce for all RF sub-bands. All MRU antennas have independently controlled azimuth and elevation positioners as well as height control.2 Figure 5 is a photo of the C-band antenna positioner with an 8 ft dish covering 4 to 5.66 GHz (C1) and a 6 ft dish covering 5.66 to 8 GHz (C2). Note also the RF enclosure which contains the high power transmitter, the receiver low noise front end, and the positioner controls.
The FRU provides coverage over 1-18 GHz and 34-36 GHz. As with the MRU, split bands are required for C, X, and Ku. The government furnished FRU antenna positioners are less flexible with split bands on the same azimuth and elevation positioner and having manual delta height control. In both the MRU and the FRU antenna positioners, automatic antenna peaking is available using the radar control computer.
 |
2.2 Instrumentation Radar(s)
The BICOMS radar subsystems are extensively modified Lintek élanTM radars. The élanTM is a wideband, pulsed, stepped-frequency system, with high-speed switching of all radar parameters. All radar parameters are agile on a pulse-to-pulse basis, including frequency, polarization, phase, range, and coherent integration. The software system for the radar is a highly flexible Windows NT application, well suited to this environment. The radar system is highly re-configurable; the core radar system consists of the digital subsystem, power supplies, RF/IF system, local oscillator (LO), A/D converters, fiber optic modems, and computer system. Figure 6 shows the MRU radar installation consisting of the two racks on the left along with the antenna control rack on the far right. At the bottom of the control rack is the fiber optic termination box, which connects the MRU to the FRU via the mobile FOL. A simplified block diagram of the MRU radar, along with the common (MRU/FRU) timing and control and coherent sources located at the DPCC, is shown in Figure 7. The FRU radar, located in the DPCC along with the coherent sources and common timing and control is not shown, but is functionally the same as the MRU radar.
 |
 |
Referring to Figure 7, the coherent sources located at the DPCC consist of a fixed 3-GHz IF and a stepped 7-15 GHz LO which, in the bistatic mode, are shared between the MRU and the FRU with the sources linked to the MRU via microwave fiber optics. In the MRU independent, stand-alone mode of operation, the MRU uses local coherent sources and local timing and control. For independent FRU operation, the DPCC coherent sources are used in a dedicated monostatic mode. All frequency sources, as well as the timing, are phase locked to an ultra-stable 100-MHz reference oscillator. The LO is generated using an ultra-fast switching synthesizer, a Comstron 5000. In the IF transmit and receive chassis and the RF conversion chassis, the IF signal is phase encoded and mixed with the LO signal to generate the desired RF (except for Ka band, which uses an additional RF conversion in the RF enclosure to up-convert 9-11 GHz to 34-36 GHz) The signal is then filtered in a high-speed switched filter bank, amplified, and sent to the selected RF remote enclosure which is located on the antenna positioner tens of feet from the RF source. In the RF enclosure the signal is further amplified by the high-power TWTA and routed to the selected polarization port (H or V) of the selected antenna (upper band or lower band for the split bands).
On receive, both co-polarized (HH, VV) and cross-polarized (HV, VH) signals are received and processed using dual-channel receivers. The received signals are immediately amplified in the RF remote enclosure using low-noise amplifiers (LNAs) and routed back to the RF conversion and IF chassis where they are down converted to IF, amplified and filtered, and finally down converted to in-phase (I) and quadrature phase (Q) baseband video. (For Ka band, the 34-36 GHz signal is down converted to 9-11 GHz in the RF remote enclosure). These signals are then sent to the digital signal processor (DSP) where they are digitized (32 bits), phase unwrapped, and integrated prior to storage.
 [fig7b.html - MISSING] |
(Click here for a larger version of Figure 7.)[fig7b.html - MISSING] |
The band switching (e.g., S to C band), polarization switching (e.g., H to V), and antenna switching (sub bands, e.g., C1 to C2) is accomplished at the pulse repetition frequency (PRF), providing the capability to sweep continuously from 1-18 GHz and 34-36 GHz at the maximum pulse repetition frequency (PRF). The radar waveform is programmed to interleave the sweeps across the different sub bands, providing optimum sampling of the target as a function of frequency. Additionally, the amount of coherent integration is selectable on a sub-band basis to provide for sensitivity optimization during a single target rotation.
2.3 Fiber optic link (FOL)
Instrumentation grade, coherent, bistatic, RCS measurement systems require a reliable low-noise method to link the coherent reference, LO, and IF signals between the widely spaced transmit and receive subsystems. Primary considerations in the selection of the method used to distribute the frequency source(s) were added phase and thermal noise, dynamic range, propagation effects, cost, and electromagnetic interference (EMI). Based on these considerations, the FOL was selected as the best overall approach for the BICOMS3. Issues of added phase noise and link stability were investigated and, as described and reported in Reference 3, the link-added phase noise has essentially no effect on the system performance where the LO and IF are linked (see Figures 8 and 9).
 |
 |
The FOL performance, however, is sensitive to temperature; that sensitivity manifests itself as a change in physical length and a change in propagation velocity through the fiber. This results in a phase change with temperature as given by
where q is the phase, T is the temperature change in degrees Celsius, lRF is the wavelength of the RF, and L is the length of the fiber optic cable.
In the BICOMS design, the frequency sources are distributed to two radars separated by 9,000 ft with a frequency range on the fiber from 3 GHz to 15 GHz. As an example, for a temperature change of 1°C and a frequency of 15 GHz, the phase change on the FOL will be 9.64 radians, or 553°. Thus a small change in temperature of the FOL will result in significant change in the phase, i.e., 0.1°C will result in a 55° phase change at 15 GHz. Since the time required to develop an image can be on the order of tens of seconds, variations in phase may occur as the fiber is heated and cooled by the environment. Although the majority of the testing will occur during stable periods, a design was developed to monitor the phase of the RF on the fiber as part of the traceability and accuracy of the system requirements. The design as implemented is shown in Figure 10 and uses the fixed 3-GHz signal as the phase reference source. The IF signal received at the MRU goes through a 3-dB optical coupler, and a sample of the IF is sent directly back to the DPCC on a dedicated fiber located in the same cable jacket as the source fiber. At the DPCC, the signals are converted to RF and are compared in an IQ detector. The IQ detector output is a DC signal proportional to the phase difference between the signals. Following the IQ detector, the signal is digitized and stored, and samples of the data are made available for monitoring on a real-time display with adjustable alarm thresholds for delta phase. The delta phase signal is continuously monitored, as well as recorded, at the waveform rate for deviations from this baseline.
The absolute phase is not of interest, as the primary impact is caused by a change in the phase during a coherent processing interval (CPI). The recorded data can be used to determine stable test periods or can be used in off-line processing to correct for phase variations with time. Note that the 3-GHz phase is measured over twice the fiber length (out and back) and represents the phase change for a 6-GHz RF.
2.4 Automated Field Probe System
The automated field probe system (AFPS) shown in Figure 11 was designed to minimize cost while providing a mobile capability to rapidly, smoothly, and accurately probe the field incident upon the various TPAs available at the RATSCAT facility. The AFPS provides the ability to probe in a horizontal raster over an area as large as 40-ft x 40-ft and to perform selective linear probing in the horizontal or vertical directions, all under computer control from the FRU or the MRU.
 |
 |
Raster scan is accomplished via vertical steps coupled with horizontal continuous scan. The total scan time required for a 40-ft x 40-ft raster with 1 foot spatial resolution is less than 30 minutes. For this calculation a scan consists of 27 different frequencies (3 per sub-band), 2 polarizations at each frequency, and 32 coherent integrations per pulse. Commercial off-the-shelf (COTS) equipment and components were used to the maximum extent possible in the development of the AFPS. The AFPS consists of a COTS, self-propelled scissors lift (JLG 500RTS) adapted for the field-probe application by the addition of a horizontally mounted 48-ft aluminum rectangular beam truss with a cable-driven probe carriage2 as depicted in Figure 11. Note in the figure, that the RF probe carriage extends down from the horizontal boom to allow probing down to ground level. Mounted on the probe carriage, and shown in Figure 12, are three dual-polarized (V and H) probe-horn antennas covering the 1-2 GHz, 2-18 GHz, and 34-36 GHz bands.
The AFPS operates as a one-way probe. The RF, phase reference, and control signals from the radar are transmitted to the AFPS over a microwave FOL from either the MRU or the FRU. The probe waveform as well as position and scan are remotely selectable from either radar control unit. In operation, the AFPS is driven under its own power to the TPA, the boom is positioned perpendicular to the line of sight of the BICOMS radar antennas (either MRU or FRU), and the entire system is leveled by means of the scissors lift's hydraulic leveling jacks. Once leveled, the system is interconnected to the BICOMS via a single-mode FOL.
As shown in Figure 13, the interface between the system control computer (SCC) at either the FRU or the MRU and the AFPS is provided by the field probe control computer (FPCC) which is located in an air-conditioned electronics enclosure on the scissors-lift platform. Under control of the FPCC, the RF signals are reconverted to microwave or millimeter wave frequencies as appropriate and routed to the correct horn for transmission back to the radar via an RF switching network situated in a cooled electronics enclosure mounted on the probe carriage. The spatial position of the probe is also under control of the FPCC via interface to an Indramat servo drive system (for horizontal positioning) and to the scissors lift hydraulic control system (for vertical positioning). The probe's horizontal and vertical position, transmit frequency, and probe polarization are transmitted back to the DPCC via the TPA FOL for data acquisition and processing. Signal coherence for the 34-36 GHz signal (up-converted from 9-11 GHz in the field probe electronics enclosure) is maintained by the use of a 100-MHz reference signal sent from the DPCC.
Selected field probe data can be displayed at the radar control computer as a function of frequency and polarization. Data>
zontal or vertical cuts) or as a 3-D plot of the entire raster scan for selected frequencies and polarizations. As with all quick look data processing, up to 12 plots can be generated simultaneously.
 |
 |
2.5 System Calibration Approach
A key task in the development and validation of any RCS measurement system is the ability to provide repeatable, accurate, and traceable measurements. These include absolute RCS as well as accurate high resolution imaging of the target under test. This is usually accomplished by applying amplitude and phase calibration correction factors to the data. For polarimetric radars, additional processing may be required to improve the polarization isolation and thus the quality for the co-polarized and cross-polarized data. For BICOMS both calibration processes will be used; amplitude and phase for the quick look data and full polarization scattering matrix (PSM) calibration for the off-line processed data. Even with excellent cross polarization isolation, the later PSM calibration process is necessary to calibrate bistatic cross-polarization data using the approach selected for BICOMS.
Full polarization calibration used in the BICOMS process results in the separation of the correction factors derived for each of the two transmitters and receivers using the approach described in references 4 and 5. The bistatic calibration is a simple extension of this approach and takes advantage of the BICOMS dual monostatic and bistatic capabilities.
In the development of the approach, major consideration was given to the broad frequency coverage, the time required to calibrate, and various site limitations. To minimize target handling and thus calibration time, the resulting approach uses only two calibration targets to cover the entire band; a horizontal cylinder and a 90 ° square dihedral. In each case the targets are rotated about one of two axes to achieve the desired response. In the case of the cylinder, it is used for both monostatic and bistatic co-polarization data collection. For the monostatic case, measurements of the end cap as well as broadside are used. This is necessary since the co-polarized responses for the broadside cylinder are not equal for 1-4 GHz. Greater than 4 GHz, the broadside response is adequate. The dihedral is used for monostatic only and provides two of the three inputs to the full PSM calibration (the other being the cylinder). The dihedral is measured with the seam horizontal and at a 22.5° tilt about the horizontal.
Initial bistatic testing at a 1.5° bistatic angle began in September 1998 and is scheduled for completion in April 1999. A second bistatic angle of 62.5° will be used for final validation of bistatic calibration. These tests are scheduled for June 1999. For the later tests, an obtuse (120°) square dihedral will be used to validate the bistatic cross-polarization calibration procedures. To support these tests and future testing, a series of precision calibration reflectors were fabricated consisting of a 90° dihedral, 120 ° dihedral, and a cylinder. Theoretical calibration tables have been generated for each reflector to be used in the calibration process. In generating these tables, a number of RCS prediction techniques have been evaluated and used including Method of Moments (MoM), Finite Difference Time Domain (FDTD), Physical Optics (PO), and Method of Equivalent Currents (MEC). This was required to maintain a calibration error of less than 0.25dB.
3. Summary
The Bistatic Coherent Measurement System (BICOMS) is designed to make highly accurate nearfield and far-field monostatic and bistatic RCS measurements. It will provide state-of-the-art coherent, full polarimetric scattering matrix radar signature data, both narrowband (total RCS) and wideband (one- and two-dimensional), and offers a high degree of efficiency and flexibility. BICOMS will provide an order of magnitude reduction in test setup and data-collection time while providing higher quality and broader test capabilities.
Operational capability is scheduled for May 1999. The BICOMS is part of the five-project Advanced Static RCS Measurement Program, which was established to correct known test capability shortfalls, to meet new technology requirements for measuring Very Low Observable platforms, to effectively reduce operational costs, and to improve the quality of critical RCS data for tactical mission planning.
4. Acknowledgements
This work has been funded by the USAF, 46th Test Group, RATSCAT Division, Holloman AFB, NM as part of the RATSCAT Advanced Static RCS Measurement Program.
The authors would like to recognize the contributions of the following GTRI staff: David Asbell (antenna positioner and field probe), Mike Brinkmann (lead digital and software engineer), Henry Cotten (lead mechanical engineer), Lacey Moore (lead RF engineer), Tim Smith (MRU transporter), Mike Tuley (calibration) and George Whitley (system integration). In addition, the authors would like to thank Johnson Controls World Services, RATSCAT Operations, for their installation support. We would like to particularly recognize Tim Espinoza, John Hennessy, David Dedmone, George Burnell, and Marvin Baker for their technical support during the development and installation of the BICOMS.
References
1.) L. Ning-jing, "Radar ECCM's New Area: Antistealth and Anti-ARM," IEEE Trans. on Aerospace and Electronic Systems, Vol. 31, pp. 1120-1127, July 1995.
2.) O. David Asbell and J. Mark Hudgens, "BICOMS Antenna Positioner System (APS) and Automated Field Probe (AFP)," Proceedings of the 1998 Antenna Measurements Techniques Association (AMTA) Symposium, Montreal, Quebec, Canada, October 1998.
3.) T. L. Lane and J. A. Scheer, "Fiber Optic Link Phase Noise and Drift Effects on Bistatic Imaging Radar Performance," Proceedings of the National Aerospace and Electronics Conference (NAECON), pp. 648-654, May 1996.
4.) Whitt, M., and F. Ulaby, "A Polarimetric Radar Calibration Technique with Insensitivity to Target Orientation," Radio Science, Vol. 25, No. 6, Nov/Dec 1990, pp. 1137-1143.
5.) M. W. Whitt, et al., "A General Polarimetric Radar Calibration Technique," IEEE Trans. on Ant. and Prop., Vol. 39, No 1, Jan 1991, pp. 62-67.