Electronically steerable plasma mirror-based radar: concept and characteristics.
A plasma mirror performs like a metal mirror, but it is inertialess. Compared to high-performance phased-array systems, a plasma-mirror-based radar system is much simpler and is therefore more affordable. Electronic steering of microwave beams using a plasma mirror permits the use of wide instantaneous bandwidth waveforms. Potential areas of application for a plasma-mirror-based antenna system include ship self-defense, high-resolution radar imaging, target identification, electronic countermeasures, high-data-rate communications, spread-spectrum links and remote sensing.
As a reflector, the plasma mirror exhibits extremely low loss and the reflectivity is very nearly 100%. Since a perfectly reflecting object cannot radiate, the noise temperature contribution of the plasma mirror to the antenna temperature is likely to be small. The plasma sheet can be steered in elevation by tilting the magnetic field, and steering in azimuth may be accomplished by designating cathode initiation sites. Switching times between successive mirror orientations may be less than 20 [micro]sec.
Phased-array antennas are extremely versatile microwave beam radiators for operation over octave bandwidths. The capability of rapidly and accurately switching beams enables multiple functions to be performed simultaneously. The transmit and receive beams may be independently controlled and optimized, permitting adaptive pulling, low-sidelobe receive beams and accurate target-angle determination. However, high-performance phased arrays are complex and expensive. There are, in addition, areas of application where phased arrays are not the most appropriate choice.
One of the drawbacks of phased arrays is that almost all of them achieve beam steering using modulo 2[Pi] phase shifters. This makes the beam direction frequency dependent, which limits the instantaneous bandwidth of the array. This limitation can be overcome by employing time-delay steering, which uses switched delay lines at each element. However, the line lengths are large for beam-steering angles [greater than] 5 [degrees], which makes this approach impractical. Another limitation of phased arrays is the requirement for the radiating elements to be spaced by [approximately][Lambda]/2 to avoid grating lobes. This makes the number of elements in the array very large if operation over three to four octaves in frequency is desired, since the element spacing is determined by the highest operating frequency.
With the increased availability of wide-bandwidth microwave components, high-speed digital processors and digitally controlled frequency sources, high-resolution radar imaging is becoming increasingly attractive. Although the fractional bandwidth of high-resolution waveforms (e.g., chirp pulses) may be small, the absolute bandwidth may range from 10 MHz to 1 GHz. Phased arrays are not appropriate for these large instantaneous bandwidth waveforms, especially for bandwidths exceeding 200 MHz. High-data-rate communications and spread-spectrum techniques also require large bandwidths.
One way of accomplishing electronic steering with large instantaneous bandwidth signals is to decouple the beam-forming and beam-steering functions of the antenna. An inertialess plasma mirror[2, 3] may be used in the near field of a wideband offset-fed reflector (beamformer) to enable electronic beam steering. The design of the reflector could incorporate low-sidelobe monopulse beams with lowcross-polarization features, and the addition of the plasma mirror could provide agility to these beams. Such a system would permit wideband operation, and it could be less expensive compared to a conventional phased-array radar system.
A low-loss plasma reflector requires discharge gas pressures less than 300 mTorr. The plasma mirror is formed[2, 4] by driving a glow discharge between a U-shaped linear hollow cathode and a flat plate brass anode which are immersed in an axial magnetic field of strength 15-25 mTesla. The discharge operates in the negative glow regime at neutral gas pressures in the 40- to 50-mTorr range. Negative glow plasmas yield higher electron densities than positive column plasmas. The plasma mirror may be steered in elevation by tilting the magnetic field, since the orientation of the plasma sheet is determined by the direction of the magnetic field lines. Steering in azimuth may be accomplished by designating cathode initiation sites or by switching between a set of linear cathodes. The plasma mirror may be made agile by forming the mirror in one location, turning it off for a short duration and then reforming it in another orientation. The decay of plasma density is governed by ion-electron recombination. Since these recombination times are typically less than 10 [micro]sec, the switching time between successive mirror orientations may also be less than 10 [micro]sec if the volume magnetic field can be changed on these time scales.
The refractive index of a plasma is given by [Mu] = [(1-[[w.sup.2].sub.p]/[w.sup.2]).sup.1/2], where w is the radian frequency of the electromagnetic wave, and the plasma frequency [w.sub.p] is related to the electron density n, in [cm.sup.-3], by [w.sub.p] = 5.64 x [10.sup.4] [square root of [n.sub.e]]. Since [[w.sup.2].sub.p]/[w.sup.2] is always positive, the refractive index of a plasma is always less than 1. This is unlike glass and plastics which have refractive indices greater than 1.
When an electromagnetic wave is incident on a plasma surface [ILLUSTRATION FOR FIGURE 1 OMITTED], the wave is entering a medium of lower refractive index. Thus, a reflection from a plasma surface is similar to total internal reflection in optics. Since a plasma reflects all microwave frequencies below the plasma frequency, the electron density at the critical surface where reflection occurs is given by [n.sub.e] = 1.24 x [10.sup.10] [f.sup.2] [cos.sup.2] [Theta], where [Theta] is the angle of incidence (measured from the surface normal) and f is the frequency of the incident microwave beam in gigahertz. The highest density is needed for normal incidence.
The plasma mirror is not a blackbody radiator. A detailed analysis takes into consideration the poor reflecting characteristics of the under-dense plasma edge region in front of the critical surface. This analysis gives a 250 [degrees] K radiating temperature. Thus the noise temperature contribution of the plasma mirror to the antenna temperature is likely to be small. The microwave power handling capability of the plasma mirror is determined by using the criterion that the microwave pulse should not by itself cause appreciable ionization of the discharge gas. The allowable power densities are then a function of the microwave pulse length. The estimated CW capability of the plasma mirror is [approximately] 100 W/[cm.sup.2]. For 1-[micro]sec-wide microwave pulses at a few kilohertz repetition rate, the allowable power density is 3 kW/[cm.sup.2] and for 10-nsec pulses the power density is 200 kW/[cm.sup.2] at X band. Thus, the peak power-handling capability of a plasma mirror placed in front of a 2-m-diameter X-band dish for 1-[micro]sec-wide pulses is 90 MW. The allowable power densities tend to increase with the frequency of the microwave beam.
Agile Mirror experiments[2, 4] at the Naval Research Laboratory have demonstrated that a planar 50 cm x 60 cm x 1 cm plasma mirror can be formed with densities high enough to reflect X-band microwave beams. The plasma discharge is formed inside a 1.2-cm thick Lexan vacuum chamber. The discharge is immersed in a magnetic field. The driver for the plasma mirror applies negative high-voltage pulses to the cathode. For typical operation at 120 mTorr (air), the discharge current is 15 A and the voltage across the plasma discharge is 4.6 kV. In a negative glow plasma, the energetic electrons emitted by the cathode propagate toward the anode, ionizing the neutral gas atoms along the way. The pulse duration of these negative glow discharges is 300 [micro]sec, and the repetition rate is 0.5 Hz.
A 30-cm-diameter X-band dish illuminates the plasma mirror at 10 GHz. The dish employs a rectangular wave-guide Cutler feed, and the radiated beam is vertically polarized. The reflected microwave signals (for a 90 [degrees] reflection) are measured using a horn located 4.3 m from the plasma mirror. Space restrictions prevented measurements at larger distances. The characteristics of the dish-plasma mirror configuration are determined by measuring the radiation pattern in the transmit mode. It is convenient to measure this pattern by rotating the plasma mirror while keeping the receiver horn stationary a fixed distance from the mirror. Figure 2 shows a comparison of the H-plane radiation patterns of the vertically polarized 10-GHz beam reflected by the plasma mirror and by a 60-cm x 60-cm metal mirror substituted in its place.[2,3] The slight differences between the two patterns may be attributed to the difference in the widths of the metal and plasma mirrors. The sidelobe levels are roughly those measured for the 30-cm-diameter dish in a compact range. The two patterns in Figure 2 use the same normalization factor for the received power; this means that the reflectivity of the plasma mirror is very close to 100%.
A ship-based radar system using a plasma mirror would operate in the X band, since lower frequencies would require much larger mirrors and radomes. X-band operation precludes long-range search missions. Therefore, the functions assigned to a multifunction ship-based X-band radar may include horizon search, detection and tracking of low-altitude targets and sea-skimmer missiles, target illumination for terminal missile guidance, medium-range tracking of all targets, target identification, fire-control operations, ship navigation and communications. It may be possible to use such a surveillance radar for medium-range volume search, but the critical issue here is the budgeting of radar time. Proper choice of radar waveforms, which may include 1,000:1 pulse compression waveforms, is essential if volume search functions are included.
Figure 3 shows a conceptual sketch (drawn to scale) of a plasma-mirror-based radar configuration. There are no moving parts in this system. The effective aperture area of the illuminating reflector dish may be 2 [m.sup.2]. Shown in Figure 3 are the three mirror positions for the -5 [degrees], +60 [degrees] and +90 [degrees] elevation beams. It is assumed here that for elevation angles greater than 60 [degrees], the dish feed-horn array will be switched to illuminate only the central portion of the dish. This avoids beam blockage at the cathode. Figure 3 shows the cathode dotted because in this view only the narrow "waist" appears. The beam is steered in azimuth by switching appropriate cathode initiation points in the array. A combination of a much smaller central cathode array with a set of radial linear cathodes may also be used in place of the large cathode array. A +45 [degrees] azimuthal angle change in the cathode orientation results in a +90 [degrees] change in the azimuthal angle of the reflected beam.
The anode is a flat metal plate. The coils AA, BB and CC generate the dc vertical magnetic field. The coil DD generates a programmable (time varying) horizontal magnetic field. The addition of this field tilts the direction of the net magnetic field, which in turn steers the plasma sheet in elevation. The one-sided configuration shown in Figure 3 provides 180 [degrees] azimuthal coverage with no blockages. By adding another reflector dish diametrically opposite in the same horizontal plane, the azimuthal coverage may be extended to 360 [degrees]. In this case, another coil may be added to complement DD. The blockage from this coil will then have to be [less than] 5% to maintain low sidelobes.
Although Figure 3 appears to show a center-fed parabolic dish, offset designs may provide greatly improved performance. Offset feeding, together with the use of more advanced feeds (multimode, corrugated and array feeds), has proven to be a very effective way of achieving low and ultra-low sidelobe reflector antennas.[5,6] Measured far sidelobe levels are down by [greater than] 60 dB for an optimized precision reflector antenna. Poor cross-polarization discrimination is one of the problems of offset-fed reflector designs. Multimode feeds may be used to improve the cross polarization performance, but this is inherently a narrow-band solution. Dual-offset-reflector configurations may instead be used, since these reflector geometries yield superior performance over large bandwidths. Synthesis techniques may be applied to these configurations to achieve not only low sidelobes, but also low cross-polarization and high aperture efficiencies.
A plasma mirror may also be used in front of the feed horns in limited scan antenna systems.[5,9] The radiation from the feed horns is redirected by the plasma mirror across the surface of an offset hyperbolic reflector surface, which scans the main beam generated by the reflector. The advantage here is that a much smaller plasma mirror may be used to steer beams at the lower microwave frequencies (below X band). Such a system could be used in applications where a single, limited scan antenna system is desired that can operate (possibly with narrow bandwidths) over three to four octaves in frequency. The X-band AN/TPN-25 precision approach radar is a limited scan antenna system which uses a small phased array to illuminate a reflector having an offset hyperbolic surface. This radar has a limited coverage sector: 15 [degrees] in elevation by 20 [degrees] in azimuth.
The radar system shown in Figure 3 is much simpler than other high-performance phased-array radar systems and is therefore more affordable. However, it is not very compact. Affordable high-power phased-array radar systems are presently being developed,[11,12] but they have the instantaneous bandwidth limitations that are common to all phased arrays.
COMPARISON WITH PHASED ARRAYS
Only a general comparison of the plasma-mirror-based antenna system with a phased-array system can be made, since phased arrays can take different forms depending upon the application. Similarly, a plasma-mirror-based system designed for one function may be somewhat different from that built for another application.
A rule of thumb for the instantaneous bandwidth of phased arrays is:
Bandwidth (%) = Beamwidth (deg).
For example, an L-band phased array generating a 1 [degree] beam will have a 15-MHz instantaneous bandwidth at 1.5 GHz. However, substantially larger instantaneous bandwidths can be achieved by partitioning the array into smaller subarrays and delaying the signal to each subarray by the amount necessary to make the signals from all subarrays arrive simultaneously at the target. This process is called "time delay steering." Phase steering is used over each subarray. The Cobra Dane (AN/FPS-108) L-band radar achieves 200-MHz instantaneous bandwidth using this technique.
In comparison, a plasma-mirror-based antenna system provides true time delay steering, since all ray path lengths are equal for each mirror orientation. Thus, the instantaneous bandwidth can be very large. In practice, however, the bandwidth is limited by the feed horns and the waveguides used before these feed horns. With the increased availability of wide-bandwidth microwave components, instantaneous bandwidths exceeding an octave may be attainable.
Optical beamforming techniques may provide extremely wideband (instantaneous bandwidth) distribution networks that are lightweight and easily adaptable to conformal surfaces. Here, microwave signals that are modulated onto an optical carrier can either be delayed or given a relative phase shift which, upon detection, maintains their desired responses. The desired time delays are produced using switchable optical-fiber lengths or dispersive optical fibers. The most elegant of these is the fiber-optic prism antenna feed, which uses numerous optical-fiber links having varying lengths of dispersive and nondispersive fibers to feed the antenna elements. By changing the frequency of the tunable wavelength laser, the relative time delay between the elements and hence the microwave beam-steering angle can be altered. Since the time delay ramp needed on receive is the opposite of that used on transmit, the laser frequency needs to be changed to switch between the transmit and receive modes. A high-performance two-dimensional fiber-optic beamformer featuring low sidelobes is likely to be complex and expensive. In spite of these limitations, optical beamforming may be an attractive option for large instantaneous bandwidth systems, particularly at the lower microwave frequencies (L band and lower).
The agility bandwidth of phased arrays is limited primarily by aperture matching and the requirement for the elements to be spaced by less than [Lambda]/2 to avoid grating lobes. If the array is required to operate over three to four octaves in frequency with substantial gain, then the number of elements in the array becomes very large, since the element spacing is determined by the highest operating frequency. This makes the array complex and expensive. In contrast, the plasma mirror reflects all microwave frequencies below the plasma frequency. If the plasma density is high enough to reflect 20-GHz microwaves, then the same plasma mirror can be used for all frequencies less than 20 GHz. The only limitation here is that the finite size of the mirror imposes a lower frequency limit below which performance degrades because the mirror size is only a small multiple of the microwave wavelength. The limited scan system described earlier is most suitable for wideband operation (over three to four octaves). By placing the plasma mirror in front of a feed horn, operation at frequencies below X band is possible, since a much smaller plasma mirror is required.
Beam Switching Speed
Ferrite-phase-shifter-based phased arrays can have switching times [less than or equal to] 10 [micro]sec. Nonreciprocal ferrite phase shifters additionally have to switch phases between transmit and receive, which makes operation in the high-PRF pulse-Doppler mode difficult. GaAs T/R-module-based phased arrays can be designed to be reciprocal and they can have much faster switching times. The agility of the plasma mirror, in comparison, is mainly determined by the speed with which the magnetic field can be changed and by the recombination time, which is typically [less than or equal to] 10 [micro]sec. For plasma mirrors larger than 1 [m.sup.2], the switching time between successive mirror orientations may be [less than or equal to] 20 [micro]sec. Mirrors smaller than 1 [m.sup.2] may be reoriented in less than 10 [micro]sec. Since a plasma-mirror-based antenna system is reciprocal, operation in the medium-PRF and high-PRF pulse-Doppler modes is possible.
Target angle tracking is best achieved using monopulse beams. The sum beam is used on transmit, and the sum and difference beams are used on receive to accurately determine the target angle. Monopulse horn feeds may be used in space-fed phased arrays. Corporate feeds are elaborate, but they provide a high degree of control in the design of illumination functions to obtain desired sum and difference patterns. Essentially independent shaping of the sum and difference patterns may be achieved by forming weighted column sums and differences and combining these to form the array sum, transverse difference and elevation difference.
The accurate determination of the direction of targets is made using the null position of the difference pattern. Thermal noise limits the accuracy of this estimate. The error caused by noise is the smallest when the target is exactly on the beam axis. Tracking radars feedback control the beam-pointing angle to minimize the errors in the determination of the target angle. Element phase errors degrade the beam-pointing accuracy. For a 10 [degrees] random phase error at each element, a 10,000-element array has a beam-pointing accuracy of [approximately]6% of the beamwidth. For a 1 [degree] beam, this error is [approximately]0.06 [degrees]. One component of the phase error is the error from quantization; as an example, a 5-bit phase shifter will have an 11% quantization error.
In comparison, the plasma-mirror orientation is controlled by the steering magnetic field. The control is therefore analog as opposed to digital. The accuracy would depend partly on the stability of the magnetic-field power supplies. However, appropriate techniques are available that can yield beam-pointing accuracies in both azimuth and elevation better than 0.02 [degrees]. The monopulse focal plane feed horn array for a plasma mirror based antenna system may be a 4-, 5-, 12- or 44-horn array or some other optimized feed array set.
Multipath introduces errors in target angle measurement. One approach used in multipath situations is to determine the target height by using a 200-MHz instantaneous bandwidth transmission and determining the time difference between the direct wave and reflected wave returns. The plasma mirror is ideally suited for such wide-bandwidth transmissions.
Compared to high-performance phased-array radar systems, a plasma-mirror-based radar system is much simpler and is therefore more affordable. An added advantage is that a plasma-mirror-based system permits the use of wide instantaneous bandwidth waveforms. The combination of a high-performance offset-fed reflector with an electronically steerable plasma mirror may form the basis for an extremely capable radar system capable of operating with bandwidths exceeding an octave over the X and Ku bands. Limited scan antenna systems using a plasma mirror may offer multi-octave agility bandwidths over a range of operating frequencies extending from the L to Ku bands.
1. D.R. Wehner, High-Resolution Radar (Artech House: Boston), 1995.
2. J. Mathew, et al., "Electronically Steerable Plasma Mirror for Radar Applications" Proceedings IEEE International Radar Conference, Alexandria, VA, USA, pp. 742-747, May 1995.
3. J. Mathew, et al., "Electronically Steerable Plasma Mirror for Surveillance Radar Applications," Proceedings of the IEEE National Radar Conference, Ann Arbor, MI, pp. 196-201, May 1996.
4. R. A. Meger, et al., "Experimental Investigations of the Formation of a Plasma Mirror for High Frequency Microwave Beam Steering," Physics of Plasmas, vol. 2, p. 2532, 1995.
5. E. Brookner, Aspects of Modern Radar, (Artech House: Boston), 1988.
6. E. Carlsson, et al., "Search Radar Reflector Antennas with Extremely Low Sidelobes," Proceedings on Military Microwaves, London. England, pp. 500-505, October 1982.
7. A. W. Rudge and N.A Adatia, "Offset-Parabolic-Reflector Antennas: A Review," Proceedings of the IEEE, vol. 66, p. 1592, 1978.
8. V. Galindo-Israel, et al., "On the Theory of the Synthesis of Offset Dual-Shaped Reflectors-Case Examples," IEEE Transactions on Antenna Propagation, vol. 39, p. 620, 1991.
9. J. M. Howell, "Limited Scan Antennas," IEEE AP-S International Symposium, 1972.
10. AN/TPN-25 radar, manufactured by Raytheon Co., Tewksbury, MA.
11. J. B. L. Rao, et al., "Affordable Phased Array for Ship Self-Defense Engagement Radar," Proceedings of the IEEE National Radar Conference, Ann Arbor, MI, pp. 32-37, May 1996.
12. J.B.L Rao and D.P. Patel, "Voltage Controlled Ferroelectric-Lens Phased Arrays," IEEE AP-S International Symposium, Baltimore, MD, July 1996.
13. M.I. Skolnik, Radar Handbook (McGraw Hill: New York), 1990.
14. M. G. Parent, "A Survey of Optical Beamforming Techniques," Proceedings, Antenna Applications Symposium, Allerton Park, Monticello, IL, vol. 1, September 1995.
15. A. Goutzoulis, K. Davies and J. Zomps, "Hybrid Electronic Fiber Optic Wavelength Multiplexed System for True Time-Delay Steering of Phased Array Antennas," Optical Engineering, vol. 31, p. 2312, 1992.
16. R. Esman, et al., "Fiber-Optic Prism True Time-Delay Antenna Feed," IEEE Photonics Tech. Letter, vol. 5, p. 1347, 1993.
17. M. Y. Frankel and R. D. Esman, "True Time-Delay Fiber-Optic Control of an Ultrawideband Array Transmitter/Receiver with Multibeam Capability," IEEE Trans. MTT, vol. 43, p. 2387, 1995.
18. C. Herper, et al., "Combined Radar and Illuminator for Sea Skimmers," Proceedings of the IEEE National Radar Conference, Ann Arbor, MI, pp. 243-248, May 1996.
Joseph Mathew is with the Plasma Physics Division at the Naval Research Laboratory. He is presently the principal scientist on the Agile Mirror experiment at NRL. Dr. Mathew is a member of the American Physical Society. Author's Current Address: Naval Research Laboratory, Code 6700, Building A59, 4555 Overlook Avenue, SW, Washington, DC 20375.
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of Electronic Defense|
|Date:||Jan 1, 1997|
|Previous Article:||Changing world - changing nature of conflicts - a critical role for military radar.|
|Next Article:||The passive mm-wave scenario.|