The Aegis SPY-1 Radar
The Aegis SPY-1 radar is part of the Aegis combat system deployed on U.S. Navy cruisers and destroyers as well as on a number of foreign ships. Originally designed as an air defense system, the Aegis system on many U.S. Navy ships has been or is being upgraded to include a ballistic missile defense (BMD) capability.
The U.S. Navy currently operates 22 Aegis cruisers (CG-47 or Ticonderoga class), although it currently plans to retire seven of these in FY2013 and FY 2014. Five of the cruisers have so far received BMD upgrades (although one of these is among ones scheduled for retirement).
By the end of 2012, all 62 Aegis destroyers (DDG-51 or Arleigh Burke class) procured through 2005 will have been delivered, with 24 of these having received the BMD upgrades. In 2010, procurement of an additional ten Aegis destroyers began, with first scheduled to be operational in 2016. These ships will be delivered with a BMD capability built-in. The number of BMD-capable Aegis ships (both cruisers and destroyers) is projected to reach at least 39 by 2020. Beginning in 2016, the Navy plans on beginning procurement of a new type of destroyer (the Aegis Flight III) with a more capable (and not yet completely defined) radar, with the first ship scheduled to be operational in 2023.
Aegis Cruiser (CG 72, Vella Gulf). Two of the Aegis antenna array faces are visible on the rear deckhouse. The other two antenna faces are on the forward deckhouse but are not visible here. (Picture source: U.S. Navy)
Aegis Radar Versions
Four different versions of the Aegis SPY-1 radar are currently deployed on U.S. ships. The SPY-1 was a test version of the radar that was never deployed.
The SPY-1A is currently deployed on the two oldest cruisers still in service (five earlier cruisers have already been retired). These two ships are not among the seven planned for retirement in the next few years, nor have they received BMD upgrades.
The SPY-1B is deployed on the fifteen subsequent cruisers. The 1B version has a new antenna, with much better sidelobe characteristics, which is important when operating in an environment with clutter as the Aegis radars must often do. The SPY-1B also has about twice the average power of the 1A version and other improvements. The power increase was achieved by increasing the radar’s duty factor (the percentage of the time the radar is emitting) without changing its peak power.
Each SPY-1 radar has four antenna faces, each covering slightly more than 90˚ in azimuth. In the 1A and 1B versions on the cruisers, there are two transmitters, each multiplexed between the two antenna faces on each of the two deckhouses.
The SPY-1D is deployed on the first 40 destroyers. It is very similar to the 1B version, except that one transmitter is used to drive all four radar faces, all located on one deckhouse. (Thus there is at most one beam in the air at any given time). All the U.S. Aegis systems which have been upgraded for BMD use have either the 1B or 1D version of the radar. The four ships to be based at Rota, Spain by 2015 as part of the European Phased Adaptive (EPAA) missile defense plan are all destroyers with the 1D radar.
The SPY-1D(V) version (the “littoral warfare” radar) is deployed on the subsequent Aegis destroyers (about 22 so far), starting with DDG-91 in 2005. This upgrade added a number of waveforms for improved clutter rejection and moving target detection to improve the capability of the Aegis radar in environments with ground and other near surface clutter. It also increased the transmitter average power (by at least 33%) and it also added a dual-beam capability which enabled it to put out two beams simultaneously (out of opposite faces).
Aegis Destroyer (DDG-74, U.S.S. McFaul). The two forward Aegis antenna faces are visible on the forward deckhouse. The other two faces are on the same deckhouse but are not visible here. (Photograph source: U.S. Navy)
AEGIS Radar Characteristics
The focus of the description here is on the 1B/D versions, and on their physical characteristics relevant to BMD use.
Frequency and Bandwidth:
The Aegis system operates in S-band, from about 3.1 to 3.5 GHz (λ = 8.6 to 9.7 cm). Early descriptions indicated that the system reportedly had a “sustained coherent bandwidth” of 10 MHz and instantaneous bandwidth of 40 MHz. An early paper on the SPY-1 radar discusses three sub-bands, FL, FC, and FH, each 40 MHz wide, in the context of measuring antenna gain. However, data was also collected over much wider bands than the defined ones — 160 MHz at broadside, and 120 MHz at a 60 degrees scan angle.
The Aegis system’s bandwidth was apparently subsequently increased, perhaps up to its maximum frequency extent of 400 MHz. The 4.0.1 version of the Aegis Ballistic Missile Defense system, which is now entering service, added an adjunct BMD Signal Processor that, among other things, allows the formation of two-dimensional inverse synthetic aperture images with better resolution than had previously been possible, which implies a wideband capability. A 1999 Lincoln Laboratory briefing slide shows a “Wideband Waveform Concept for AN/SPY-1 Radar” using a 400 MHz wideband waveform constructed from ten 40 MHz bandwidth pulses frequency jumping from 3.1 to 3.5 GHz. A 2002 paper cites a bandwidth of 300 MHz for Aegis. Such a bandwidth would likely permit a range resolution of about 0.5-1.0 meters.
Antennas and Beamwidths:
Each Aegis radar system has four radar antenna faces. Starting with the SPY-1B, a new antenna was introduced, that although outwardly similar in appearance to the antenna of the SPY-1A, incorporated significant improvements. In particular, it has improved peak and average sidelobes relative to the 1A version and eliminates grating lobes within the antenna scan angles. These improvements were accomplished by subdividing the antenna into many more subarrays (2,175, each with two elements, for a total of 4,350 elements) than the 1A antenna (68 subarrays of 64 elements each, for a total of 4,352 elements) and by improved machining tolerances and alignment techniques.
The antenna face physical structure is octagonal, with a height of 4.06 m and a width of 3.94 m. In the 1A version, the antenna elements themselves are contained within a similar hexagonal shape with dimensions of roughly 3.84 m in height and 3.67 m in width. The area populated by the antenna elements appears to be about 12 m2. In the 1B/D version, the antenna face itself (the area occupied by the elements) is more nearly circular than in the 1A antenna, but since the number of elements is essentially the same, it is likely that its aperture area is also about the same.
The Aegis radar reportedly has a gain of G = 42 dB (= 15,800) and a beamwidth of 1.7˚x 1.7˚. This gain figure is consistent with G = ρ(4πA/λ2) with A =12 m2 and λ = 9.1 cm, only if ρ = 0.87 (which seems too high). Moreover, a gain of 42 dB appears to be inconsistent with the stated beamwidth of 1.7˚, which indicates a lower gain of about G =9,000.
As noted above, the 1B and 1D versions are nearly identical except that the 1B version uses two transmitters for each pair of two antenna faces, whereas the 1D version has one transmitter for all four faces. However, since a transmitter can apparently be used with only one face at a time, the maximum power that can be put out of any antenna face should be same for both versions.
The original SPY-1A version reportedly has a peak power of up to 5 MW and an average power of 32 kW. The SPY-A’s transmitter output is provided by 32 crossed field amplifiers (CFAs), each with peak power of 132 kw, which would give a combined peak power of 4.2 MW. This seems to indicate that reported peak and average powers for the radar are the transmitter power, not the power actually emitted, which will be less due to losses between the transmitter and antenna.
The SPY-1B reportedly has an average power of 58 kW with a peak power of 4-6 MW. This is roughly consistent with reports that the 1B version had the same peak power but twice the average power (that is, its duty cycle was doubled) of the 1A, and that, more specifically, the SPY-1B/D used a new CFA with a doubled duty cycle.
According to a 2004 Defense Science Board Report, “the average radiated power aperture for the Aegis System is 485 kwm2.” Assuming that statement applies to the SPY-1D (since the SPY-1D(V) version was not yet operational) and an antenna area of 12 m2, this would give an average emitted power of about 40 kW.
The Aegis radar (1B version) can produce pulses with lengths of 6.4, 12.7, 25 and 51 microseconds, with a pulse compression ratio of 128. This 51 μs maximum pulse length is consistent with a 1997 study that stated that the electromagnetic interference produced by an Aegis radar pulse would last for at most 52 microseconds. However, given the many upgrades to the Aegis system, including the BMD upgrades, these pulse lengths may have changed significantly.
A 1978 paper states that the noise figure for the Aegis SPY-1A receiver was about 4.25 dB = 2.66.
The version of the AEGIS radar currently being built, the SPY-1D(V) was first deployed on U.S. Navy destroyers in 2005, beginning with DDG-91. This upgrade does not appear to involve significant changes to the antenna. A 33% increase in duty cycle was apparently set as a requirement for the SPY-1D(V) upgrade. An increase of “over 33%” in amplifier duty cycle over that of SFD-262 CFA was achieved in the SFD-268 CFA, intended for use in the D(V) radar, in part by using improved cooling techniques. This would give an average transmitter power of at least 77 kW, based on a 58 kW average transmitter power for the 1B/D version.
The only public numerical figure on Aegis detection range against a specific target (that I have seen) is that the SPY-1D “can track golf ball-sized targets at ranges in excess of 165 kilometers.” A golf ball-size (1.68 inches diameter) sphere corresponds to radar cross section of about 0.0025 m2 at 3.3 GHz. This statement was made in the context of the soon-to-be deployed SPY-1D(V) radar to detect mortar and artillery shell and small-caliber rockets against a clutter background, so presumably it applies to the D(V) version. Scaling to a radar cross section more typical of a ballistic missile warhead (0.03 m2 at 3.3 GHz) gives a range of at least 310 km.
 The numbers in this paragraph and the next are from: Ronald O’Rourke, “Navy Aegis Ballistic Missile Defense (BMD) Program: Background and Issues for Congress,” Congressional Research Service Report RL33745, July 2, 2012. Available at: http://www.fas.org/sgp/crs/weapons/RL33745.pdf.
 The designation SPY-1C was apparently used for a manufacturer’s proposed but never built version for aircraft carriers.
 These are the Ross (DDG-71), Donald Cook (DDG-75), Porter (DDG-78), and Carney (DDG-64). Kate Wiltrout, “Norfolk To Lose 3 Destroyers as Navy Boosts Missile Defense,” The Virginia Pilot, February 17, 2012, p. A1.
 Several other countries have built or are planning to build ships similar to the U.S. Aegis destroyers. Japan has built four Kongo class (SPY-1D) and two Atago class SPY-1D(V) destroyers. Spain has built five Aegis-equipped frigates (four SPY-1D, one SPY-1D(V). South Korea and Australia both plan to build three SPY-1D(V) equipped destroyers. Norway has deployed five frigates equipped with the SPY-1F, a smaller version of the radar.
 Norman Friedman, World Naval Weapons Systems, Fifth Ed. (Annapolis, MD: Naval Institute Press, 2006) p. 316. However all the information from Friedman cited in this post also appears in earlier book editions going back to at least the 1991/92 edition.
W.T. Patton, “Compact, Constrained Feed Phased Array for AN/SPY-1,” in Eli Brookner, Practical Phased-Array Antenna Systems (Norwood, Mass.: Artech, 1991), pp. 8-1 – 8-32.
Bernard Ulfers and George LeFurjah, “AN/SPY-1B/D Radar Design Changes Supporting Ballistic Missile Defense, “ Leading Edge, Vol. 7, No. 2, Naval Sea Systems Command, pp. 100-105. Available at: http://www.navsea.navy.mil/nswc/dahlgren/Leading%20Edge/Sensors/05_Modernization.pdf.
 Eric D. Evans, “Missile Defense Technology (Can BMD Systems Work?)” Mini DTS Course, MIT Lincoln Laboratory, December 10, 1999.
 Philip A. Ingerwesen, William W. Camp and Alan J. Fenn, “Radar Technology for Ballistic Missile Defense,” Lincoln Laboratory Journal, Vol. 13, No. 1 (2002), pp. 109-147 (p. 141).
 Richard L. Britton, T.W. Kimbrell, C.E. Caldwell, and Gerald C. Rose, “AN/SPY-1 Planned Improvements,” IEEE EASCON (1982), reprinted in Merrill I. Skolnik, Radar Applications (New York: IEEE Press, 1988), pp. 192-199.
 W.T. Patton, “Compact, Constrained Feed Phased Array for AN/SPY-1,” in Eli Brookner, Practical Phased-Array Antenna Systems (Norwood, Mass.: Artech, 1991), pp. 8-1 – 8-32 (Figure 8.1). This is actually for the SPY-1, which has the same 64×68 element arrangement as the 1A.
 Estimated from Figure 8.1 of Patton, “Compact, Constrained Feed.”.
 Friedman, “World Naval Weapon Systems,” p. 316.
 According to Skolnik (Merrill I. Skolnik, Introduction to Radar Systems, 3rd Ed. (New York: McGraw-Hill, 2001), p. 541) a good approximation when other information is lacking is G = 26,000/θBφB, where θB and φB are the half power beamwidths in degrees. This gives G = 9,000.
 John A. Adam, “Pinning Defense Hopes of Aegis,” IEEE Spectrum, June 1988, pp. 27. Another description of the system gives a peak power of 4 to 6 MW (Frank Elliot, “An Inside Look at Aegis, Shield for the Fleet,” Navy Times, August 25, 1986, pp. 35-36.
 Friedman, “World Naval Weapons Systems,” p. 316.
 Friedman, “World Naval Weapons Systems,” p. 316. This (along with numerous other similar titles by Friedman) appears to the only public source for the 58 kW average power figure).
 As of 1982, both the Varian SFD-262 CFA and the Litton L4707 CFA had met the duty cycle requirements but still required additional testing. Richard L. Britton, T.W. Kimbrell, C.E. Caldwell, and Gerald C. Rose, “AN/SPY-1 Planned Improvements,” IEEE EASCON (1982), reprinted in Merrill I. Skolnik, Radar Applications (New York: IEEE Press, 1988), pp. 192-199.
 Friedman, “World Naval Weapon Systems,” p. 316.
 Sanders, F.H., Ramsey, B.J. and Hinkle, R.L., “Summary of Results of Tests and Measurements Related to RF Interference at Bath, Maine,” Institute for Telecommunication Sciences and Office of Spectrum Management, National Telecommunications and Information Administration, U.S. Department of Commerce, September 17, 1997.
 R. J. Socci, “The AEGIS Radar Receiver,” Microwave Journal, October 1978, pp. 38-47. This applies to all receiver output channels except for the sidelobe blanking channel, which was somewhat lower (3.75 dB).
 Mrityunjoy Mazumdar, “USS Pinckney Sails with Latest Aegis Suite,” Jane’s Defence Weekly, October 12, 2005. p. 31.
 C.L. Wheeland, M.S. Worthington, K.F. Ramacher, and E.M. Doyle, “Ultra Low-Noise CFA Design and Development for the AN/SPY-1B/D Radar, “ 1996 IEEE International Conference on Plasma Science (Abstract), p. 235
 Stephen Einarson, “Development and Production of the SFD-268 CFA for the Aegis AN/SPY-1D(V) Radar,” International Vacuum Electronics Conference 2000 (Abstract), Monterey, CA, May 2-4, 2000. Input power was increased by 45%, which also produced improvements in noise and jitter.
 John A. Robinson, “Force Protection from the Sea: Employing the SPY-1D Radar,” Field Artillery, March-June 2004, pp. 24-25.
 RCS calculated using NASA’s Size Estimation Model. For a discussion of this, see, for example, C.L. Stokely, J.L. Foster, Jr., E.G. Stansbery, J.R. Benbrook and Q. Juarez, “Haystack and HAX Radar Measurements of the Orbital Debris Environment; 2003, NASA Lyndon B. Johnson Space Center, November 2006., pp. 20-22. Available at: http://www.orbitaldebris.jsc.nasa.gov/library/Haystack_HAX_radar2003.pdf.