Airborne Early Warning and Control

The rotodome arrangement is readily applicable to active array systems, as a three or four sided array may be fixed in the same position as the rotodome, providing 360 degree coverage and good clearance from the aircraft’s structure. Some weight will be saved by eliminating the machinery required to rotate the antenna assembly. Lockheed had proposed a three sided array in this configuration, fitted to either a new aircraft or the existing S-3 Viking airframe, to meet USN E-2C replacement requirements. The three or four sided array arrangement may be applied to any airframe able to accommodate a rotodome, and may well become the standard in years to come. Its only significant failing is that it retains much of the cost and weight penalties of the rotating rotodome.

Another configuration derived from this idea is that of the Swedish Ericsson Erieye, which uses a two sided array in a beam shaped structure, carried above the fuselage of a twin-engined commuter or larger airframe. The two sided array used in this arrangement is almost as long as the APY-2 antenna of the E-3 AWACS, potentially providing similar angular resolution performance at range, on a very small airframe. This arrangement however suffers from an obvious and significant operational limitation, as it cannot provide 360 degree coverage, using conventional active phased array technology. With each array scanning a 120 degree sector, the two sided array has a blind sector over the nose and the tail of the aircraft, and degraded antenna performance beyond 45 degrees off the beam of the aircraft.

With Sweden’s relatively compact geography this would not be an issue, as multiple platforms would cover a single area, and operating in pairs, the aircraft could patrol in two racetrack orbits set 90 degrees apart to provide overlapping coverage. The success of this scheme then devolves down to the capability of the computer datalink networking which links the platforms to each other or the ground air defence centre, to ensure that a comprehensive picture of the air situation exists at whatever is the central command point. In a heavy ECM environment, where platform to platform or platform to ground datalink function is interfered with, the two sided array has a major limitation. Producing a three or four sided array with similar array length results in a structure with a size comparable to an E-3 AWACS radome, which in turn requires at least a C-130 sized aircraft to carry it, thereby largely defeating the apparent cost advantage of the linear array concept.

Another alternative is the use of a rhombic four sided array geometry, with a 60-120-60-120 degree arrangement of arrays. While the rhombic arrangement will provide full 360 degree coverage, its effective antenna length is halved in the nose and tail sectors. The result is a compromise between the bulky but excellent four or three sided array, and the compact but partially blind two sided array. No publicly discussed proposals to date have involved the rhombic arrangement.

An idea which has created some excitement in the engineering community is the concept of conformal active arrays or “smart skins”, where active arrays are embedded in the skin of the aircraft, thus avoiding the structural, aerodynamic and weight penalties of an external radome. However, close examination of most existing airframe designs suggests they may not always accommodate this concept without some other penalties, such as coverage limitations like those suffered by the two sided array concept. The Israeli Phalcon system, which uses a 707 airframe, was reported initially to have been the first implementation of this scheme. The aircraft’s public debut has however shown this not to be true, as the aircraft uses fuselage mounted boxes for its main sidelooking arrays, and a nose mounted radome for a smaller forward looking array. The fuselage mounted linear arrays provide for excellent coverage over the beam aspect 120 degree sectors, but the nose and (reported optional) tail mounted arrays which “plug” the holes in beam array coverage are much shorter because of the fuselage diameter and thus would suffer a loss in resolution performance fore and aft.

As with two sided and rhombic array configurations, the Phalcon arrangement may or may not be suitable for a given operational environment. Where the threat axis is defined unambiguously and the aircraft’s patrol racetrack can be aligned appropriately, the coverage limitations may not be of significant importance. Where the threat can approach from multiple axes concurrently, full 360 degree coverage is almost mandatory. Westinghouse’s MESA system, which is currently in development, uses a podded arrangement of sidelooking linear arrays carried by a C-130, with some reports suggesting that supplementary nose and tail arrays could be fitted to plug nose and tail coverage.

Other alternatives do exist. An arrangement publicised by Boeing for the now defunct USN E-X program involved the use of an airframe with a trapezoidal (diamond shaped) wing, with arrays embedded in the wing surfaces to provide 360 degree coverage. The S-3 sized aircraft had been proposed to replace the E-2C. Recent budget cuts have put its future very much in doubt. The dilemma faced by designers is simple: active arrays provide the potential for a small airframe to have E-3 AWACS class antenna performance, however antenna coverage and performance requirements will force the use of either a mast mounted radome or a new airframe geometry, both negating the potential cost advantage offered by the antenna technology.

Other alternatives may yet exist, but investigating their suitability will require some effort by airframe manufacturers. Most modern airliner airframes have a wing sweep of about 60 degrees, which suggests that the leading edges of the wings have the almost ideal geometry to accommodate conformal arrays. A six to eight metre conformal array embedded in the leading edge, inboard, would automatically provide two sides of the three sided array configuration. The problem is that this would preclude the use of leading edge lift devices over at least 30% of the span and also preclude the use of wing mounted engines, which would obstruct the array’s field of view. Another issue is structural rigidity, and a flexing wing structure is not the ideal base for fitting an antenna. The unresolved issue is the third side of the array, which could only be implemented by placing an array on the tail of the aircraft. A six to eight metre array length will require, by default, a beam structure of similar size.

Close examination of available airframes suggests that those with aft mounted engines (727, MD-80) would be geometrically most suited to leading edge array placement, but the positioning of the powerplants would cause difficulties in positioning the tail array. The optimal geometry would see engine pods mounted above the wings (as with the VFW-614), and the tail array beam structure fitted to the end of the fuselage, or at the top of the vertical stabiliser. The scale of change would again force a new airframe design or substantial rework of an existing design. The application of supergain array techniques could of course alleviate some of these difficulties, and it remains to be seen how soon this technology becomes adapted to an AEW&C application. Clearly as the technology of active phased arrays matures, designers will settle upon the most suitable configurations, but as is often the case the simple and orthodox solutions may ultimately prevail merely because they were a good idea in the first place. The pylon mounted radome may be with us for some time to come.

Phased Arrays – A Perspective View

The emergence of active phased array technology in AEW&C applications promises substantial gains in capability, if the technology is developed to its full potential. The phased array offers more flexible scanning and tracking in comparison with mechanically pointed antennas, which in turn allows the antenna’s behaviour to be adaptively configured to a given situation. The usefulness of this capability should not be underestimated. The US Aegis SPY-1 radars for air defence duty during the Gulf campaign provided a case study in the operational gains to be had with this type of technology. In that situation the radar was configured to achieve much higher scan rates in sectors of particular interest, while still maintaining 360 degree situational awareness. Equally so, the software adaptations to the Patriot SAM enabled it to engage Scud ballistic missiles, and this again could not have been achieved with a conventional tracking radar.

Another area where substantial gains are available is resilience in an intense electronic warfare environment. A phased array system offers two important capabilities in this context. The first is the ability to adaptively configure the antenna to place nulls on hostile jammers, thereby confining the effect of the jammer only to the narrow sector which it physically occupies. The second is the ability to use unconventional antenna scan patterns; thereby reducing the probability that the radar’s transmissions will be detected and firmly tracked by an opponent. A conventional rotating antenna provides the opponent’s ESM with a regularly repeated transmission, once per revolution, which makes detection and tracking quite straightforward. Reliability is another strength of the phased array, particularly the active array, and it is reasonable to say that a well implemented phased array is immune to single point catastrophic failure, and substantially immune to multiple failures. In this respect it differs from conventional antennas, where mechanical and electrical failures can cripple the whole system, where items are not multiply redundant.

The central weakness of the phased array concept lies in the degradation of antenna performance as the beam is slewed away from the antenna boresight. While the loss in sensitivity can be compensated by adjusting receiver gains, the loss in angular resolution or beam sharpness requires more sophisticated methods, which will typically increase the complexity of the antenna control software. Even so, the result may be a reduction in target track confidence levels, as the antenna looks away from boresight, and this in turn can add further complexity to target tracking algorithms. Where the radar is to interface with existing systems, such as secondary radar or track management or fire control systems, where the software assumes a constant target track confidence level, the whole system design must be altered to accommodate the antenna’s idiosyncrasies.

Another area where system performance may be compromised in a phased array is in heightfinding. Just as the resolution of the beam in azimuth is determined by antenna length, so its resolution in height is determined by the height (or depth) of the array. Should the array be too short, heightfinding accuracy will suffer, and it is worth noting that the Patriot and Aegis radars, where heightfinding is critical for SAM fire control, have a 1:1 height to width ratio. In summary it is fair to say that phased arrays, while not a panacea, can provide some very substantial gains over established antenna technology. Whether a particular array design achieves this is dependent upon the implementation, as is the case with established antenna technology.

At the time of writing, the ADF’s proposed AEW&C program is still uncertain, with a Project Definition Study currently being done to determine, in relation to strategic guidance, whether such a capability is required, and if so, with what performance and system capabilities. When the PDS is completed, Technology Explained will review the ADF requirement and the possible contenders should a capability be sought. Readers interested in the technical and strategic issues involved are referred to the March and May issues of Australian Aviation, which provide an introduction to the subject.