Radar revolution: the arrival of gallium nitride components opens up new applications for radars, including jamming and telecommunications.

The active array antenna has virtually taken over the radar market--having won a berth on every new fighter or surveillance aircraft programme launched over the past 10 years. Even earlier programmes, such as the Rafale, Gripen, Typhoon or MiG-35 are preparing for the transition to active-array technology. The pressure do to so has been stepped up following the first US export sales of aircraft equipped with active-array radars--F-15s to Singapore in 2005, followed by Australia's recent order for the F/A-18E/F Super Hornet.

Weapons labs around the globe, however, are already gearing up for the next generation--antenna with the power and bandwith to perform offensive jamming and communications in addition to the radar function, and with a sufficiently compact and modular design to be housed not only in the nose of the fighter aircraft but also in conformal units elsewhere on the airframe surface.

The trigger for this revolution is a semiconductor material called gallium nitride (GaN), which, like the gallium arsenide (GaAs) used in current active antennas, is composed of elements from columns 3 and 5 of the periodic table and can be used to produce high frequency amplifiers.

The emergence of GaN from the laboratory has been delayed by epitaxy issues--growth of the semi-conductor layer on the silicon (Si) substrate, or silicon carbide (SIC) in the case of cutting-edge military applications. GaN and the substrate are made of crystals with different interatomic distances, hence the difficulty in assembling the two materials at a microscopic scale. The largest slices of high-performance GaN that have been obtained to date have a diameter of three inches, compared with six inches for GaAs and up to twelve inches for silicon. The size of the slice determines the number of chips that the machine can produce in a single pass.

GaN is clearly destined to remain expensive and its utilisation unlikely to expand beyond a limited number of applications, particularly since suppliers of SiC substrates are themselves limited. This situation could change, however, as GaN slices are expected to increase to four inches in the near future, and the arrival of new players should help to drive prices down, predicts Dominique Pons who heads the Alcatel Lucent/Thales III-V Lab. (The name reflects the columns of the periodic table mentioned above.) In any case, the intrinsic qualities of GaN have convinced the US Defense Advanced Research Projects Agency (Darpa) to invest heavily--tens of millions of dollars per year--in the technology.

The involvement of telecommunications giant Alcatel Lucent reflects the inherent duality of the technology--GaAs components are widely used in cell phones. Though power applications, such as radar, are largely confined to the military and space sectors, they are gradually finding their way into the civil domain. In the 1990s, EADS and Thales formed a joint company, United Monolithic Semiconductors (UMS), to produce GaAs chips and monolithic microwave integrated circuits (MMICs) for their new-generation radars.

Series production

UMS launched series production of MMICs for S- and C-band radars at the beginning of this decade, followed more recently by X-band radars like the active-array RBE2 AA that will equip the Rafale starting in 2012. On the civil side, lead times between technology incubation and application are much shorter, and the company has managed to find GaAs applications in a number of professional or top-end civil markets, such as wireless telecom infrastructures and anti-collision radars for cars.

In this way the military potential of GaAs has opened up an industrial capability that has found market openings in Europe ... the same openings that GaN will be able to exploit in its turn. Agreements are already in place with NXP (ex-Philips Semiconductors), explains Thierry Laboureau, UMS sales and marketing director, to develop power components for base stations for third- and fourth-generation cell phone networks and for WiMax base stations for mobile internet users. Ultimately, once prices have come down far enough, GaN could conceivably make its way into the kitchen, replacing the magnetron in the micro-wave oven.

However, these longer-term commercial perspectives will not be enough to cover the investment required to launch foundry operations. Nor is there any prospect of procuring components for military applications from the US or Japan--both countries have already placed an embargo on GaAs circuits, and there is no reason for them to be any more flexible concerning GaN. This explains why the defence procurement authorities in France and Germany are both helping to support industrial research efforts.

According to Dominique Pons, the III-V Lab should produce its first X-band or wide-band GaN MMICs this year. Following validation and industrialisation by UMS, series production should get under way by 2009.

EDA funding

GaN is also one of the very first research areas to receive funding from the European Defence Agency (EDA) under a 40 million [euro] programme called Korrigan that brings together 23 companies and laboratories in seven countries to accelerate the development of one or more European GaN foundries with associated supply chain by 2009. The programme leader is Thales Airborne Systems. Other participants include EADS, Selex Sistemi Integrati, Saab Ericsson and BAE Insyte. Their role initially is to define requirements for the foundries, before becoming directly involved, from 2008 onwards, in integrating the microchips into a variety of specialised modules covering a range of land-based and airborne radar applications, as well as self-protection or offensive jammers.

In this way, explains Thales Airborne Systems technical director Pierre Fossier, it should be possible to launch the first system applications in 2010. In France, one of the leading candidates for the new technology is the offensive jammer, a capacity that the French Air Force has had its eyes on for several years, and which has already given rise to the Carbone airborne demonstrator. The performance of the system attracted a lot of attention at NATO's Mace X electronic warfare exercise in the year 2000.

The DGA procurement branch of the French MoD is continuing to provide limited funding for exploratory work by Thales while awaiting for national budgets to kick in to complete development. GaN would allow for a reduction in the size of the jammer, potentially clearing the way for integration into a combat aircraft. One of the first European acquisition programmes to integrate GaN technology could well be the Maritime Airborne Surveillance and Control (MASC) programme to replace Royal Navy Sea King Mk7 airborne surveillance helicopters, as required for the future CVF aircraft carriers. The three candidates for this mission are the Hawkeye aircraft, the EH-101 helicopter and the tiltwing V-22, though the Hawkeye would appear to be ruled out by the absence of a catapult in the current CVF definition. Both the other candidates would require a compact and powerful radar to meet missions requirements. The potential advantages of a GaN radar in this context have prompted the British MoD to finance some upstream development work in preparation for a programme launch in the 2009 timeframe--the same year that the first European GaN modules are scheduled to come off the production line.

Rafale lead

As far as Europe's combat aircraft programmes are concerned, the Rafale seems to have established a lead over Typhoon and Gripen in the race to integrate an active array antenna. This is primarily because--unlike its competitors--the transition to active-array technology on the Rafale's electronically scanned RBE2 was planned from the outset, avoiding the need for the more extensive (and expensive) modifications required on the mechanical antennas of the Typhoon and Gripen. The increase in range that the new technology will bring is deemed essential if the aircraft is to fully exploit the potential of the future ramjet-powered Meteor missile, due to enter service in the early years of the next decade. Without it, pilots will rely on target designation from another platform to strike targets at the limits of the Meteor envelope.

All aircraft will benefit from the collaborative work accomplished under the trinational Airborne Multirole Solid State Active Array Radar (AMSAR) programme, which was launched in 1993 to develop a European capability in GaAs power devices and subsequently gave rise to UMS (EADS/Thales). Work under AMSAR is currently focused on beam forming through computation. The goal is to cancel reception in jammed sectors and improve rejection of parasitic ground echos, though at the cost of a more complex antenna architecture.

In France, Thales launched its own active antenna radar demonstrator programme in the late 1990s incorporating US components. The resulting mockup was tested at the CEV flight test centre in 2002 on a Mystere XX test bed, and the following year on Rafale. In February 2004, the French MoD's DGA procurement branch awarded 85 million [euro] under the DRAMA programme to develop a prototype active-module radar representative of an operational system.