Sensor research and development is assuming a high profile today. Research in materials, structures, chips, modules and subsystems is leading the way to radars that perform electronic warfare (EW) functions; communications systems that take on signals intelligence (sigint) and EW functions; and unmanned aircraft whose radar arrays double as wing structures. Emerging platforms like the F/A-18E/F and the Joint Strike Fighter plan to adopt some of these features.
SensorCraft, an unmanned air vehicle (UAV) concept for intelligence, surveillance and reconnaissance (ISR), reverses the usual model of aircraft development: sensors are optimized and the air vehicle is built around them. The classical approach is to design the airframe and then bolt on the radar and other sensors, explains John Perdzock, SensorCraft air vehicles lead with the Air Vehicles Directorate of the Air Force Research Laboratory (AFRL). The current concept draws upon research not only in sensors, but also in propulsion and aerodynamics.
"I consider it more of a second-generation Global Hawk [UAV]," Perdzock says, with the air-to-air functionality of an AWACS [E-3 Airborne Warning And Control System aircraft] and the air-to-ground functionality of a JSTARS [E-8 Joint Surveillance Target Attack Radar System aircraft]. The SensorCraft would collect data at standoff range, relaying it to a data exploitation center. If all goes well, a prototype could be available by 2010, according to press reports.
One possible airframe, an open diamond, "draws a lot of interest [because] it can put the sensors as far outboard of the other interfering elements of the airframe as possible," Perdzock explains. In that vehicle the sensors would be embedded in the outboard edges of the airframe, giving them maximum visibility to the ground and air around them. SensorCraft would be a subsonic, high-altitude, long-endurance (HALE) vehicle, flying at 60,000 feet or higher for 40 hours or more per mission.
Cutting Cost and Weight
The SensorCraft radar would combine air and ground moving target identification (GMTI), imaging and foliage-penetration applications; electro-optical/infrared sensors also would be used. Building a lightweight, low-cost sensor and then integrating it into the wing structure are key challenges on the radio frequency (RF) side, which is regarded as the most difficult aspect of SensorCraft. The active, electronically scanned radar must be lighter in weight–in the thousand-fold range–and much lower in cost than today’s technology. Using lightweight materials would enable affordable radars that are "five to six times bigger in area than what we have today," Perdzock says. "It’s fundamentally a reduction in the size of the electronics."
Key to the SensorCraft are load-bearing antennas, where the sensor becomes part of the wing, rather than a "parasitic" load bolted onto the airframe. The resulting antenna would be more susceptible to aerodynamic pressures–less stable than traditional structures. So engineers would embed sensors in the wing to track antenna movement and deformation in order for software to compensate for these factors. Trillions of floating-point operations per second (teraflops) of processing power will be necessary, but AFRL assumes the electronics industry will be able to meet these needs.
Advanced concepts such as SensorCraft will rely on progress in many areas, including work on transmit/receive (T/R) modules, ranging from chips to materials. For example, Raytheon, which is developing the APG-79 active electronically scanned array (AESA) for the Boeing F/A-18E/F Super Hornet, is using "tile" packaging technology at the T/R module level. Tile technology places RF components more closely together than was possible before.
Small Antenna Research
Tile technology "probably produces the smallest-sized, AESA-quality antenna in the world," says Frank Fleischer, director of business development for Raytheon Air Combat and Strike Systems. For the same frequency range and application, tile packaging yields T/R modules one-quarter to one-third the size of earlier technology. "We have been able to take the traditional design and alter it to reduce the size and weight of the device, with equivalent performance," Fleischer says.
Research may yield T/R modules with multifunctional, wider-bandwidth components, says Jack Urbaniak, Raytheon manager of business development. And tile backplanes could be made stronger, lighter and multilayered.
AFRL’s Sensors Directorate is working to make smaller, lighter, more flexible, larger-bandwidth and lower-loss RF components. Substrates in T/R modules, which hold the RF transmission line that guides the microwave energy, are being built with multiple layers, "making better use of the vertical dimension to route signals" and reducing weight, says Robert Kemerley, chief of the Aerospace Components Division at AFRL’s Sensors Directorate.
AFRL is looking at embedding transmission lines inside the substrate, using stacked polymer material. The object is to make modules lighter weight and more flexible so that they can conform to the airframe surfaces. Raytheon and Northrop Grumman are involved in this work.
More Powerful Chips
The use of indium phosphide and gallium nitride semiconductor materials also promises higher-power, higher-temperature microwave chips with greater power density, Kemerley says. "We are hoping to apply the chip technology to the polymer-based [substrate] material." Raytheon, Northrop Grumman, TRW, Lockheed Martin, IBM and Cree are developing chip technology in a collection of 10 to 15 small programs.
Likewise AFRL is investigating lower-loss phase shifters, the components that switch microwave signals to steer the antenna beam. Lower energy loss is important on UAVs or in space because the sensor is limited by battery power. In a manned aircraft more efficient switching would minimize cooling requirements and reduce size and weight.
Micro electromechanical systems (MEMS) technology promises more efficient operation than today’s transistor-based phase shifters. Although transistor-based devices switch at much faster speeds, they probably have two to four times the energy loss, Kemerley says. A 5-bit, transistor-based phase shifter, for example, could have a 4- to 6-dB loss. Researchers hope that a MEMS-based phase shifter could operate with less than a 2-dB loss, an improvement of up to 50 percent. But reliability is an issue. AFRL wants to get 109 to 1012 switching operations over the life of these components.
There is also great interest in systems that can perform multiple functions, such as radar, EW and communications, Kemerley says. Multifunctional systems will save money and real estate and perhaps accomplish the missions more effectively.
Such systems will need wider-bandwidth, highly integrated components that can be controlled more effectively and process information more efficiently than traditionally has been the case. And as bandwidth increases, maintaining power, dynamic range and low-noise operations becomes more difficult, Kemerley says. An amplifier, for example, has to match input and output signals over the frequency range; as the range expands, it becomes more difficult to do so. But if the signals aren’t matched, the radar won’t get the desired power output or range. "A number of component projects are investigating these problems, and system design and analysis has to be done," Kemerley says.
Raytheon’s active electronically scanned array for the F/A-18E/F Super Hornet–planned for delivery beginning in 2005–is expected to support active and passive EW, as well as traditional X-band radar functions. The F/A-18E/F AESA will have much broader bandwidth than traditional mechanically scanned radars, says Paul Summers, Boeing’s director of F/A-18 derivative programs. The F/A-18E/F AESA includes an embedded interferometry array, which will allow target azimuth and elevation data to be collected passively. The bandwidth is "wide enough to address ‘shooters’–X-band aircraft radars–but not broad enough to handle typical IADS [integrated air defense systems] assets, low-frequency emitters," Summers says. The radar also boasts higher power output, which means longer range and better performance against low-signature targets.
If radar bandwidth is widened and T/R modules operate over a wider frequency range, "you could look at something that has respectable operation as an EW system," Fleischer says.
However, the nose-mounted AESA is forward-looking, with +/-70-degree coverage, Summers says. So the AESA is supplemented by a radar warning receiver–the AN/ALR-67(V)3–which provides 360-degree passive coverage.
The Airborne Communications Node (ACN) program at the Defense Advanced Research Projects Agency (DARPA) is evolving into a multifunctional system, as well. ACN has been described as a software-reprogrammable radio that could provide in-theater communications switching. While the first phase of the program focused on communications relay, the current phase emphasizes the development of simultaneous communications, signals intelligence and EW functions. ACN would interoperate with multiple waveforms across a 30-MHz to 2-GHz spectrum.
The ACN system could scale from UAVs to large, manned aircraft. It also will comply with the Joint Tactical Radio System (JTRS) architecture (March 2002). Raytheon is competing with BAE Systems in phase 2 of the ACN program, which will culminate in a lab demo in August 2002. After that DARPA will select one team to prepare for an advanced concept technology demonstration (ACTD) in 2004 or 2005.
The onboard system could reuse core components, says Roger Bache, Raytheon’s ACN program manager. "You can use the same antenna to communicate and intercept signals," he says, and the same RF processing chain can perform multiple functions. Among the challenges are co-site and co-channel interference, the former from multiple transmitters on the same platform and the latter from ground emitters, such as radio and television stations, transmitting on the same frequency.
RF nanotechnology, the investigation of materials and components at the one-billionth-meter level, seems to be the new frontier, however. "Nanotechnology gets to the atomic level in putting together new materials," where boundaries between circuits and devices blur, AFRL’s Kemerley says. Nanoscale structures can be manipulated and integrated into larger components and subsystems with new properties and functions.
New electromagnetic analysis tools and thermal analysis tools will be needed to design integrated components. Kemerley expects this research area to expand significantly over the next five years.
Sensors for JSF
Sensors for the F-35 Joint Strike Fighter (JSF)–now in the system development and demonstration (SDD) phase–will provide pilots with information that "has not been available in tactical aircraft before," claims Robert Thompson, director of JSF combat avionics for Northrop Grumman’s Electronic Systems Sector.
Northrop Grumman’s distributed aperture system (DAS) development project for JSF seems to bear Thompson out. DAS is to provide the pilot a protective, electro-optical (EO) sphere for situational awareness and missile warning. (Previous EO missile warning systems provided only a directional cue.)
DAS, with six self-contained sensors distributed around the airframe, would generate a complete infrared image of the pilot’s environment viewable on a Kaiser helmet-mounted display. DAS and JSF’s electro-optical targeting system (EOTS) are being developed as part of a partnership with Lockheed Martin Missiles and Fire Control, the EOTS lead.
Northrop Grumman is contributing focal plane array technology, software and cryogenic coolers for EOTS. Mounted behind a sapphire window under the aircraft’s nose, the sensor provides long-range, high-resolution IR/video imagery, targeting forward-looking infrared (FLIR) with IR search and track functions, and continuous digital zoom.
The JSF’s active electronically scanned array (AESA) supports not only radar but also electronic warfare (EW) functions.
The JSF radar will build on Northrop Grumman’s experience with the F-16 and F-22 radar systems. The prototype radar was flown 100 flight test hours during the program’s concept demonstration phase, demonstrating the real-time autonomous acquisition and targeting of stationary and moving targets. The radar operates in air-to-air, ground mapping and ground moving target identification (GMTI) modes, the last of which leveraged Joint Surveillance Target Attack Radar System (JSTARS) aircraft technology.
Northrop Grumman claims that the radar will be built to last two and a half times the life of the aircraft. The unit will cost and weigh half as much as, but double the reliability of current systems, the company asserts.
Armed with a prototype based on reasonably mature technology, Northrop Grumman now will have to demonstrate the sensor’s operability inside the aircraft. "We don’t feel there’s a lot of risk going forward," Thompson says. The first F-35 in the SDD phase is slated to fly in four years. The radar will be first sensor in the airplane, followed by the EO systems.
Signal Processing: KASSPER
One of the most difficult signal processing tasks is sorting out targets from clutter in ground moving target identification (GMTI) radar operations. This task becomes more difficult in areas where strong reflections from buildings, power lines, water towers, complex mountain surfaces and other structures compete with the target signal.
Older methods of removing clutter are felt to be insufficient to detect and track slow-moving targets on the ground. In response to the problem, the Defense Advanced Research Projects Agency (DARPA) has started the Knowledge-Aided Sensor Signal Processing Expert Reasoning (KASSPER) program. KASSPER attempts to apply the potential of voluminous terrain, ground cover and land use databases and "expert reasoning" techniques to raw sensor data in order to optimize clutter extraction. These databases and reasoning techniques "have never had a role in the radar front-end," says Joe Guerci, deputy director of DARPA’s Special Projects Office and manager of the KASSPER program. The two-phase effort will first undertake mathematical research and develop algorithms. Later the program will apply the knowledge amassed in the first phase to develop a real-time, embedded signal processor and software to implement these algorithms.
If the terrain information in these databases could be matched with the radar information of the same area–automatically, in real time–it would be possible to optimize filters to extract clutter, reducing false alarms. Expert reasoning techniques, derived from radar operator practice, would further contribute to an automated, front-end radar detection scheme.
Raw digital data arrives at rates of trillions of bits per second (terabits/sec) but must be understood and applied to target detection in real time. Today’s processors compute at billions of floating-point operations per second (gigaflops), which may not be adequate. "We’re trying to figure out whether we need teraflops [trillions of floating-point operations per second] or a new computer architecture," Guerci says.