Radiometer Ils Monitor Studied
Moffett Field, Calif.-Pilot making a low visibility approach may in the future obtain at least a partial view of the airport from an airborne microwave radiometer scanning forward and down across the runway from his airplane. The radiometer output would be processed,formatted and displayed in the cockpit as a pictorial representation of the runway and its environs. The ground scene depicted in the display will lack the detail of direct visual observation, but could portray such key features as approach lights, runway outline, touchdown point and the presence of other aircraft or surface vehicles on the runway. The scale and perspective of the displays will change in the same way as the pilot's view through the windscreen in clear weather. Advocates of visual landing aids that function independently of the instrument landing system (ILS) maintain that these aids are needed to establish pilot confidence in the performance of advanced landing systems being designed and implemented for Category 2 and 3 operations. These visual aids, generically called independent landing monitors (ILM), also are needed, their proponents say, to give the pilot the visual cues he needs if he must assume manual control close to the ground under instrument landing conditions. The ILM concept has been strongly recommended, for example, by the allweather flying committee of the Air Line Pilots Assn, (AW&ST, July 27, 1970, P27). The National Aeronautics and Space Administration’s Ames Research Center is studying the feasibility of a radiometric ILM, and plans to begin testing a 37-gc. system in about a year. The frequency is one at which electromagnetic radiation can penetrate atmospheric moisture that reduces or blocks human vision. Another radiation “window,” at 94 gc., was considered and rejected in earlier trade studies. The shorter wavelength at that frequency would result in higher image resolution and permit the use of a smaller antenna, but signal attenuation is greater, and device development less advanced at the higher frequency. The Ames study, which will include both ground and flight tests using an Aerojet General Corp. radiometer, is designed to determine whether, and under what conditions a radiometer is an appropriate sensor for an ILM. Some of the questions to be answered as a result of the study are: ■ Whether the radiometer will function as predicted at the 3-10-deg. grazing angles encountered during fixed and rotary wing aircraft approaches. Most radiometric viewing of the earth to date has been from high altitude aircraft and satellites at a viewing angle of nearly 90 deg. ■ The effects of changing weather conditions and signal polarization. The antenna to be used in the test program will detect horizontally polarized signals only. ■ What kind of system performance can be obtained when sensing natural radiation only, and how that performance may be upgraded by supplementary illuminators and passive reflectors on thep ground. Passive reflectors pointed at the cold sky could outline a runway or mark features such as the touchdown point. They would have to be large enough to be visible to the radiometer and be provided with suitable means of maintaining their reflectivity. Discrete low power emitters could be very effective for outlining the runway, said John Dimeff, chief of the instrumentation division at Ames. He predicted that low-cost, low-power solid state sources using Gunn or Impatt oscillators would become available at a cost at least an order of magnitude lower than Category 2 runway lights. Milliwatt sources with a small horn probably could be picked up at 5-10-mi. ranges, and possibly could be built into runway lighting pods, he suggested. Additional emitters could be deployed on the landscape to facilitate alignment with the runway. The use of such emitters to outline the runway is similar in concept to the Bendix Micro-Vision system. However, a radiometric ILM could remain operative outside the range of the emitters-in a taxiing mode, for example—and its wide bandwidth should eliminate problems with overflights, fading and signal coherence, Dimeff said. Another type of supplementary illumination that will be considered is microwave “floodlighting” over the runway or a broader area of the airport. Basically, a radiometer is a sensitive, special purpose radio receiver capable of accurately measuring low intensity signals. Unlike radar, which also is being advocated and developed for the ILM function, the radiometer is a completely passive instrument. It senses forward scattered energy emitted or reflected from objects within its field of view. Present equipment is expensive, primarily because of low volume, but in quantity production, an ILM radiometer should cost less than a comparable radar, said an Aerojet spokesman. The radiometer does not require a transmitter section, and it acquires information with essentially the same perspective as the pilot’s eye. Therefore, it is less complicated to process the radiometer output for the display. Because the radiometer is a passive sensor, it will not produce spurious signals in other nearby receivers. On the other hand, a large number of aircraft operating ILM radars in the vicinity of a busy airport could create an electromagnetic interference problem, said Richard Olson, advanced development manager at Aerojet’s Microwave Div. All objects not at absolute zero temperature emit random electromagnetic energy continuously as thermal noise. Because different materials radiate energy at different intensities, they can be distinguished by the radiometer according to their apparent, or brightness temperature. Metal objects are reflectors rather than emitters, and therefore appear at the temperature of the objects they retlect—cold, if on a runway reflecting the cold sky. The brightness temperature of nonmetals depends on the object’s actual temperature—as sensed by a thermometer—and dielectric constant, and the frequency of observation. Different materials at the same temperature will appear to the radiometer to have a different brightness temperature. Thus, as the antenna beam scans across the border between two dissimilar materials on an airport’s surface—earth and concrete, for example—the radiometer should detect a change of signal strength. Ideally, the antenna beam will be kept narrow to improve image resolution. Receiver sensitivity determines the length of time a scanning beam must dwell in one position before it can move on. Recent improvements in receiver noise figure and antenna design at 37 gc. are making full raster scans marginally possible now, but an unaided system can not yet scan fast enough for a practical ILM, said Ames’ Dimeff. Scan rate requirements are determined by the rates at which the display data must be updated. Rates as low as once per second may be enough for the relatively static scene near the horizon, but 20 frames/sec. may be needed for the rapidly changing foreground, according to Everett Palmer, of Ames’ biotechnology division. The Aerojet system, operating with supplementary ground illumination, will scan initially at 6 frames/sec., but this rate can be increased by raising the power of the ground illumination. The Aerojet radiometer will consist of a flat phased array antenna and a separate receiver electronics subassembly. The antenna, ground illuminator, and output processor are being designed and built under a $270,000 contract with Ames. The receiver, a standard product, is being loaned to Ames for the tests. The antenna array, consisting of about 150 vertical waveguide elements, will scan electronically in the horizontal plane (across the runway) and mechanically in the vertical plane (along the runway). The mechanical scan feature is primarily for simplicity and economy in the experimental system, and will be replaced by a fully electronic scan system, capable of higher scan rates, in an operational system. The antenna will be 38 in. wide by 13 in. high, producing a beam 0.6 deg. wide by 2 deg. high at the half-power points, and will scan a volume 20 deg. (vertical) by 30 deg. (horizontal). The beam position will shift half the beam’s width, or 0.3 deg. with each step, while the mechanical stepping will shift the beam 2 deg./step, resulting in no vertical overlap. Angular resolution in azimuth will be 0.6 deg. said Olson. The antenna’s relatively small vertical aperture, and the resultant 4:1 aspect ratio of the antenna beam, would not, if unaided, produce acceptable angular resolution in elevation, and was incorporated mainly to save money. The lost resolution will be regained by the use of supplementary ground equipment, a narrow pulse transmitter that will be located about 5,000 ft. in front of the runway. The noise pulses emitted by this device will progress through the beamprint on the ground, subdividing it, in effect, into a number of smaller resolution cells. As a result, resolution in elevation will be limited by the 525-line display system, rather than by the radiometer, according to Olson. The big obstacle has been the lack of a suitable ferrite phase shifter for highspeed electronic beam shifting, Dimeff said. If the desired resolution were to be obtained solely by electronically switching a narrow beam, the switching speed would have to increase substantially. Fast switching ferrites are costly and require high levels of power to accomplish the switching, hence the alternate approach. Although this technique was devised to circumvent the current unavailability of suitable components, it might be applicable in operational systems, Dimeff said, for it would permit the use of smaller, less expensive components. However, a way would have to be found to prevent the ground illuminator from interfering with other equipment. The receiver has a common front end that feeds two intermediate frequency amplifiers. One, with 100 me. bandwidth, will be used in conjunction with the ground illuminator. The second, with 300 me. bandwith, will be used when the radiometer is functioning without supplemental illumination. The receiver will not use Dicke switching, but will stabilize gain with a second order servo loop. As currently conceived, the radiometer will become operative a few miles from the touchdown point, and could continue to function in a ground navigation mode after touchdown when surface visibility is poor. Eventually, it might be adapted to function as an aircraft warning indicator when it is not being utilized as a landing aid. A 15,000 ft. range, from the airplane to the start of the runway, in Category 3C weather is specified for the test system with the illuminator working and an update rate of 6 frames/sec. The illuminator will not be effective once the airplane has landed, and all ground navigation would be performed unaided and with the wideband IF amplifier. Update requirements are not as stringent once the airplane is on the ground, of course. The Ames biotechnology division has been studying the value of cockpit displays as monitors and manual landing aids, the symbology requirements of an effective display, and the effects of factors such as degraded resolution and update rate on pilot performance, as-observed in simulators. The earliest tests, with a television type display that provided only an aircraft symbol, a horizon, and the runway outline, showed that performance did not fall off noticeably with degraded resolution or update information, but the basic performance level was not very good anyway, said Palmer. Glide slope performance was particularly bad, and the pilots could not, for example, distinguish between a 3 deg. and a 4 deg. glide slope from the display. Runway outlines alone were not adequate for providing height or sink rate information near touchdown. Adding symbols improved pilot performance. With a glide slope reference, for example, performance for Category 2 manual approaches tended to become marginal, though still mostly marginally unacceptable. As before, lateral control tended to be consistently better than vertical. When pilots were asked to monitor their own performance and judge whether they were within a prescribed “window” at a 100 ft. decision height, about 15% showed an error in judgment, with most tending to be optimistic about their performance. The window bounded the area ±0.3 deg. from the glide slope and ±75 ft. laterally from the extended runway centerline. In another study, performed by Serendipity, Inc., for Ames, pilots flew simulated Category 2 approaches and landings with conventional instrumentation—a flight director with an expanded localizer—and two versions of a runway perspective display. The pilots did not use the displays for manual control. The pilots’ assessment of their situation generally was improved by use of the displays, the Serendipity study reported, with greater accuracy noted for a display modified for more realistic and less cluttered portrayal of the “real world.” For example, six pilots evaluating 10 different approaches committed no errors in judging lateral offset at the decision height with the modified display, compared to a total of 20 with the flight director. Whatever the ultimate value of pictorial cockpit displays—and Palmer is cautious in predicting this—glide slope reference symbology is the most useful that can be added, he believes. Ultimately, the chief benefit from these displays may lie in their subjective value to the pilot in monitoring—even in a relatively crude way—the performance of his automatic equipment. three rockets in the rear fuselage.