The Deep Space Network
The Deep Space Network (DSN), managed by the California Institute of Technology’s Jet Propulsion laboratory (JPL), has provided vital communications and navigation services for NASA deep space exploration missions for more than 40 years. The remarkable technical achievements of the planetary exploration program executed by NASA would not have been possible without the extensive and sophisticated communications and data handling systems that comprise the Deep Space Network. The high-resolution pictures of Jupiter’s satellites that were transmitted by the Galileo spacecraft are an example of this. Another is the ability to continue communicating with a spacecraft that was launched 30 years ago and is now more than 7 billion miles from Earth (Pioneer 10). These capabilities are the result of a sustained effort to improve the data rate capability and the sensitivity of the receivers by the research and development staff of the Jet Propulsion Laboratory in Pasadena California.
The principal facilities of the DSN are three major ground station complexes, one in the United States (Goldstone, California), one near Madrid, Spain, and one near Canberra, Australia. Each of the complexes has several tracking antennae; the largest has a diameter of 70 meters, and smaller ones have diameters of 11-34 meters. In addition to antennae, the DSN also has a complex of computers and signal processing capabilities that permit very thorough and sophisticated analysis of signals sent back from distant spacecraft.
In addition to receiving data from very distant spacecraft, the facilities of the network are also used to control the spacecraft. The network and the associated spacecraft are designed so that if a failure of a spacecraft should occur, there are often means available to effect a recovery. Perhaps the best example of this is the work-around that was developed as a result of the failure of the high-gain antenna on the Galileo spacecraft. In spite of this setback, the Galileo mission was judged more than 70% successful because of the measures taken to adapt the data receivers and the processors to the failure. Finally, the continuing improvements in the technology of the network have greatly increased the useful lives and capabilities of the spacecraft that use the network. The example of Pioneer 10 has already been mentioned. It was originally designed for a life of about 5 years, and because of continued technical improvements in the network, can still be tracked by the network 30 years later.
A thorough history of the DSN is available in a recent published topic Uplink-Downlink: A History of the Deep Space Network (1) in which these achievements are described in detail.
No history of the DSN would be complete without full appreciation of the contribution made by advanced technology to the successful development of the Network. The wellspring of new and innovative ideas for increasing the existing capability of the Network, improving reliability, operability, and cost-effectiveness, and for enabling recovery from potential mission-threatening situations has resided, from the very beginning of the Network’s history, in a strong program of advanced technology, research, and development. Many of the accomplishments of the technology program, known during much of this time as the DSN Advanced Systems Program, can be found in Ref. 2.
The Great Antennae of the Deep Space Network
To enable 24-hour coverage of deep space spacecraft, NASA/JPL maintains a complex of large antennae at each of three geographic locations; Goldstone, California, Madrid, Spain, and Canberra, Australia. A photograph of the NASA Deep Space Communications Complex (DSCC) near Canberra, Australia, is shown in Fig. 1.
The large antennae at the complexes are quasi-parabolic reflector antennae; one has a diameter of 70 meters, the others are 34 meters in diameter. These are used for deep space mission support. Other smaller (26-m or 11-m) antennae provide occasional tracking support for selected Earth-orbiting missions.
Each of the antennae has what is termed a Cassegrain configuration and has a secondary reflector mounted on the center axis just below the focal point of the primary reflector ”dish.” The secondary reflector relocates the antenna focal point closer to the surface of the main dish and thus establishes a more convenient location for low-noise amplifiers, receivers, and powerful transmitters.
The efficiency with which parabolic antennas collect radio signals from distant spacecraft is degraded to some extent by radio noise radiated by Earth’s terrain surrounding the antenna. This form of radio noise, known scientifically as ”blackbody radiation,” is a physical characteristic of all material that have a temperature above absolute zero (— 273° F. or 0K). The magnitude of noise power radiated by a material body depends on its temperature. It is incredibly small at the temperature of typical Earth surfaces, but it is enormous at the temperature of the Sun, for instance. All parabolic, radio antennae have side lobes in their beam patterns, and the magnitudes of those side lobes can increase as the antennae deviate from the ideal shape. The shape of the beam and its ”side lobes” are essentially the same whether the antenna is used for transmitting or receiving signals. The ”side lobes” are analogous to the circles of light surrounding the main beam of a flashlight when it is held close to a reflecting surface.
Figure 1. Deep Space Communications Complex; Canberra, Australia.
When an antenna is used for transmitting a strong signal to a spacecraft, the side lobes are of no great consequence. However, when the antenna is receiving a weak signal from a distant spacecraft, particularly at or near the horizon, ”Earth noise” picked up by the side lobes can be sufficient to obscure the spacecraft signal in the extremely sensitive receivers used on the DSN antennae. This produces errors in the data stream being delivered to spacecraft engineers and scientists and may cause the DSN receivers and antennae to lose the spacecraft signal altogether; in that case, the data stream is completely lost.
The earliest antennae in the DSN were of commercial design and were parabolic. Then, as now, the actual efficiency of the antenna represented a compromise between maximum signal gathering capability and minimum susceptibility to radio noise picked up from surrounding Earth.
Improved technology to reduce the side lobes and increase the signal collection capability (gain) of future DSN antennae appeared in the DSN’s Advanced Systems Program in the early 1970s. The new technology was based on a ”dual-shape” design wherein the surface shapes of both the primary and secondary reflectors were modified to illuminate the slightly reshaped ”quasi-parabolic” surface of the main reflector more uniformly (3). However, it was not until the 1980s when the first 34-meter high-efficiency antennae were built that the new ”dual-shape” design saw operational service in the Network.
These antennae were needed by the DSN to support the two Voyager spacecraft in their tour of the outer planets. At the time, the DSN was in transition from the lower, less capable, S-band (2.3 GHz) operating frequency for which the early spacecraft and antennae were designed, to the X band (8.4 GHz), a higher, more capable operating frequency. When X-band technology became available (1975) in the DSN, all later spacecraft and DSN antennae were designed to operate at X-band frequencies. These antennae were, therefore, the first to be optimized for best performance in the X band.
As the Voyager 2 spacecraft headed outward toward Neptune, it was recognized that increased signal collecting area was needed on Earth to support this unique scienctific opportunity effectively. The DSN’s largest antennae at the time were 64-m parabolas of the original design. Calculations showed that the best investment of scarce construction funds would be to modify these antennas using the dual-shape design and expanding their diameter to 70 m. It was also apparent that the upgraded large antennas would benefit the planned Galileo and Magellan missions.
Completed in time for support of Voyager 2 at Neptune, the 70-m enhancement project (Fig. 2) resulted in an increase of more than 60% in the effective collecting area of these large antennae. Fully half of the increase was attributed to the dual-shape design.
Figure 2. The 70-meter antenna with dual-shaped reflector design.
By the end of the century, several new 34-m antennae employing the dual-shaped reflector design in conjunction with beam waveguide (BWG) techniques had been constructed for operational use in the Network. The dual-shaped reflector design enhanced the radio performance of the antenna, whereas the beam waveguide configuration greatly facilitated maintaining and operating of the microwave receivers and transmitters. Using a series of additional secondary reflectors to relocate the focal point into a stationary room below the main dish, the BWG design feature enabled mounting these critical components in a fixed environment rather than in the more conventional type of moving and tipping enclosure mounted on the antenna itself.
Beam waveguide antennae had been used for many years in Earth communications satellite terminals where ease of maintenance and operation outweighed the consideration of losses introduced into the microwave signal path by the additional microwave reflecting mirrors. For deep space applications, however, where received signal power levels were orders of magnitude smaller, any losses in the signal path were a matter of great concern, and the losses associated with BWG designs kept such antennas out of consideration for DSN purposes for many years. Researchers in the DSN Advanced Systems Program, nevertheless, pursued the idea of BWG antennas for the DSN and by 1985 were ready to conduct a collaborative experiment with the Japanese Institute for Space and Aeronautical Sciences (ISAS), using its new 64-m beam waveguide antenna at Usuda, Japan (4). Using one of the DSN’s low-noise microwave receivers installed on the Usuda antenna to receive a signal from the International Come-tary Explorer (ICE) spacecraft, the researchers made very precise measurements of the microwave losses, or degradation, of the downlink signal.
The results of the experiment were very surprising. The measured losses attributable to the BWG design were much smaller than expected, exhibited similar performance at zenith and better performance at low elevation angles than traditional antennae, and confirmed the efficacy of the BWG configuration.
Encouraged by this field demonstration, researchers sponsored by the Advanced Systems Program moved forward with the construction of a prototype BWG antenna for potential application in the Network.
This new prototype BWG antenna, built at the Venus site at Goldstone, replaced an aging 26-m antenna that had served for many years as a field test site for technology research and development (R&D) programs. The designers used microwave optics analysis software, an evolving product of the technology program, to optimize the antenna for operation across a wide range of current and future DSN operating frequency bands. When completed, the antenna successfully demonstrated its ability to operate effectively in the S band, X band, and Ka band (approximately 2, 8, and 32 GHz, respectively).
Figure 3 shows the completed BWG antenna, and Fig. 4 shows the interior of the equipment room below the antenna structure.
The various frequencies and modes of operation for the BWG antenna were selected by rotating the single microwave mirror at the center of this room. Lessons learned by Advanced Systems Program personnel in constructing and evaluating this antenna were incorporated into the design of the operational BWG antennae for the rest of the Network; the result was that their performance somewhat exceeded that of the prototype, especially at lower frequencies.
Figure 3. The 34-meter beam waveguide antenna at DSS 13.
Forward Command/Data Link (Uplink)
The large antennae of the DSN are used for transmitting radio signals carrying instructions and data to the spacecraft, as well as for receiving signals back from them. Getting data to distant spacecraft safely and successfully requires transmitting substantial power from the ground, directed in a narrow beam toward the spacecraft. For most “normal” situations, the compatible design of spacecraft and the DSN is such that power of about 2 to 20 kW is adequate. However, situations in space are not always normal. Unexpected events can redirect a spacecraft’s main antenna away from Earth, leaving only a low-gain or omnidirectional antenna capable of receiving anything from Earth. Transmitter power up to 400 kW in the S band can be sent from the 70-m antenna during attempts to regain contact with a spacecraft in an emergency.
Figure 4. Stationary equipment room below BWG Antenna.
The initial design and evaluation of R&D models of the high-power transmitters and their associated instrumentation were carried out under the Advanced Systems Program. Much of the essential field testing was carried out as part of the planetary radar experiments. This cooperative and productive arrangement provided a realistic environment for testing without exposing an inflight spacecraft to an operationally unqualified uplink transmission. Later, DSN engineers implemented fully qualified operational versions of these transmitters in the Network at all sites.
Pointing the narrow forward link signal to the spacecraft is critical, especially when making initial contact without having received a signal for reference, as is typical of emergencies. The beam width of the signal from the 70-m antenna in the S band is about 0.030°, and that of the 34-m antenna in the X the band is about 0.017°. Achieving blind pointing to that precision requires thorough understanding of the mechanics of the antenna, including the effects of gravity and wind on the dish, specifics of the antenna bearing and positioning mechanisms, as well as knowledge of the spacecraft and antenna positions, atmospheric refraction, and other interferences.
Forward link data delivered to a spacecraft, if incorrectly interpreted, may cause that spacecraft to take undesirable actions, including some that could result in an emergency for the spacecraft. To guard against that possibility, the forward link signal is coded with additional redundant data that allow the spacecraft data system to detect or correct any corruption in that signal. Operating on the presumption that it is always better to take no action than an erroneous one, the forward link decoder accepts only data sets for which the probability of error is extremely small and discards those that cannot be trusted (5).
Return Telemetry/Data Link (Downlink)
Throughout the Network, the stations use the same antennae for both the forward link and the return data link signals. Because the strength of a signal decreases as the square of the distance it must travel, these two signals may differ in strength by a factor of 1024 in a single DSN antenna. Isolating the return signal path from interference by the much stronger forward signal poses a significant technical challenge. Normally, these two signals differ somewhat in frequency, so at least a part of this isolation can be accomplished via dichroic or frequency selective reflectors. These reflectors consist of periodic arrays of metallic/dielectric elements tuned for the specific frequencies that either reflect or pass the incident radiation. These devices must be frequency selective, and they also must be designed to minimize the addition of extraneous radio noise picked up from the antenna and its surroundings, which would corrupt the incredibly weak signals collected by the antenna from the desired radio source in deep space.
The DSN Advanced Systems Program developed the prototypes for almost all the reflectors of this type currently used in the Network. As an adjunct to this work, powerful microwave analytical tools that can be used to affix design details for almost any conceivable dichroic reflector applicable to the frequency bands of the DSN were also developed under the Program.
Low-Noise Amplifiers. The typical return data link signal is incredibly small and must be amplified before it can be processed and the data itself reconstructed. The low-noise amplifiers that reside in the antennae of the DSN are the most sophisticated in the world and provide this amplification while adding the least amount of noise of any other such devices.
Known as traveling-wave masers (TWMs), the quietest (in adding radio noise) of these operational devices amplify signals that are propagated along the length of a tuned, ruby crystal. Noise in a TWM depends on the physical temperature of the crystal; those in the DSN operate in a liquid helium bath at 4.2 K. (Zero K is equivalent to a temperature of — 273.18°C. Therefore, the temperature of the helium bath is equivalent to approximately — 269°C) Invented by researchers at the University of Michigan, early development of practical amplifiers for the DSN was carried out under the DSN Advanced Systems Program (6). The quietest amplifiers in the world today (Fig. 5), which operate at a temperature of 1.2 K, were developed by the DSN Advanced Systems Program and demonstrated at the Technology Development Field Test Site, DSS 13 (7).
Some of the low-noise amplifiers in the DSN today are not TWMs but are a special kind of transistor amplifier using high-electron mobility transistors (HEMTs) in amplifiers cooled to a temperature of about 15 K (8).
Developed initially at the University of California at Berkeley, these amplifiers were quickly adopted by the scientific community for radio astronomy applications. This, in turn, spawned the JPL development work that was carried out via collaboration involving JPL, radio astronomers at the National Radio Astronomy Observatory (NRAO), and device developers at General Electric (9,10). This work built on progress in the commercial sector with uncooled transistor amplifiers. In the 2-GHz DSN band, the cooled HEMT amplifiers were almost as noise-free as the corresponding TWMs, and the refrigeration equipment needed to cool the HEMTs to 15 K was much less troublesome than that for TWMs. Primarily for this reason, current development efforts in the DSN are focused on improving the noise performance of HEMT amplifiers for higher DSN frequency bands.
The first DSN application of cooled HEMT amplifiers came with outfitting the NRAO Very Large Array (VLA) in Socorro, New Mexico, for collaborative support of the Voyager-Neptune encounter (9). The VLA, designed to map radio emissions from distant stars and galaxies, consists of 27 antennas, each 25 m in diameter, arranged in a triaxial configuration. Within the funding constraints, only a small part of the VLA could be outfitted with TWMs, whereas HEMTs for the entire array were affordable and were expected to give an equivalent sensitivity for the combined full array. In actuality, technical progress with the HEMTs during the several years taken to build and deploy the needed X-band (8 GHz) amplifiers resulted in better performance of the fully equipped VLA than would have been possible with the VLA partially equipped with more expensive TWMs. Since that time, many of the DSN operational antennae have had cooled HEMT amplifiers installed for the 2- and 8-GHz bands.
Figure 5. Ultra-low-noise amplifier at DSS 13.
Phase-Lock Tracking. Once through the first stages of processing in low-noise amplifiers, there are still many transformations needed to convert a radio signal from a spacecraft into a replica of the data stream originating on that spacecraft. Some of these transformations are by nature analog and linear, and others are digital with discrete quantization. All must be performed with virtually no loss in fidelity in the resultant data stream.
Typically, the downlink signal consists of a narrowband ”residual carrier” sine wave, together with a symmetrical pair of modulation sidebands, each of which carries a replica of the spacecraft data. (Specifics of the signal values vary greatly but are not essential for this general discussion.) If this signal is cross-correlated with a pure identical copy of the residual carrier, the two sidebands will fold together, creating a baseband signal that contains a cleaner replica of the spacecraft data than either sideband alone. Of course, such a pure copy of the carrier signal does not exist, but must be created, typically via an adaptive narrowband filter known as a phase-locked loop (11). The re-created carrier reference is thus used to extract the sidebands. The strength of the resultant data signal is diminished to the extent that this local carrier reference fails to be an identical copy of the received residual carrier. Noise in the spectral neighborhood of the received residual carrier and dynamic variations in the phase of the carrier itself limit the ability to phase lock the local reference to it.
These dynamic variations are dominated by the Doppler effect due to the relative motion between a distant spacecraft and the DSN antenna on the surface of spinning Earth. Over the years, the DSN Advanced Systems Program has contributed significantly to the theory and design practices for phase-locked loops and to the impacts of imperfect reference tracking on phase-coherent communications (12,13).
Synchronization and Detection. Further steps in converting a spacecraft signal into a replica of the spacecraft data stream are accomplished by averaging the signal across brief intervals of time that correspond to each symbol (or bit) transmitted from the spacecraft and by sampling these averages to create a sequence of numbers, often referred to as a ”symbol stream.” These averages must be precisely synchronized with the transitions in the signal as sent from the spacecraft, so that each contains as much as possible of the desired symbol and as little as possible of the adjacent ones. Usually, a subcarrier, or secondary carrier, is employed to shape the spectrum of the spacecraft signal, and it must be phase-tracked and removed before final processing of the data themselves. The Network contains several different generations of equipment that perform this stage of processing. Designs for all of these have their roots in the products of the DSN Advanced Systems Program. The oldest equipment is of a design developed in the late 1960s. This equipment was mostly analog and, while still effective, was subject to component value shifts with time and temperature and thus, required periodic tending and adjustments to maintain desired performance.
As digital devices became faster and more complex, it became possible to develop digital equipment that could perform this stage of signal processing. Digital demodulation techniques were demonstrated by the Advanced Systems Program in the early 1970s in an all-digital ranging system. Similar techniques were subsequently employed for data detection in the second generation of the Demodulator-Synchronizer Assembly (14,15).
A Digital Receiver. Rapidly evolving digital technology in the 1980s led researchers to explore the application of digital techniques to various complex processes found in receiving systems such as those used in the Network. The processes of filtering, detection, and phase-lock carrier tracking, formerly based on analog techniques, were prime candidates for the new digital technology.
In this context, the Advanced Systems Program supported the development of an all-digital receiver for Network use. Known as the Advanced Receiver (ARX), the developmental model embodied most of these new ideas and demonstrated capabilities far exceeding those of the conventional analog receivers then installed throughout the Network (16,17).
Encouraged by the performance of the laboratory model, an engineering prototype was built and installed for evaluation in an operational environment at the Canberra, Australia, tracking complex. Tests with the very weak signal from Pioneer 10 spacecraft, then approaching the limits of the current DSN receiving capability, confirmed the design’s performance and significantly extended the working life of that spacecraft.
As a result of these tests, the DSN decided to implement a new operational receiver for the Network that would be based on the design techniques demonstrated by the ARX. The new operational equipment, designated the Block V receiver (BVR), would include all of the functions of the existing receiver in addition to several other data processing functions, such as demodulation and synchronization, formerly carried out in separate units.
As the older generation receivers were replaced with all-digital BVR equipment, the Network generally improved weak signal tracking performance and operational reliability. The receiver replacement program was completed throughout the Network by 1998.
Encoding and Decoding. Data generated by scientific instruments must be reliably communicated from the spacecraft to the ground, despite the fact that the signal received is extremely weak and the ground receiver corrupts the signal with additive noise. Even with optimum integration and threshold detection, individual bits usually do not have adequate signal energy to ensure error-free decisions. To overcome this problem, structured redundancy (channel encoding) is added to the data-bit stream at the spacecraft. Despite the fact that individual ”symbols” resulting from this encoding have even less energy at the receiver, the overall contextual information, used properly in the decoding process on the ground, results in more reliable detection of the original data stream.
High-performance codes to be used for reliable data transfer from spacecraft to DSN were identified by research performed under the Advanced Systems Program and adopted for standard use in the Network, while the search for even more powerful and efficient codes continued. New, more efficient, block codes that used the limited spacecraft transmitter power better by avoiding the need to transmit separate synchronizing signals were developed and first demonstrated on the Mariner 6 and 7 spacecraft in 1969 (18). By putting the extra available spacecraft transmitter power into the data-carrying signal, the new block code enabled the return of Mars imaging data at the astonishing (for the time) rate of 16,200 bits per second, an enormous improvement over the 270 bits per second data rate for which the basic mission had been designed. Of course, conversion of the encoded data stream back to its original error-free form required a special decoder. The experimental block decoder developed for this purpose formed the basis for the operational block-decoders implemented in the Network as part of the Multimission Telemetry System, shortly thereafter (19).
While the JPL designers of the Mariner spacecraft were pursuing the advantages of block-coded data, the designers of the Pioneer 9 spacecraft at the NASA Ames Research Center (ARC) were looking to very complex convolutional codes to satisfy their scientists. The scientists agreed to accept intermittent gaps in the data, caused by decoding failure in exchange for the knowledge that successfully decoded data would be virtually error-free. In theory, a convolutional code of length k = 25 would meet the requirement, but it had a most significant drawback. The decoding process was (at the time) extraordinarily difficult.
Known technically as ”sequential decoding,” this was a continuous decoding operation rather than the ”one block at a time” process used by the DSN for decoding Voyager data.
Originally, it was planned to perform the decoding operation for Pioneer 9 in nonreal time at ARC, using tape-recorded data provided by the DSN. However, Pioneer engineers, working in conjunction with the DSN Advanced Systems Program, explored and demonstrated the potential for decoding this code in real time via a very high-speed engineering model sequential decoder (20). The rapid evolution in the capability of small computers made it apparent that decoding Pioneer’s data in such computers was both feasible and economical. Subsequent implementation of sequential decoding in the Network was done via microprogramming of a small computer, guided by the knowledge gained via the efforts of the technology development program. The subsequent Pioneer 10 and Pioneer 11 spacecraft flew with a related code of length k = 32 and were supported by the DSN in a computer-based decoder.
The DSN standard code, flown on Voyager and Galileo, consisted of a short convolutional code that was combined with a large block-size Reed-Solomon code (21,22). The standard algorithm for decoding convolutional codes was devised in consultation with JPL researchers and demonstrated by simulations. Prototypes of the decoding equipment were fabricated and subsequently demonstrated at JPL.
The application of coding and decoding technology in the DSN was paced by the evolution of digital processing capability. At the time of the Voyager design, a convolutional code of length k = 7 was chosen as a compromise between performance and decoding complexity that would grow exponentially with code length. Equipment was implemented around the DSN to handle this code for Voyager and subsequently for Magellan, Galileo, and others. But modern digital technology permitted constructing much more complex decoders, so a code of length k = 15 was devised (23). This code was installed as an experiment on the Galileo spacecraft shortly before its launch. The corresponding prototype decoder was completed soon afterward. Though not used for Galileo because of its antenna problem, the more complex decoder was implemented around the Network to support the Cassini and subsequent missions.
Research in the technology development program provided the theoretical understanding to predict the performance of these new codes. Figure 6 displays the reliability of the communication error (actually, the probability of erroneous data bits), as it depends upon the spacecraft signal energy allocated to each data bit for uncoded communication and several different codes. In Fig. 6, the first set of curves shows the Voyager k = 7 code, both alone and in combination with the Reed-Solomon code. The second set of codes illustrates the k = 15 code, which was to be demonstrated with the Galileo’s original high-rate channel, shown alone and in combination with the Reed-Solomon code, either as constrained by the Galileo spacecraft data system (I = 2), or in ideal combination. The third set shows the k = 14 code, devised by the Advanced Systems Program researchers for the actual Galileo low-rate mission, both alone and in combination with the selected variable-redundancy Reed-Solomon code and a complex four-stage decoder. The added complexity of the codes, which has its greatest effect in the size of the ground decoder, clearly provides increased reliability in correct communication.
Figure 6. Telemetry communication channel performance for various coding schemes.
Research on new and even more powerful coding schemes such as turbo codes continued to occupy an important place in the Advanced Systems Program. Turbo codes are composite codes made up of short constraint length convolu-tional codes and a data stream interleaver (24,25). The decoding likewise consists of decoders for the simple component codes, but use iterative sharing of information between them. These codes that push hard on the fundamental theoretical limits of signal detection can result in almost a full decibel of performance gain over the best previous concatenated coding systems.
Data Compression. Source encoding and data compression are not typically considered part of the DSN’s downlink functions, but the mathematics that underlie coding and decoding are a counterpart of those that guide the development of data compression. Simply stated, channel encoding is the insertion of structured redundancy into a data stream, whereas data compression is finding and removing of intrinsic redundancy. Imaging data are often highly redundant and can be compressed by factors of at least 2, and often 4 or more, without loss in quality. For Voyager, the combined effect of a very simplified image compression process, constrained to fit into available onboard memory, and the corresponding changes to the channel coding, was about a factor-of-2 increase in the number of images returned from Uranus and Neptune.
The success of data compression technology in enhancing the data return from the Voyager missions firmly established the technique as an important consideration in designing all future planetary downlinks. The original telecommunication link design for the Galileo spacecraft used data compression to double almost the amount of imaging data that the spacecraft could transmit from its orbital mission around Jupiter. The failure of the spacecraft’s high-gain antenna before Galileo’s arrival at Jupiter prompted an intense effort to find even more complex data compression schemes that would recover some of the Jupiter imaging data that otherwise could not have been returned (26).
Arraying of Antennae
The technique of antenna arraying, as practiced in the Deep Space Network, used the physical fact that a weak radio signal from a distant spacecraft received simultaneously by several antennae at different locations is degraded by a component of radio noise that is independent at each receiving station. By contrast, the transmitted spacecraft signal is dependent, or coherent, at each receiving site. In theory, therefore, the power of the signal, relative to the power of the noise, or signal-to-noise ratio, (SNR) can be improved by combining the individual antennae so that coherent spacecraft signals are reinforced and independent or noncoherent noise components are averaged.
In practice, this involved a complex digital process to compensate for the time, or phase, delays caused by the different distances between each station and the spacecraft and by the different distances between the various antenna locations and the reference station, where the combining function was carried out. This technique became known as antenna arraying, and the digital processing function that realized the theoretical ”gain” of the entire process was called ”signal combining” (27).
By 1970, conceptual studies had described and analyzed the performance of several levels of signal combining, and two of these schemes, carrier and baseband combining, were of potential interest to the Network. Both techniques involved compensation for the time, or phase, delays due to the various locations of the arrayed antennas. The difference lay in the frequency at which the combining function was performed. ”Carrier” combining was carried out at the carrier frequency of the received signal, whereas ”baseband” combining was carried out at the frequencies of the subcarrier and data signal that modulated it. Each had its advantages and disadvantages, but ”baseband” combining proved easier to implement and was, obviously, tried first.
The ”arraying and signal combining” concept was first developed and demonstrated in 1969 and 1970 by J. Urech, a Spanish engineer working at the Madrid tracking station (28,29). Using signals from the Pioneer 8 spacecraft and a microwave link to connect two 26-m stations located 20 m apart (DSS 61 and DSS 62), he succeeded in demonstrating for the first time the practical application of the principle of baseband combining in the Network. Because of the low baseband frequency of the Pioneer 8 data stream (8 bits per second) and close proximity of the antennae, no compensation for time delay was necessary.
Within the bounds of experimental error, this demonstration confirmed the R&D theoretical estimates of performance gain and encouraged the JPL researchers to press forward with a more complex form of baseband combining at a much higher data rate (117 kilobits per second) in real time.
The demonstration took place at Goldstone in September 1974, using the downlink signals from the Mariner-Venus-Mercury (MVM) spacecraft during its second encounter with the planet Mercury (30). Spacecraft signals from the two 26-m antennae, DSS 12 and DSS 13, were combined in an R&D combiner with signals from the DSS 14 64-m antenna in real time at 117 kbps. The less-than-predicted arraying gain obtained in this demonstration (9 versus 17%) was attributed to small differences in performance between key elements of the several data processing systems involved in the test. Although this experience demonstrated the practical difficulty of achieving the full theoretical gain of an arrayed antenna system and the critical effect of very small variations in the performance of its components, it also established the technical feasibility of baseband arraying of very weak high-rate signals.
In 1977, using the lessons learned from these demonstrations as background, the DSN started to develop an operational arraying capability for the Network. The Voyager 1 and 2 encounters with Saturn in 1980 and 1981, respectively, would be the first to use the arraying in the Network. A prototype baseband real-time combiner (RTC), based on the analysis and design techniques developed by the earlier R&D activity, was completed in the fall of 1978. Designed to combine the signals from DSS 12 and DSS 14 at Goldstone, it was used with varying degrees of success to enhance the signals from the Voyagers at Jupiter in March and July 1979 and the Pioneer 11 encounter with Saturn in August and September of that year.
Like the previous demonstration, this experience emphasized the critical importance of having all elements of the array—receivers, antennae, and instrumentation—operating precisely according to their specified performance capabilities. With this very much in mind, the DSN proceeded to the design of operational versions of the RTC for use at all three complexes to support the Voyager 1 and 2 encounters with Saturn. The operational versions of the RTC embodied many improvements derived from the experience with the R&D prototype. By mid-August 1980, they were installed and being used to array the 64-m and 34-m antennae at all three complexes, as Voyager 1 began its far-encounter operations. During this period, the average arraying gain was 0.62 dB, that is, about 15% greater than that of the 64-m antenna alone. While this was “good,” improvement came slowly as more rigorous control and calibration measures for the array elements were instituted throughout the Network. By the time Voyager 2 reached Saturn in August 1981, these measures, supplemented by additional training and calibration procedures, had paid off. The average arraying gain around the Network increased to 0.8dB (approximately 20%), relative to the 64-m antenna alone, clearly a most satisfactory result and the best up to that time. Antenna arraying had become a permanent addition to the capability of the Network.
While researchers working within the Advanced Systems Program continued to explore new processes for arraying antennas, engineers within the DSN took advantage of the long flight time between the Voyager Saturn and
Uranus encounters to refine the existing RTC configuration. During the next 5 years, the formerly separate data processing functions of combining, demodulation, and synchronization were integrated into a single assembly, and performance, stability, and operational convenience were improved. By the time Voyager 2 approached Uranus in 1985, the new baseband assemblies (BBAs), as they were called, had been installed at all three complexes. In addition, a special version of the basic four-antenna BBA was installed at the Canberra Complex. This provided for combining the Canberra array of one 64-m and two 34-m antennae with signals from the 64-m Parkes Radio Telescope, 200 km distant (Fig. 7).
In January 1986, this arrangement was a key factor in the successful return of Voyager imaging data from the unprecedented range of Uranus (20 AU). But even greater achievements in antenna arraying lay ahead.
In 1989, the DSN used a similar arrangement with great success to capture the Voyager imaging data at a still greater range—from Neptune (30 AU). This time the Goldstone 70-m and 34-m antennas were arrayed with the 27 antennas of the Very Large Array (VLA) (Fig. 8) of the National Radio Astronomy Observatory at Socorro, New Mexico (31).
Figure 7. Parkes Radio Telescope, Australia.
Figure 8. The Very Large Array (VLA) of the National Radio Astronomy Observatory (NRAO) at Socorro, New Mexico.
DSN support for the Voyager encounter with Uranus was further augmented by the Canberra-Parkes array in Australia, which the DSN had reinstated with the addition of new BBAs, new 34-m antennae, and the upgraded 70-m antenna.
The success of these applications of the multiple antenna arraying technique provided the DSN with a solid background of operational experience. The DSN drew heavily on this experience a few years later when it was called upon to recover the scientific data from Galileo after the failure of the spacecraft’s high-gain antenna in 1991. Together with the data compression and coding techniques discussed earlier, the Network’s Canberra/Parkes/Goldstone antenna arrays succeeded in recovering a volume of data that, according to the Galileo project, was equivalent to about 70% of that of the original mission.
With time, arraying of multiple antennae within complexes, between complexes, or between international space agencies, came into general use to enhance the downlink capability of the Network. In the latter years of the century, most of the enhancements in arraying in the DSN were driven by implementation and operational considerations, rather than by new technology, although the Advanced Systems Program continued to explore the boundaries of performance of various alternative arraying architectures and combining techniques.
Radio-Metric Techniques
In addition to being able to exchange forward and return link data with an exploring spacecraft, it is equally important to know the precise location of the spacecraft, its speed, and direction (velocity). Information about the position and velocity of the spacecraft can be extracted from the one-way or two-way radio signals passing between the spacecraft and the DSN. When these data are extracted by appropriate processing and further refined to remove aberrations introduced by the propagative medium along the radio path between spacecraft and Earth, it can be used for spacecraft “navigation.”
Radio-metric techniques similar to those used for spacecraft navigation can also be used for more explicit scientific purposes, notably radio science, radio astronomy, and radio interferometry on very long baselines (VLBI).
Since its inception, the DSN Advanced Systems Program has worked to develop effective radio-metric tools, techniques, observing strategies, and analytical techniques that furthered the DSN preeminence in these unique fields of science. More recently, the program demonstrated the application of Global Positioning System (GPS) technology to refine further radio-metric data generated by the DSN.
Doppler and Range Data. If Earth and the spacecraft were standing still, the time for a radio signal to travel from Earth to the spacecraft and back would be a measurement of the distance between them. This is referred to as the round-trip light time (RTLT). However, because Earth and the spacecraft are both in motion, the RTLT contains both position and velocity information, which can be disentangled through multiple measurements and suitable analysis. The precision of such measurements is limited by the precision at which one can attach a time-tag marker to the radio signals and by the strength of the signal in proportion to the noise mixed with it, or by the signal-to-noise ratio (SNR).
Precise measurements of changes in this light time are far easier to obtain by observing the Doppler effect resulting from the relative motions. Such measurements are mechanized via the phase-locked loops in both spacecraft and ground receivers using the spacecraft’s replica of the forward link residual carrier signal to generate the return link signal and counting the local replica of the return link residual carrier against the original carrier for the forward link signal. The raw precision of these measurements is comparable to the wavelength of the residual carrier signal, that is, a few centimeters for an X-band signal (8 GHz). Numerous error sources tend to corrupt the accuracy of the measurement and the inferred position and velocity of the spacecraft derived there from. The observed Doppler contains numerous distinct contributions, including the very significant component due to the rotation of Earth. As Earth turns, the position of any specific site on the surface describes a circle, centered at the spin axis of Earth, falling in a plane defined by the latitude of that site. The resultant Doppler component varies diurnally with a sinusoidal variation, which is at its maximum positive value when the spacecraft is first observable over the eastern horizon, and is at its corresponding negative value as it approaches the western horizon. A full-pass Doppler observation from horizon to horizon can be analyzed to extract the apparent spacecraft position in the sky, although the determination is somewhat weak near the equatorial plane. Direct measurements of the RTLT are useful for resolving this difficulty (32).
Three distinct generations of instruments, designed to measure the RTLT, were developed by the Advanced Systems Program and used in an ad hoc fashion for spacecraft support before a hybrid version was designed and implemented around the DSN (33-35). The third instrument designed, the Mu-II Ranging Machine, was used with Viking Landers in a celestial mechanics experiment, which provided the most precise test, until that time, of the general theory of relativity (36).
These devices function by imposing an additional “ranging” modulation signal on the forward link, which is copied on the spacecraft (within the limits imposed by noise) and then imposed on the return link. The ranging signal is actually a very long-period coded sequence that provides the effect of a discrete time tag. The bandwidth of the signal is of the order of 1 MHz, giving the measurement a raw precision of a few hundred meters, resolvable with care to a few meters. Among other features, the Mu-II Ranging Machine included the first demonstrated application of digital detection techniques that would figure strongly in future developments for the DSN.
Timing Standards. The basic units of measurement for all radio-metric observations, Doppler or range, derive from the wavelength of the transmitted signal. Uncertainties or errors in the knowledge of that wavelength are equivalent to errors in the derived spacecraft position. The need for accurate radio metrics has motivated the DSN to develop some of the most precise, most stable frequency standards in the world. Although the current suite of hydrogen maser frequency standards in DSN field sites was built outside of JPL, the design is the end product of a long collaboration in technology development; research units were built at JPL and elsewhere (37,38).
Continued research for improved frequency standards resulted in the development of a new linear ion trap standard (Fig. 9) that offered improved long-term stability of a few parts in 1016, as well as simpler and easier maintenance than that required by hydrogen masers (39).
Earth’s Rotation and Propagative Media. Radio-metric Doppler and range data enable determining the apparent location of a spacecraft relative to the position and attitude of rotating Earth. Earth, however, is not a perfectly rigid body at constant rotation, but contains fluid components as well, which slosh about and induce variations in rotation of perhaps a few milliseconds per day.
Figure 9. New linear ion trap frequency standard.
Calibration of Earth’s attitude is necessary so that the spacecraft’s position in inertial space can be determined, which is a necessary factor in navigating it toward a target planet. Such calibration is available via the world’s optical observatories, and with greater precision via radio techniques, which will be discussed further in later sections titled “VLBI and Radio Astronomy” and ”Global Positioning System.”
The interplanetary media along the signal path between Earth and the spacecraft affect the accuracy of Doppler and range observations. The charged ions in the tenuous plasma that spreads out from the Sun, known as the solar wind, slightly bend and delay the radio signal. Likewise, charged ions in Earth’s own ionosphere and water vapor and other gases of the denser lower atmosphere bend and delay the radio signal. All of these factors are highly variable because of other factors, such as the intensity of solar activity, season, time of day, and weather. All factors must be calibrated, modeled, or measured to achieve the needed accuracy. Over the years, the DSN Advanced Systems Program has devised an increasingly accurate series of tools and techniques for these calibrations (37-43).
Radio Science. Radio science is the term used to describe the scientific information obtained from the intervening pathway between Earth and a spacecraft by using radio links. The effects of the solar wind on the radio signal path interfere with our efforts to determine the location of a spacecraft, but if the relative motions of Earth and a spacecraft are modeled and removed from the radio-metric data, much of what remains is information about the solar wind and, thus, about the Sun itself. Other interfering factors are also of scientific interest.
In some situations, the signal path passes close by a planet or other object, and the signal itself is bent, delayed, obscured, or reflected by that object and its surrounding atmosphere. These situations provide a unique opportunity for scientists to extract information from the signal about object size, atmospheric density profiles, and other factors not otherwise observable. Algorithms and other tools devised to help calibrate and remove interfering signatures from radio-metric data for use in locating a spacecraft often become part of the process for extracting scientific information from the same radio-metric data stream. The precision frequency standards, low-noise amplifiers, and other elements of the DSN are key factors in the ability to extract this information with scientifically interesting accuracy. Occasionally, engineering model equipment is placed in the Network in parallel with operational instrumentation for ad hoc support of metric data gathering for some unique scientific event.
The effects of gravity can also be observed by means of the radio link. Several situations are of interest. If the spacecraft is passing by or in orbit about an object that has a lumpy uneven density, that unevenness will cause a variation in the spacecraft’s pathway that will be observable via radio-metric data. If the radio signal passes near a massive object such as the Sun, the radio signal’s path will be bent by the intense gravity field, according to the theories of general relativity. And in concept, gravitational waves (a yet-to-be observed aspect of gravity field theory) should be observable in the Doppler data from a distant spacecraft (44).
VLBI and Radio Astronomy. The technical excellence of the current DSN is, at least in part, a result of long and fruitful collaboration with an active radio astronomy community at the California Institute of Technology (Caltech) and elsewhere. Many distant stars, galaxies, and quasars are detectable by the DSN at radio frequencies. The furthest of them are virtually motionless when viewed from Earth and can be considered a fixed-coordinate system to which spacecraft and other observations can be referenced. Observations relative to this coordinate set help to reduce the distorting effects of intervening material in the radio signal path and uncertainties in the exact rotational attitude of Earth during spacecraft observations.
Little precise information can be extracted by observing these objects one at a time and from a single site, but concurrent observation at a pair of sites will determine the relative position of the two sites referenced to the distant object. The observing technique is known as very long baseline interferometry (VLBI). If three sites are used in VLBI pairs and multiple objects are observed, the positional attitude of Earth and the relative positions of the observed objects can be determined. If one of the observed items is a spacecraft transmitting a suitable signal, its position and velocity in the sky can be very accurately defined (45). A demonstration of this technique led to operational use for spacecraft such as Voyager and Magellan.
VLBI can also be used in conjunction with conventional radio-metric data types to provide calibration for the positional attitude of Earth. Such observations can be made without interfering with spacecraft communication, except for the time use of the DSN antennae. In addition to determining the attitude of Earth, the observations measure the relative behavior of the frequency standards at the widely separated DSN sites, and thus help to maintain their precise performance.
The DSN equipment and software needed for VLBI signal acquisition and signal processing (correlation) was designed and developed in a collaboration involving the Advanced Systems Program, the operational DSN, and the Caltech radio astronomy community. The tools needed to produce VLBI metric observations for the DSN were essentially the same as those for interferometric radio astronomy. Caltech was funded by the National Science Foundation for this activity. Both Caltech and the DSN shared in the efforts of the design and obtained products that were substantially better than any that they could have been obtained independently.
Another area of common interest between the DSN and the radio astronomy community is that of precision wideband spectral analysis. Development efforts produced spectral analytical tools that have been employed by the DSN in spacecraft emergencies and in examining the DSN’s radio interference environment and have served as preprototype models for equipment for the DSN (46,47). Demonstration of the technical feasibility of very wide band spectral analysis and preliminary observations by a mega channel spectrum analyzer fielded by the Advanced Systems Program helped establish the sky survey planned as part of the former SETI (Search for Extraterrestrial Intelligence) Program.
Another technique (one similar to using of VLBI as a radio-metric reference) is used if two spacecraft are flown to the same target. The second can be observed relative to the first and can provide better target-relative guidance, once the first has arrived at the target.
The Global Positioning System. The Global Positioning System (GPS) is a constellation of Earth-orbiting satellites designed (initially) to provide for military navigation on Earth’s surface. Research under the Advanced Systems Program showed that these satellites could provide an excellent tool to calibrate and assist in radio-metric observations of distant spacecraft. GPS satellites fly above Earth’s atmosphere and ionosphere in well-defined orbits, so that their signals can be used to measure the delay through these media in a number of directions. Using suitable modeling and analysis, these measurements can be used to develop the atmospheric and ionospheric calibrations for the radio path to a distant spacecraft (48).
Additionally, because GPS satellites are in free orbit about Earth, their positions are defined relative to the center of mass of Earth, not its surface. They provide another method for observing the uneven rotation of Earth.
GPS techniques can also be used to determine the position of an Earth-orbiting spacecraft relative to the GPS satellites, as long as the spacecraft carries a receiver for the GPS signals. GPS signals were subsequently used by the TOPEX/POSEIDON Project for precise orbit determination and consequent enhancement of its scientific return (49).
Goldstone Solar System Radar
The Goldstone Solar System Radar (GSSR) is a unique scientific instrument for observing nearby asteroids, the surfaces of Venus and Mars, the satellites of Jupiter, and other objects in the solar system. Although the GSSR uses the DSN 70-m antenna for its scheduled observing sessions, its receiving, transmitting, and data processing equipment is unique to the radar program. The GSSR is a product of many years of development. In the early days of the DSN, the Advanced Systems Program took ownership of the radar capability at the DSN’s Goldstone, California, site and evolved and nurtured it as a vehicle for developing and demonstrating many of the RF and signal processing capabilities that the Network would need elsewhere.
Scientific results abounded as well, but were not its primary product. Timely development of DSN capabilities was the major result. Preparations for radar observations of asteroids at the DSN Technology Development Field Site bore many resemblances to those for a spacecraft planetary encounter, because the radar observations could be successfully accomplished only during the few days when Earth and the radar target were closest together.
In the conventional formulation of the radar sensitivity equations, that sensitivity depends on the aperture, temperature, power, and gain of the system elements. Here, aperture refers to the effective size, or collecting area of the receiving antenna, and temperature is a way of referring to the noise in the receiving system, a lower temperature means less noise. Power refers to the raw power level from the transmitter, and gain is the effective gain of the transmitting antenna, which depends in turn upon its size, its surface efficiency, and the frequency of the transmitted signal. When the same antenna is used both to transmit and receive, the antenna size and efficiency appear twice in the radar equations.
Significant improvements in the DSN’s capability for telemetry reception were to come from the move upward in frequency from S band (2 GHz) to X band (8 GHz) on the large 64-m antennae. Performance of these antennae at higher frequencies and the ability to point them successfully were uncertain, however, and these uncertainties would best be removed by radar observations before spacecraft with X-band capabilities were launched (50). The radar had obvious benefit from the large antenna and the higher frequency. The first flight experiment for X-band communication was carried out on the 1973 Mariner Venus Mars mission. Successful radar observations from the Goldstone 64-m antenna demonstrated that the challenge of operating the large antennae at the higher X-band frequency could be surmounted.
High-power transmitters were needed by the DSN for its emergency forward-link functions, but were plagued by problems such as arcing in the waveguide path when power densities became too high. High-power transmitters were essential for the radar to “see” at increased distances and with increased resolution. Intense development efforts at the DSN Technology Development Field Site could take place without interference or risk to spacecraft support in the Network. Successful resolution of the high-power problems for the radar under the Advanced Systems Program enabled successful implementation of the high-power capability needed by the Network for uplink communications.
Low-noise amplifiers were needed by the DSN to increase data return from distant spacecraft. Low-noise amplifiers are also essential for the radar to enable it to detect echoes from increasingly distant targets or to provide increased resolution of already detectable targets. The synergistic needs of both the radar system and the Network led to development of the extremely low-noise maser amplifiers that became part of the standard operational inventory of the DSN.
Digital systems technology was rapidly evolving during this period and would play an increasing role in the developing DSN. Equipment developed by the Advanced Systems Program for its radar application included (1) digital encoders to provide spatial resolution of parts of the radar echo, (2) computer-driven programmable oscillators to accommodate Doppler effects on the signal path from Earth-to-target-to-Earth, and (3) complex high-speed digital signal processing and spectral analysis equipment. Much of the digital technology developed this way would transfer quickly to other parts of the signal processing work under the Advanced Systems Program and eventually into the operational DSN. Some of the elements would find direct application, such as programmable oscillators, which became essential for maintaining contact with the Voyager 2 spacecraft following a partial failure in its receiver soon after launch. And the signal analysis tools developed to understand and optimize radar would be called on many times over the years to help respond to spacecraft emergencies.
Some of the products of early radar observations were both scientific in nature and essential for providing information for planning and executing of NASA’s missions. One notable “first” was direct measurement of the astronomical unit (51). (One astronomical unit (AU) is equal to 1.5 x 108km , i.e., the mean distance between Earth and the Sun.) It sets the scale size for describing distances in the solar system. The measurement was made to support preparations for Mariner 2 to Venus and provided a correction of 66,000 km from conventional belief at that time. It also enabled corrections that brought the mission into the desired trajectory for its close flyby of the planet. The GSSR was also used in qualifying potential Mars landing sites for the Viking Landers and continues to provide information about the position and motion of the planets, which is used to update the predicted orbits for the planets of the solar system.
Telecommunications Performance of the Network
The progress of deep space communications capability during the 40-year history of the Network is shown in Fig. 10. In Fig. 10, the timeline on the horizontal axis of the figure covers the first 40 years of actual Network operational experience through the close of the century, and extends for a further 20 years to forecast the potential for future improvements through the year 2020. The vertical axis displays the growth in space-to-Earth, downlink capability of the Network. The downlink capability is given in units of the telemetry data rate (bits per second) on a logarithmic scale, and represents the equivalent data rate capability for a typical spacecraft at Jupiter distance (750 million kilometers). Significant events in the history of deep space telecommunications and deep space exploration are appropriately annotated on both axes.
In interpreting the data presented in Fig. 10, it will be observed that the logarithmic scale that displays the data rate gives an impression that the early improvements are more significant than the later improvements because the steps represent fractional or percentage increases, rather than the magnitudes of the actual increases, which are much larger in the later years.
Figure 10. Profile of deep space telecommunications performance: 1960-2020.
Presented this way, however, the figure clearly shows that, from inception though 1997, the downlink capability of the Network grew from an equivalent Jupiter data rate of 10 ~6 bits/s to nearly 10+ 6 bits/s. Note, however, that the drive to switch to “faster-better-cheaper” missions in the late 1990s resulted in decreased capabilities on the spacecraft and eroded the actually implemented capabilities by approximately two orders of magnitude. Nevertheless, that still represents 10 orders of magnitude of improvement.
This remarkable progress is not, of course, solely due to improvements in the Network. Many of the steps result from “cooperative” changes on the part of both the DSN and the spacecraft. Coding, for example, is applied to the data on the spacecraft and removed on Earth. A change in frequency has resulted in some of the larger steps shown by making the radio beam from the spacecraft more narrowly focused. Such a change necessitates equipment changes on both the spacecraft and on Earth.
Other steps represent advances that are strictly spacecraft related, such as increases in return-link transmitter power or increases in spacecraft antenna size, which improves performance by more narrowly focusing the radio beam from the spacecraft. Still other steps depict improvements strictly resulting from the DSN, such as reducing receiving system temperature, increasing the size of ground antennae, or using arrays of antennae to increase the effective surface area available for collecting signal power.
Other Deep Space Network Activities
A number of other DSN activities and developments are worth mentioning. Cost Reduction Initiatives. Pressures have always existed to reduce costs in the DSN. In 1994, a Network automation work area was set up to develop automated procedures that would replace the extremely operator-intensive work of running a DSN tracking station during a spacecraft “pass.” This effort soon produced demonstrable results. A fully automated satellite-tracking terminal for near-Earth satellites was demonstrated in 1994 (52). A software prototype controller that reduced the number of manual inputs for a typical 8-hour track from 900 to 3 was installed at DSS 13 and used to support Ka-band operations at that site (53). Eventually, this technology found its way into the operational Network. A contract for a new, small, deep space transponder offering lower size and power needs, and most importantly lower production costs, was initiated with Motorola in July 1995 (54). This became an element in JPL’s future low-cost (faster-better-cheaper) spacecraft.
Prototypes of a new class of low-cost, fully automated, autonomous ground stations that would simplify implementation and operation and reduce the life-cycle cost of tracking stations in the DSN were introduced in 1995, 1996, and 1997. The first of these terminals was designed for tracking spacecraft in low Earth orbit and was named LEO-T. It was enclosed in a radome and mounted on the roof of a building at JPL where it accumulated more than 2 years of unattended satellite tracking operations without problems. Prompted by the success of LEO-T, the program undertook a fast-track effort to develop a similar automated terminal for deep space applications. It would be called DS-T. The prototype DS-T was to be implemented at the 26-m BWG antenna at Goldstone. Automation technology was carried one step further into the area of Network operations in 1997, when Automated Real-time Spacecraft Navigation (ARTSN) was introduced (55). In addition to the antenna system, the DS-T included an X-band microwave system, a 4-kW transmitter, and an electronics rack containing commercial, off-the-shelf equipment to carry out all baseband telemetry downlink, command uplink, and Doppler and ranging functions. It was planned to demonstrate DS-T with the Mars Global Surveyor early in 1998 and to use this technology (autonomous uplink and downlink) in the Network with the New Millennium DS-1 spacecraft later that year.
Ka-Band Development. In the past, the most significant improvements in the Network’s communication capabilities were made by moving to higher frequency bands. Recognizing this, research and development in the Ka band (32 GHz) was started in 1980. Initial efforts were directed toward low-noise amplifier development and system benefit studies. However, it was also clear that the performance of existing antennas (which were designed for much lower frequencies) would severely limit the improvement in performance that could be realized from the higher operating frequency. Accordingly, in 1991, a new antenna specifically designed for research and development in the Ka band was installed at the DSS 13 Venus site at Goldstone. It was used as the pathfinder for developing large-aperture beam waveguide (BWG) antennae that were, in due course, implemented throughout the Network.
Small imperfections in the surface of an antenna cause larger degradations in the Ka band than at lower frequencies, and, because of the narrower beam width, small pointing errors have a much larger effect. In 1994, improvements in antenna efficiency and in antenna pointing were made in the research and development Ka-band antenna at DSS 13 and in the new operational antenna at DSS 24 (56). These improvements were effected by using microwave holography for precise determination of antenna efficiency and a special gravity compensation system to counteract the effect of gravitational sag as a function of antenna elevation angle. In a search for further downlink improvement, a new feed system consisting of a maximally compact array of seven, circular, Ka-band horns was designed and tested at DSS 13 (57). Each horn was connected to a cryogenically cooled low-noise amplifier, a frequency down-converter, an analog to digital converter, and a digital signal processor. The signals from each horn were optimally combined in a signal processor and presented as a single output whose quality was equivalent to that of a signal from an undistorted antenna. The measured gain in downlink performance was 0.7 dB.
The technology program continued to develop operational concepts that eventually led to the adoption of the Ka band for deep space missions. Trade-off studies between the X band and Ka band showed that the overall advantage of the Ka band, taking due account of the negative effects inherent in its use, was about a factor of 4, or 6 dB, in data return capability (58). Obviously, end-to-end system demonstrations were needed to instill confidence in the new technology. In 1993, the Mars Observer spacecraft carried a non linear element in its transmission feed to produce a fourth harmonic of the X-band signal. The demonstration (called KABLE for KA-band link experiment) provided a weak Ka-band signal for the tracking antenna at DSS 13. A second demonstration (KABLE II) was conducted using the Mars Global Surveyor mission in 1996. This experiment was used to characterize Ka-band link performance under real flight conditions and to validate the theoretical models derived from the studies mentioned above.
KABLE II required additions to the MGS spacecraft radio transponder to generate a modulated Ka-band downlink from which the improvements that had been made to the DSS 13 to improve Ka-band performance could be evaluated. Though the main objective of KABLE II was to evaluate the Ka band for future operational use, it also served as a test bed for new Ka-band technology applications in both flight systems and ground systems.
In the course of transition to simultaneous X-/Ka-band operation, the DSN needed the capability to support various combinations of X- and Ka-band uplinks and downlinks. These included Ka-band receive only, X-/Ka-band simultaneous receive, with or without X-band transmit, and full X-band transmit/ receive simultaneously with Ka-band receive/transmit. The technology program developed new microwave techniques using frequency selective surfaces and feed junction diplexers to provide the frequency and power isolation necessary to realize the performance required by a practical device. In late 1996, a demonstration at DSS 13 showed that these four different modes of operation could coexist on a single, beam waveguide antenna within acceptable performance limits (59). This work provided a viable solution to the problem of simultaneous X/Ka-band operation on a single antenna that would be needed by the operational network to support the Cassini radio science experiment (search for gravitational waves) in 2000.
Recognizing the need for a cheaper, smaller, less power-consuming radio transponder to replace the existing device on future deep space missions, the Advanced Development Program embarked on a joint program with other JPL organizations to develop the Small Deep Space Transponder (SDST). The concept employed microwave monolithic integrated circuits in the RF circuits and application specific integrated circuit (ASIC) techniques to perform digital signal processing functions and a RISC microprocessor to orchestrate overall transponder operation (60). The transponder would transmit coherent X-band and Ka-band downlinks and receive an X-band uplink. Besides minimizing production costs, the principal design drivers were reductions in mass, power consumption, and volume. The Small Deep Space Transponder was flown in space for the first time onboard the DS-1 New Millennium spacecraft in July 1998.
To provide better understanding of the performance of Ka-band links relative to X-band links from the vantage point of a spaceborne radio source, the DSN technology program engaged in developing a small, low Earth orbiting spacecraft called SURFSAT-1. Launched in 1995, the experiment provided an end-to-end test of Ka-band signals under all weather conditions and DSS 13 antenna elevation angles, as the spacecraft passed over Goldstone. The SURF-SAT data was also used for comparison with the KABLE data received from MGS. Later, the SURFSAT X-band and Ka-band downlinks were used to great advantage to test and calibrate the DSN’s new 11-meter antennas, before their support of the VSOP (HALCA) Orbiting VLBI mission in 1996. Optical Communications Development. Beginning in 1980, the technology program supported theoretical analyses that predicted, under certain system and background light conditions typical of deep space applications, the ability to communicate at more than 2.5 bits of information per detected photon at the receiver. Laboratory tests later confirmed these theoretical predictions (61,62). However, detection power efficiency was only one of the many factors that needed to be studied to bring optical communications to reality. Others included laser transmitter efficiency, spatial beam acquisition, tracking and pointing, link performance tools, flight terminal systems design, definition of cost-effective ground stations, and mitigation of Earth’s atmospheric effects on ground stations.
Several system-level demonstrations were carried out as this work progressed. The first, in December 1992, involved detecting a ground-based pulsed laser transmission by the Galileo spacecraft during its second Earth fly by. Transmitted laser signals generated from two ground-based telescope facilities were successfully detected at spacecraft-Earth distances up to 6 million km (63). The second demonstration was carried out during the period November 1995 to May 1996 between JPL’s optical telescope facility at Table Mountain, California, and the Japanese Earth-orbiting satellite ETS VI at geosynchronous Earth-orbit altitudes (40,000 km). Data-modulated transmissions were successfully detected in both uplink and downlink directions at a data rate of 1 Mbps (64).
Both experiments yielded important observational data in support of theoretical studies and encouraged the further development of optical communications technology with follow-on supportive flight demonstrations. DSN Science. Science and technology were always closely coupled in the DSN. Since the very beginnings of the DSN, its radio telescopes provided world-class instruments for radio astronomy, planetary radar, and radio science. Many technology program achievements were of direct benefit to these scientific endeavors, and DSN science activities frequently resulted in new techniques that eventually found their way into the operational Network.
In the period reported here, the program supported radio astronomy investigations related to the formation of stars, to the study of microwave radio emissions from Jupiter, and to radio science measurements of the electron density in the solar plasma outside the plane of the ecliptic (Ulysses spacecraft). The DSN also supported a program of tropospheric delay measurements, which would be of direct benefit to the Cassini gravitational wave experiment. The Goldstone Solar System Radar (GSSR) continued its highly successful series of Earth-crossing asteroid (ECA) observations, which began in 1992 with images of Toutatis (asteroid 4179) and Geographos (asteroid 1620) in 1994 and Golevka (asteroid 6489) in 1995 (65). This work was expected to increase in the years ahead as new and improved optical search programs enabled the discovery of more ECAs.
