Compromising Stealth?
In mid-October Lockheed-Martin announced a revolutionary new sensor—Silent Sentry—based on the emerging reality that the world is bathed in man-made electromagnetic radiation, such as radio and television signals. Objects reflect or block such radiation, and thus can be detected, much as the eye detects objects by their reflection of surrounding light. The U.S. Navy discovered its version of radar by observing this same effect when passing airplanes affected radio transmissions across the Potomac from the Naval Research Laboratory. It occurred to researchers that the airplanes were reflecting the signals, and that radio could be used to detect them. They proposed an aircraft detection system in which an area of sky would be continuously illuminated, receivers picking up reflections.
This idea was not particularly practical at the time, and radars today still generally operate by sending out signals for reflection because ambient radiation is difficult to use. It covers a very wide range of frequencies, and the strength of the signal at any one frequency may vary radically. Thus the prerequisites for using some form of electromagnetic ambient "light" include a receiver with a very wide bandwidth and a very wide dynamic range, neither very common in the past.
Over the past few years, however, the situation has changed. Digital receivers can handle a much wider range of signals and there is considerable interest in the shared-aperture antennas they make possible. The Lockheed Martin system employs three such receivers to detect objects. That makes sense. A single human eye, observing an object at a distance, sees mainly its bearing. Two eyes can triangulate; three can triangulate in three dimensions.
Lockheed-Martin emphasizes Silent Sentry's entirely passive nature; it does not radiate. It is likely to have a much more interesting capability, however—it may be capable of dealing with stealthy targets. Unlike passive sensors such as electronic support measures (ESM) and infrared systems, it does not require cooperation on the part of the target. Any object must either reflect or absorb the radiation in which it is bathed. If it reflects, inevitably a distributed net of sensors will see it. Stealthy shapes avoid detection only because they reflect incoming radiation away from the illuminator. Silent Sentry sensors, however, operate nowhere near that line of bearing. Moreover, if the array is large enough (three clearly is a minimum), many sensors should pick up reflected energy. If the stealthy object absorbs most radiation, then it should leave, in effect, a shadow; that, too, ought to be readily detectable. Many current mine detection sonars, for example, are designed to switch between image and shadow modes, exactly because some mines are sheathed in sonar-absorbing materials.
The key technology, the wideband antenna, already is under intensive development for air and naval applications apparently unrelated to Silent Sentry. The main reason is that real estate is very limited on board ships and aircraft, and multiple antennas make poor use of it. Thus the Joint Strike Fighter (JSF) is to use common apertures and common front-end processors for all electromagnetic functions: radar, ESM, communications, and navigation (e.g., Global Positioning System reception). The electromagnetic signals it picks up are translated into digital form and placed on a common bus from which the relevant processors pick off the signals they need. These signals exist at widely varying strengths; a returning radar pulse generally will be considerably weaker than an enemy's radar pulse that has traveled only one way to the target (that is one reason ESM antennas can be much smaller than radar antennas). Clearly the Joint Strike Fighter's avionics can handle the difference—just as Silent Sentry must.
A shipboard shared-aperture project, which is based on much the same logic, is ongoing. Sharing already occurs on an involuntary basis, as electromagnetic interference. During the Falklands Conflict, for example, the British had to turn off the ESM set on board HMS Sheffield when she was transmitting via satellite, because the satellite dish operated at a radar frequency and the ESM set was likely to pick it up as an unknown radar. With the ESM set turned off, of course, the ship was unaware of an approaching Argentine airplane, which delivered an Exocet missile.
This problem was solved, at the time, by time-sharing. Satellite transmission was reduced to short bursts, during which the ESM set was turned off automatically. Since the transmissions did not take up a large fraction of the time, a ship had sufficiently continuous ESM coverage. Unfortunately, there is now a strong desire that any transmission be stealthy (low probability of intercept). That is not a fundamental problem: the transmission is spread out in time (so that it can be made much weaker) and in frequency. Unfortunately, time-sharing is no longer at all possible, because the spread-out transmission is effectively continuous. Something more must be done. A shared aperture would seem to promise a solution.
The idea of ambient electromagnetic light recalls a parallel acoustic concept. Some years ago, EDO Inc. proposed that minehunting sonars be replaced by devices using ambient acoustic energy. Normally sonar developers try very hard to exclude this noise from their sets, because it crowds out the weak signals they are trying to detect; ambient noise , however, clearly reflects off underwater objects. Recent interest in explosive echo ranging (for example, in the experimental Deep Thunder system) suggests that a wide frequency range is not a problem for underwater sensors (explosions produce very broad band pulses).
Both above and below water, it is unlikely that systems will produce real images. It is difficult enough to produce an image using a signal with a fairly narrow frequency band. That is one reason the human eye uses only some of the light that is around it, not the infrared or ultraviolet or beyond. At least initially, Silent Sentry probably will be limited to point detectors, which will produce a simple track in three dimensions. That may be quite enough to be very useful.
For example, Silent Sentry would seem to be a passive complement to the Navy's cooperative engagement capability, in which a group of ships exchange detailed radar detection data so as to form consistent target tracks despite interference, e.g., from nearby terrain. They are already exchanging data; could Silent Sentry-type data be added? That might not be particularly useful well out to sea, but areas closer inshore are likely to be brightly illuminated by civilian radars and radios. Aircraft might usefully exchange Silent Sentry data, perhaps using fairly covert data links of their own, such as ultraviolet links.
At the very least, the wide diffusion of Silent Sentry-type technology would seem to doom current efforts to develop stealthy aircraft, which would presumably be quite visible over populated areas, in which ambient illumination would be particularly strong. On the other hand, current means of evading radar, by flying very low and by using terrain aggressively, probably would remain effective. It is still the case, after all, that an airplane popping up suddenly at very short range, from behind a hill, is difficult to counter. The main effect of Silent Sentry is likely to be to make the night seem more like the day.
HMAS Collins Is Noisy
The Royal Australian Navy's diesel submarine Collins recently underwent testing at the U.S. Navy's Atlantic Undersea Test and Evaluation Center and apparently experienced considerable problems. A Royal Australian Navy spokesman later told a press conference that the builder, the Australian Submarine Corporation, would have to spend about $1 billion over the next year to fix the ship. Problems, many of them reported in the past, include incomplete combat system performance and excessive noise.
The recent announcement may owe more to an Australian policy dispute, however, than to the builder's incompetence. Australia had no submarine construction facility when the Royal Australian Navy ordered the Collins class. An industry had to be created. Considerable efforts produced a first-class modern industry—with just six submarines on order, and without good export prospects. The Australian government had absolutely no intention to sell to the likeliest customer, Taiwan. Another prospective customer, Canada, dithered before deciding to accept the four existing British Upholder-class submarines, and it seems unlikely that South Africa, another prospect, can afford anything nearly as sophisticated as a Collins. The new submarine builder, fighting to survive, began to agitate for a shift in Australian policy, away from a surface fleet and toward more submarines. The recent official republication of old reports of problems suggests a political motive, a navy counter-offensive.
On the other hand, the problems are real. Interestingly, the navy is fairly certain that they can be solved fairly quickly, which suggests that many of them are related to software rather than hardware or, rather, to the way the two interact. Military software is increasingly complex, and over the past few years it has caused several serious problems. The British Type 23 system suffered from serious delays caused by the need to fix software (it already had been delayed by the need to develop a system from scratch to replace that originally planned). Malaysia's two new frigates also were delayed.
In these ships, as well as in the Collins class, developers hoped to simplify matters by using multiple computers, all connected to a digital bus—a configuration that should protect any one computer from errors in programs in the others. However, it has two pitfalls. One is the capacity of the bus itself. Computers communicate with each other and with other elements of the combat system by sending and receiving messages via the bus. As the rate of sending and receiving rises, messages may begin to collide in the bus, destroying each other. In the case of the Type 23s, reportedly the system could not react quickly enough. That suggests that the bus was overloaded, and the solution was probably to change the way it operated (as the French did in their new carrier Charles de Gaulle—the bus changes operating mode above a certain level of usage). A second pitfall is that the computers really do interact, because the system generally was designed to survive crashes by some of its computers. In the event a computer collapses, the others reconfigure to do its job as well as their own. That entails a higher degree of coordination than might otherwise be imagined.
Reports concerning the Collins class indicate that the propeller cavitates and cracks blades. Noise may come directly from cavitation, or from blade-rate vibration (which travels up the propeller shaft). At least in the latter case, noise is traceable to interaction between the submarine sail, tail fins, and propeller blades. The Collins design may have suffered because the submarine was scaled up from a much smaller Swedish prototype; the interaction in question may depend on the distance between sail and stern. Blade cracking suggests a particularly severe vibration problem, but that would not necessarily produce noise radiating aft. Much of blade-rate radiation goes up the propeller shaft and emerges from the bow of the submarine, where it might modulate the flow of water, and so make the entire boat much noisier. The solution is likely to be a better propeller design. It also is possible that, as in some Soviet Charlie-class submarines, the sail itself creates vortices that produce noise. The Charlie problem was so bad that paint visibly peeled off the hull where the vortices formed. Eventually, the Soviets fitted these submarines with collars that presumably altered the flow onto their unfortunately shaped or located sails. The problem may have arisen because the Charlie design began as an inexpensive mass-production torpedo submarine. The missiles were added later, and the enlarged bow-section housing them may have affected other features of the design.