Analysis:
Ocean surveillance
Picking up tiny things
Ocean surveillance has become a huge challenge. Space-based assets bring in significant wide-area capability for various activities specifically in regions where maritime traffic is affected by potential threats or sources of illegal actions such as maritime piracy. The sea sanctuary used to be fundamental in naval strategy: once ships were beyond the horizon, they were effectively invisible. They might be picked up from time to time, and reconnaissance aircraft might find them, but the sea was generally a sanctuary. During the Cold War, the superpowers began to erode this sanctuary with space-based systems such as the US Navy’s (USN) White Cloud satellites. Like other once-military technology, the one they employed seems to be moving towards wide commercial availability, with real consequences for the world’s navies. The questions are how badly naval invisibility is being eroded, and what navies can do about the problem. The issue is not the massive number of optical satellites, which seem to keep the whole world under almost continuous observation. Each such satellite actually has a limited capacity for images, and it is typically pointless to waste that capacity over the sea. Individual ships can certainly be recognized from space, but the sea is too vast for any optical system. A ship’s electronic emissions are a very different matter. The emissions a satellite is likely to pick up are unlikely to exhaust its memory.
What is new is the advent of commercial ocean surveillance based on such emissions. Several companies currently offer such capability. The two leaders seem to be HawkEye 360, a commercial satellite company specializing in radio frequency (RF) geo-analytics based in the United States, and UNSEENLABS in France, the latter providing civilian and military clients with the ability to locate ships by detecting and characterizing their passive electromagnetic signature. Australian-listed space technology operator Kleos Space also offers space-based radio frequency (RF) surveillance, but apparently it has not yet gained government contracts. In its advertising, Kleon Space emphasizes its ability to locate maritime activity where it should not be, which suggests that it hopes to sell its data to national law enforcement agencies. At least publicly, it does not claim an ability to identify emissions on a ship-by-ship basis. Horizon Technologies in the United Kingdom plans (as of April 2021) to orbit its first satellite in August 2021. The company claims SEI (specific emitter identification) capability against X- and S-band radars; it already operates an airborne maritime ESM system, called Flying Fish.
Key is direction-finding: A space-based direction-finder defines a direction in space, which pierces the ocean surface only at a particular place. White Cloud employed triplets of satellites. It and successors most likely employ one of two alternative techniques. One is time-of-arrival (ToA). A signal arrives at three different satellites at slightly different times, the relationship among which indicate direction. The other possibility is Doppler. The signal the satellite receives is frequency-shifted because the satellite is moving rapidly over the earth. This type of localisation is closely related to the technique used in early satellite navigation systems. A satellite can invert it to find out where a ship is. It seems likely that a satellite system uses ToA to distinguish pulsed signals such as radars. Doppler would be more useful when dealing with continuous-wave communications systems. HawkEye 360 uses triplets of satellites, which it claims can employ both techniques. The French UNSEENLABS system uses single satellites employing a proprietary technique. That may be some form of Doppler.
A second key issue is the receiver. When White Cloud was conceived, it presumably used tuning receivers that could handle only a limited frequency range. Now, there are wide-band, electronically-tuned receivers (software defined radios) that can synthesize the desired frequency, presumably on command. HawkEye 360 claims that its system can be tuned to cover the range between 144MHz (Megahertz) and 15GHz (Gigahertz), approximately very high frequency (VHF) up to the high end of the ‘window’ through which signals propagate into space. That a satellite can be programmed to listen anywhere in this spectrum does not, however, mean that it is wide open to all of it. Sensitivity is inversely proportional to the instantaneous range covered. The receiver needs time to determine that what it is receiving is a real signal, not noise.
The catalogue is a more difficult proposition. Big data and the cloud make it possible to list hundreds of thousands of emitters and their parameters, and increasingly fast computing makes it possible to search such a list quickly enough to be useful. HawkEye 360 claims that it is assembling a catalogue of many thousands of radars. The issue is how to be sure of which ship emits which signals. During the Cold War, the USN could order its patrol aircraft to swoop down on Soviet ships so as to identify them with their SEI parameters. No private organization has that sort of resource. The closest is probably to call up images from optical satellites of ships with known SEI parameters when they are in harbour. That is probably how HawkEye 360 was able to distinguish Chinese fishing vessels in Ecuadorian waters. It is also possible that it relied on the gross parameters of Chinese-made navigational radars, in which case it did not need SEI data. However, the question for the future is the extent to which the commercial operations are able to track warships in the open sea, using their normal radar emissions. That requires SEI in some form.
The key developments for HawkEye 360, UNSEENLABS and others are the advent of inexpensive satellites, which in turn can be small because of the explosive growth in electronic capabilities. HawkEye 360 has claimed that its SEI-based technique can overcome attempts to evade typical surveillance, which is based on the automatic identification system (AIS) emitters which are mandated for all ocean-going ships. There are already attempts to spoof AIS, and illegals often simply turn off their AIS transponders. Some warships employ spoofing to conceal their operations. HawkEye 360 points out that ships are far less likely to turn off the navigational radars they require for safety at sea. For example, the Chinese fishing boats in Ecuadorian waters had turned off their AIS transponders, but were identifiable by their navigational radars.
Are there viable counters to space-based ocean surveillance? One possibility is to operate navigational radar in a part of the spectrum not easily observable from space, one outside the window in the ionosphere. That would mean going well into K-band, beyond the 15GHz limit HawkEye 360 advertises. That presents some problems, but this sort of millimetre-wave radar is already fairly widely used. Such radars would still be detectable (and trackable) by aircraft, but tracking aircraft would be unlikely to spend much time over the vast reaches of the sea.
Another possibility is to rely on stealthy radars, which use spread-spectrum techniques to limit their observability. The first such radar was Scout, made by Signaal (now Thales Nederland). There are many others. Such radars do not need recognizable antennas, and they may be alternative feeds to conventional navigational sets. Scout has often been described as a frequency modulated (FM) radar, relying on frequency shift rather than on signal strength to detect objects. Many years ago, Canadian Marconi advertised stealthy air search radars, which used pulse compression techniques to limit their peak power. In theory any pulse-compression radar has some stealth aspects.
Another possibility, which the Soviets employed extensively during the Cold War, is to provide radars with ‘war’ modes, i.e. with operating modes never used in peacetime. In theory, a war mode switches the radar signal generator, thus overcoming any SEI technique that exploits details of that signal generator, such as scratches in its magnetron. Such a radar is still vulnerable to an SEI technique that relies on details of the radar antenna.
A variation on this theme would be to reserve an entire navigational radar for emergency use. If the radar were never used except in an emergency, its parameters would not be collected and catalogued. Once it was being used, eventually it would be identified with a particular ship, and that ship would be tracked, but that might take time. It would take considerably longer if the ship used her special radars only when well out to sea, when the chance of identification by optical satellite would be small.
It is also possible that array radars employing large numbers of active elements are inherently difficult to identify using SEI techniques. If that is so, the advent of SEI-based commercial ocean surveillance will push navies to adopt active-element techniques.
For an aircraft carrier, there is another possibility. During the Cold War, it was often pointed out that an airborne early warning airplane could in effect take over all of the radar functions of the carrier and her consorts, beaming down the radar data they needed. The carrier would not have to emit at all, and the radar airplane typically would not be flying directly overhead. Right now, the inherent limitations on ocean surveillance satellites may limit the danger involved. The satellites almost certainly search over limited frequency ranges set by command, probably usually X-band for navigation. The operators’ ability to build and maintain SEI catalogues may also be limited, although the catalogues may be enhanced when they work with major governments.
However, now is the time for navies to recognize that their usual operating environment, well beyond the horizon, may not be so safe as they have imagined.
Norman Friedman, Ph.D., holds a doctorate in theoretical physics from Columbia University, and is a regular contributor to Naval Forces.
UNSEENLABS launched its first nanosatellite into orbit in August 2019, which thanks to its proprietary on-board technology based on the identification of electromagnetic waves emitted by ships, can geolocate from space any ship at sea, in near-real time, to the nearest kilometre.
(Photo: UNSEENLABS)
The recent concentration of Chinese fishing vessels in the vicinity of the Galapagos Islands in the Eastern Pacific Ocean, accused of fishing inside the Ecuadorean economic exclusion zone around the islands, was tracked by an HawkEye 360 commercial satellite system that intercepts radio frequency signals and can detect when a vessel turns off its automatic identification system. Shown is a schematic showing how the system works.
(Photo: Airbus/HawkEye 360)
Space-based surveillance is crucial for law enforcement, border and security authorities, fisheries management agencies and other state end-users to optimize the use of traditional surveillance means such as patrol vessels and aircraft.
(Photo: Airbus)
Kleos scouting mission satellites deliver data collected over key regions of maritime interest for defence and security customers.
(Photo: Kleos Space)
The European Space Agency’s Sentinel-2 satellites provide multispectral images with pixel sizes down to 10m, allowing for fast and frequent detection, classification and discrimination of various objects in the sea.
(Photo: MDPI AG Basel, Switzerland, Peder Heiselberg/Niels Bohr Institute, Copenhagen and Henning Heiselberg/National Space Institute, Technical University of Denmark, Kongens Lyngby)