Digital Processing Radar
No More Mixed Signals
Adding digital processing to radar—both pulse and solid-state—has created a powerful tool for navigating and safety at sea.
Remember how your TV picture used to go “snowy” when your neighbor started his lawnmower, and that high-pitched buzzing whine that came on the radio when someone nearby switched on a hairdryer or electric drill? They were both examples of electromagnetic interference (EMI)—radio noise pollution that is created as an unwanted byproduct whenever an electric circuit is interrupted or changes direction.
Microprocessors, in particular, are creators of EMI as well as victims of it, so the huge increase in the use of electronic devices means that EMI could easily have become a worldwide pandemic.
The fact that the electronic revolution didn’t grind to a halt in a tidal wave of its own interference is largely due to the determined efforts of engineers and lawmakers around the world, who have devised ways of minimizing its effects, and have set increasingly stringent limits on the levels of EMI that are tolerated.
Even transmitters are affected by the clean-up legislation. Of course, they’re not prevented from transmitting on the frequencies they need to do their jobs, but there is increasing pressure to limit the “spurious emissions” on adjacent frequencies. Until 2003, radar was exempt, but since then, it has been subject to the same increasingly stringent regulations as other transmitters.
One reason for radar’s special status was that almost all marine radars depend on a device called a magnetron. Magnetrons are brilliant at producing intense pulses of high-power microwave energy, but not so good at nailing them to a single frequency. Now that cleaner versions are available though, there’s no justification for treating radar as a special case.
The latest generations of radar are pumping far less pollution into the electromagnetic environment. Even better, they are easier to use than ever before, with clearer pictures and more features. But as new technologies emerge, the question arises: Is there still a place for conventional, pulse radar? The Big Four manufacturers—Furuno, Raymarine, Navico, and Garmin—continue to offer them, even if Navico’s brands—Simrad, Lowrance, and B&G—are hedging.
Replacing dirty magnetrons with clean ones would have conformed with the letter of the law, but Navico took a far more radical step when it introduced Frequency Modulated Continuous Wave (FMCW) Broadband radar in 2008. The latest version from Simrad is the Broadband 4G radar ($2,299), which the company says has a 36-mile range.
Just like conventional marine radar, FMCW radar transmits microwave energy, and listens for the echoes that bounce back to its antenna. But instead of transmitting pulses that last less than a microsecond each (1 millionth of a second) it transmits “sweeps” that last more than a millisecond each (one thousandth of a second). The clever bit about FMCW technology is that it achieves range performance that is comparable with a typical 2-kilowatt magnetron radar, but from a solid-state transmitter with an output of just 165 milliwatts (less than one fifth of a watt).
FMCW offers quite a package of other benefits as well, including no minimum range, improved range discrimination, reduced susceptibility to rain and sea clutter, no warm-up, no tuning, and low radiation.
So why haven’t Furuno or Garmin followed suit? And most surprisingly, perhaps—why haven’t we seen an FMCW radar from Raymarine, whose parent company already produces FMCW radars for military use?
One answer is that—at the moment—it’s not possible to produce an FMCW radar with enough power to achieve the kind of long-range, small-target detection that some users want at prices they are prepared to pay.
The other companies have systems under development, however. “One of the things you’ll see is kind of a melding of our existing digital radar Ethernet sensors that we have out there with solid-state transmitters,” says Eric Kunz, senior product manager at Furuno USA. “You’ll see that from most manufacturers. It’s going to definitely improve system reliability for heavy commercial users. Magnetrons have a limited lifetime—they’re about a 5,000-hour product. Solid-state transmitters last up to 100,000 hours or more.”
The second answer is that the other companies are concentrating their R&D efforts on going digital. “Digital radar” isn’t really new: Digital signal processing (DSP) of raw radar data was an essential part of the early “daylight viewing” radars, 30-odd years ago. What makes the current generation of digital radars different is that back then, digitizing was one of the last steps in the process of converting pulses in the scanner into pixels on the screen. Now, it’s one of the first.
Going digital early in the process has several advantages. The first—though it’s easily overlooked—is that any echo returning to the radar scanner is very weak: It needs to be massively amplified before any other kind of processing can be applied to it. But if you amplify an analog signal, then you also, inevitably, amplify the radio noise that is mixed up with it, as well as adding more noise created within the receiver itself. A digital signal, on the other hand, is easy to amplify without adding noise.
To put it another way: You could say that digitizing early increases the signal-to-noise ratio—almost like increasing the transmitter power.
The more obvious advantage of digitizing though, is that it opens up countless opportunities to enhance the picture. It can even give the illusion that the scanner is considerably bigger than it really is. “With the digital signal processing that we’ve done on the receiver end of our 18-inch broadband radar, we can define targets like a conventional pulse radar—the grunty performance from our 4G dome paints targets like a 3 ½-foot 4-kilowatt open array,” says Dennis Hogan, product manager for Navico America. “It’s very compelling, but not easy to do. We’ve spent years refining our side-lobe suppression and digital processing to reduce noise and increase clarity on the screen.” So it’s not gathering more data, it’s just using it differently.
To see how, we need to accept that a radar beam is a bit like the beam from a car headlight: It’s fairly well defined, but with fuzzy edges. The more carefully you examine it, the more difficult it becomes to decide exactly what is inside or outside of the beam.
Now, imagine the effect of this fuzzy-edged beam sweeping across a small radar target, such as a buoy. A few stray whispers of radar energy start to reach the buoy and are reflected back to the scanner while the center of the beam is still several degrees to the left of the marker. As the scanner rotates, the intensity of the energy hitting the buoy will increase, reaching a maximum when the beam is pointing straight at it, and will then fade away as the beam sweeps around the horizon.
Much the same can be said about “pulse length.” When the specification says your radar has a pulse length of one microsecond, it’s natural to assume that the radar transmits nothing, then transmits at full strength for a microsecond, then stops. But again, that’s an oversimplification. Each pulse starts from nothing, builds to a maximum, and then fades away.
The combined effect of the fuzzy-edged beam and the slightly soft-start pulse means that there is really no such thing as a perfectly crisp, sharp-edged radar contact, even from a good reflector at short range.
Generations of radar operators have learned to live with the problems this produces—with buoys that appear to be as big as container ships, and inlets that don’t appear until you’re within half a mile of the entrance. But now, digital signal processing is cleaning up those smudgy images. In effect, the processor looks at each contact and removes the fuzzy edges to leave only the crisp central heart, producing pictures that look as though they came from a scanner that is twice as big as it really is.
With multiple hardware technologies in the works and new software tweaks, radar keeps evolving. And even though each manufacturer keeps its R&D plans close to the vest, radar will continue to become more useful and better integrated into the helm in whatever form it takes in the future.
This article originally appeared in the March 2013 issue of Power & Motoryacht magazine.