Integrated radar

A 338
In the 1970s Gordon Moore, a founder of Intel Corp., maker of the Pentium chip, developed a rule of thumb for the evolution of microprocessors. “Moore’s Law” predicts that every 18 months the number of transistors possible on a chip will double. This implies that the cost of transistors is cut in half every 1.5 years, putting ever more capability in our hands for less money.

This Furuno radar is an example of a modern interfaced radar unit. With input from a GPS, it can display lat/long and waypoint data, a fluxgate allows it to display in magnetic north-up mode.
   Image Credit: Courtesy Furuno

The radar market is a prime example of how cheap transistors give us wonderfully sophisticated radar units. Sub-$1,000 radar sets are now common; even the most inexpensive offer fabulous performance and amazing feature mixes.

One feature that is now almost de rigueur is some sort of networking capability. Interconnectedness, the mantra of the '90s, lets us tie our disparate navigational instruments together to give us a clear view of course and hazards in a single display.

Long before the microprocessor age, research ships lugged minicomputers to sea to log and process data acquired from a broad range of instruments. Even in those electronics-deprived days, a simple network linked disparate electronic units together. For example, "SAIL" was a standard used as a communications protocol that predated even the RS-232 standard so common on today's PCs.

Now electronics makes up an ever-increasing percentage of the cost of any modern vessel. Even small production boats sport surprising numbers of instruments. Higher-end yachts often include special masts and arches just for antennas and sensors. Wiring harnesses connect the sensors themselves (radomes, depth transducers, fluxgate senders, etc.) back to displays grouped into locations convenient to the navigator and helmsman.

No instrument is an island, though. Just as a navigator would never use a single point of information to plot a position, each bit of electronic wizardry produces but one piece of information that is part of the whole picture of the boat's position, course, and heading. Today most devices can communicate their bit of navigational intelligence to the rest of the boat via a local area network (LAN). In some casesparticularly for high-end racersthe LAN connects the yacht into the rest of the world via satellite uplinks.Standard interfaceThe NMEA 0183 standard has long been the boating world's LAN of choice due to its low cost and widespread implementation. Like any networking standard, NMEA 0183 defines the physical connection between communicating devices (wiring and connectors), the electrical parameters (voltage, communication speed), and protocol (the lingo, or composition of messages). NMEA 0183 very much resembles the RS-232 interface we use on PCs to connect modems, printers, and the like. It's a single-talker, multiple-listener convention. That is, one instrument on the network is always the source of data; many instruments may listen. At 4,800 bits per second, it's quite slow. This rate is fine for sending small chunks of data like heading info, but far from adequate for transmitting electronic charts and other graphics information.The marine electronics community recognized these limitations and is now in the process of defining a new standard. NMEA 2000 will be about 20 times as fast as its predecessor. Perhaps more important, it greatly resembles the networks used on PCs. Any instrument on the network will be able to listen and/or talk. You'll be able to use a single cable to connect all of the instruments, using a multi-drop configuration rather like the kind we have with thin Ethernet connections today.

Bi-direction communications on one data line greatly simplifies boat wiring but brings along one downside: what happens when two devices start to speak at the same time? The data will be hopelessly garbled. NMEA 2000 will use Carrier Sense, Multiple Access (CSMA) techniques to detect the garbled transmission, and then gracefully designate one device to talk while the others wait their turn.

This bidirectional signaling solves a huge problem associated with today's NMEA 0183 devices. Since only one device can talk, when connecting multiple instruments data sharing is a one-way affair. A GPS can send position data to a radar, but the radar can never send target info back to the GPS display unless dual interfaces are connected.

One alternative to NMEA 0183 is Raytheon's Seatalk. Available today, Seatalk is itself a multidrop CSMA protocol that permits bidirectional communications. Like NMEA 2000, it greatly simplifies boat wiringin fact, all new Hunter sailboats come prewired with a Seatalk bus.

While all Raytheon's equipment is compatible with Seatalk, this is a proprietary protocol and as such is no panacea to networking. Many devices aren't compatible with it, so it pays to check communications requirements carefully.

Networked radar

Today's "smart boat" usually includes at least a GPS and an electronic compass. These two instruments, when they include a NMEA or Seatalk interface, provide a radar set with the most basic and critical information needed to orient the radar in course and position.

Connect a compass to your radar, and suddenly your viewing options multiply. You'll be able to orient the display heading-up (the default display for non-networked units) or north-up, so the display's orientation matches that of most charts and the mental model we have of the world.

Though the north-up orientation is sometimes a bit confusing to experienced radar users, it does mean that displayed electronic bearing lines (EBLs) are suddenly given in compass directions, greatly simplifying the plotting of targets. It also avoids the frustration of trying to acquire an EBL bearing and compass reading at the same time, an especially trying experience in a small boat in heavy seas. Instead, move the bearing line over the target and directly read out direction to the target.

When the display includes GPS information, plotting targets is even easier. The unit will generally show the target's position directly in latitude and longitude. Many unitslike those from Furuno and Raytheonwill even draw a "lollipop," which is a circle around the next waypoint stored in the GPS, and draw a line with course and distance information from your current position to that of the waypoint.

Some units will show other sorts of data on-screen if networked sensors exist. An example is depth, quite a useful parameter when navigating in the shallow waters of the Chesapeake Bay or those rocky ones of New England.

If you have a chart plotter on board you may be able to network the radar to the plotter to superimpose radar targets on the chart display. You'll see other traffic on the electronic chart, accurately displayed in relationship to all charted land masses and other information.

Charting radar and/or GPS data does require that you have a sense of the geodetic datum used by both the GPS receiver and the chart. Mix datums and you may find positions off by a quarter mile or more.

Many vendors are working on units that will allow you to superimpose charts onto the radar screen, but there are some important philosophical issues about the nature of the displayed data. In a busy harbor, the radar data alone is quite complex, resulting in a screen that's full and rather "busy." Do you want to overlay an electronic chart on top of this already cluttered screen?

At least two vendors (Furuno and Si-Tex) sell boxes that accept Navionics or C-Map chart cartridges and then transfer this information to their radar displays. To limit the clutter they let you alternate between radar data and the chart or even create windows with views into the chart that you can position in a (hopefully) empty part of the screen. It's an attractive solution to the problem of busy screens, yet gives that one-look glance that combines "where you are" with "who is near"without plotting anything.

Not so long ago, the radar screen was a CRT coupled directly to the data coming from the amplified returned echoes. The sweeping pattern was actually created by a motor rotating electromagnets around the neck of the CRT in synchronization with the spinning antenna. No longer. Radar displays are all computer displays, paralleling the monitor on your desk. Antenna data is amplified and digitized, fed to a quite powerful, generally 32-bit processor buried inside the display unit, before being enhanced and shown on the screen. Buttons and controls are all just inputs to the computer, which then digitally creates EBLs, variable range markers (VRMs), and on-screen displays. Some units use computers that are little more than PCs stripped of their keyboard and buried inside the display. In fact, with so many navigational software packages available now for PCs, and with laptop computers becoming common in the chartroom, it's natural to wonder why we don't dispense with the radar display altogether and simply connect the laptop to the radar antenna.

At least one vendor has taken just this approach. Pinpoint Systems' Titan product consists of a board that plugs into the PC and that connects to your radar set. The board connects to the "slave" output of Furuno units, or to a Y connection the installer splices into the antenna cable for other brands of units. (Clearly, this is not something you'd install yourself; Pinpoint provides a trained technician.)

The board contains a digital signal processor (DSP), which is a very fast microprocessor designed for signal processing applications. Smart algorithms do real-time processing on the data to enhance targets and eliminate noise. For example, those frustrating targets that fade in rough seas are "bloomed" if the software determines they are real and not random surface echoes. In fact, the versions of the DSP algorithms Pinpoint sells to commercial shippers even process echoes to find subtle patterns indicating sea ice, showing these hazards clearly before the ship has a chance to do a Titanic imitation.

Pinpoint's software reads most common chart cartridges and CDs and overlays the radar data on top of the displayed chart. Coupled with a compass input it gives true north-up display. With a GPS connected, course and waypoint information shows up as well.

Much of the cost of a radar display (not the radome) and of an electronic chart is in the CRT or LCD display screen and the computer that controls the display. It seems inevitable that radars will routinely show chart overlays and charting units will show complete radar sweeps.

Less clear, however, is the effect the PC will have on the integration of navigational electronics. Will all instruments become "headless," using a single laptop for all display purposes? Fewer displays means less power consumed and less room needed. It also means a single failure would shut down all of the electronics. Non-ruggedized laptops are notoriously unreliable at sea. PCs are built to compete in the desktop world, a very benign environment. A laptop exposed to mariners dripping sea water from their oilskins just won't survive as well as a purpose-built radar or chart plotter. Probably, though, we'll see the units becoming PCs internally as they maintain the appearance of a purpose-built instrument. Regardless, costs will continue to fall while capabilities increase. You'll find more useful nav data on the screen of all instruments. We'll probably rely ever more heavily on the electronic devices for all phases of navigation.

By Ocean Navigator