AIS mandates in the USA, the Coast Guard speaketh
Just online this morning is the USCG’s latest proposed rulemaking regarding the use of AIS by commercial vessels (and also expansion of the Notice of Arrival and Departure requirements). While the PDF weighs in at 94 pages—and contains some required bureaucratic folderol (that must drive writers nuts)—the suggested regulations make a lot of sense and will significantly improve marine safety, I think. Once refined and enacted—the USCG is hoping for 2010 mandates—the new rules will also be a boon for the manufacturers and installers of Class A or Class B AIS transponders, or both. You see, while the CG has a very specific idea about which formerly-exempt vessels should be made to carry AIS—17,442 more tugs, fishing boats, dredges, passenger vessels and others, to be exact—and endorses Class B technology with vigor, it also recognizes the superior performance of Class A, and is asking all parties involved to help decide which gear should be required on which new classes of mandated vessels…
Some commenters recommended that the Coast Guard permit the use of AIS Class B devices. We agree. Since publication of the 2003 final rule (68 FR 60559) and through the diligent work of various standards bodies, we now have AIS Class B devices that are interoperable with AIS Class A devices. Class B devices differ slightly in features and nature of design, which reduce their cost (on average half the cost of Class A devices); however, their performance is somewhat limited. They report at a fixed rate (30 seconds) vice the Class A’s variable rate (2-10 seconds dependent on speed and course change). They consume less power, but also report at lower power (2 watts versus 12 watts of AIS Class A), thus impacting their broadcast range. Despite these design limitations, and after extensive testing by the Coast Guard Research and Development Center, we deem AIS Class B devices can operate properly and safely amongst Class A devices and offer similar AIS benefits. They broadcast and receive virtually the same vessel identification and other information. They have the same ability to see targets that radar may not always show (around the bend, in sea clutter, or during foul weather). For these reasons, we have concluded that AIS Class B devices do enhance navigation safety and assist in collision avoidance comparable to AIS Class A devices; however, given their design limitations, we caution users that they may not be the best alternative for vessels that are highly maneuverable, travel at high speed, or routinely transit congested waters.
The Coast Guard seeks comment in this NPRM on whether AIS Class B devices should be permitted only on certain vessels or waterways, or whether this decision should be best left to the master or owner’s discretion.
Mandated AIS usage is subject to “effective operating conditions” and the new rules smartly intend to spell those out:
(i) ability to reinitialize the AIS (knowledge of system password)
(ii) ability to access AIS information from conning position
(iii) accurate broadcast of an official MMSI
(iv) accurate input and upkeep of all AIS data fields ‘and system updates’
(v) continual operation of AIS and its associated devices when underway, at anchor, or moored in or near a channel, except when its use would compromise safety or security (securing an AIS for the later must be logged and reported to USCG)* AIS text messaging must be conducted in English and solely to exchange or communicate navigation safety information
* AIS is primarily intended for use of the master or person directing the movement of the vessel, who must maintain a periodic watch for AIS information
* Spells out that use of AIS does not relieve the vessel of Navigation Rules duties regarding sound, lights or shapes nor Bridge-to-Bridge radiotelephone requirements
Which means, I think, that all commercial vessels using AIS in U.S. waters are going to have to clean up data irregularities, and at least have target info near the conning position. (Interestingly, the CG states in the proposed ruling that it would like to mandate target display on an electronic charting system, but can’t until such systems are better defined, a process which is underway.) I’ve asked if “effective operating conditions” might also mean that a mandated vessel can not filter out Class B targets, but I think that liability issues will take care of that overblown issue anyway.
At any rate, this is just the beginning of a rulemaking process I intend to keep an eye on (e-mail notifications are available at the same Regulations.gov site). At the moment I don’t have a strong opinion as to whether any of the newly mandated commercial vessels should be required to use Class A instead of B, though those 25–knot Water Taxis I mentioned yesterday would surely plot better if A equipped. What do you think?
I strongly support the requirement that all commercial vessels be equipped with AIS.
Regarding A vs B I don’t see anything in the electronics of the A that necessitate it costing twice as much to build and I doubt the regulatory hoops are any more difficult (FCC, etc.). I suspect the price either reflects 1st generation hardware that has not been cost reduced, or high profit margins because the A systems were going to customers who were required to make the purchase.
I think there is a clear opportunity for someone to bring to market a class A system at class B prices. It won’t be any of the established vendors, they won’t gore their own ox, especially during these economic times. But we may see a new vendor who sees the opportunity to grab a lot of share, especially if the CG is about to increase market demand.
So, what if the cost were the same, why allow or require A vs B? Wouldn’t we all prefer to be receiving better information (i.e., more frequent and at greater range)? These update rates are all theoretical; your receiver does not necessarily receive every update from every AIS. Who wants to watch a target jump across their screen because you miss one class B update and the target with a speed of 15kt has travelled 1,500m and made a turn since the last update?
If the argument for B is that an “all A” world would cause too much congestion on the AIS channels, then the CG/FCC had better get some regs on the books to limit the use of A very soon because telling pleasure boats they can’t use A will cause a lot of grief as the installed base of A continues to grow.
In summary, require AIS on every commercial vessel, hope that a new vendor emerges to lower the cost of A and forget about B altogether.
I agree with Russ’s comments. I am reminded of the early days of DSC VHF radios when someone thought it was a good idea to have a lesser standard (SC101) for recreational boaters in the US, fortunately we have matured and now most boaters purchase at least a “class D” radio. (Yes, I know there are also class A radios, but that is another topic).
If you think about it, an AIS transponder is little more than a glorified VHF radio, and I rather suspect that if AIS units were to be sold in the same volume as VHF radios, the price would not be much different.
If the price is driven down by demand, and anyone with cash can put an AIS A or B on their vessel, how will all ths info be sorted out and be “useful” as opposed to confounding and chaotic (in crowded waters where it’s supposed to help)?
Jef .. not a problem. I’ve transited the Singapore Straits several times and have had over 450 targets detected at one time by my AIS receiver (single channel NASA AIS engine). What you do in practice is to zoom in on the display (I use both Yacht-AIS Pro and MaxSea 10.3) to show those vessels near you. You select the parameters of CPA and tCPA for collision alarms and reduce these in such crowded conditions. AIS has taken the chaos out of such passages and I ‘won’t leave home without it’.
Another major issue that will need to be addressed as AIS installation become more prevalent is that of accuracy. Most installations are accurate. But we have seen 350′ freighters steaming at 15 knots perpendicular to their reported heading. We’ve seen commercial boats with a length of 10 meters and beam of 35 meters. And of course there is the never ending stream of vessels reporting at anchor doing 20 knots, or tugs underway with large tow while tied firmly to the pier sans any kind of tow.
On the recreational side it only gets worse with all of the above plus inaccurate position reporting etc. What is Puget Sound, or any other body of water going to look like when we have hundreds of class B AIS transponders out there that have been haphazardly installed by an owner or even a technician that doesn’t bother thinking it through or reading the documents.
I suppose that down the road there will be mandated installation inspections and possibly periodic inspections. But until then, I’d like to see some focus on getting the present installations correct. After all, a lack of information is far preferable to incorrect information, at least from my perspective.
Lets say for the sake of argument that the class B units aren’t grossly overpriced. With this in mind, how do you argue that you can amplify the output by 6 times, and add a dedicated display without any impact on price. If you want to argue that the class B units are overpriced, you can do that and hope that both types of units will come down as production increases, but to say that there isn’t an inherent price difference between the base requirements of the units doesn’t seem realistic.
I think the primary reason for Class A’s extra expense is the additional memory requirements needed for Class A operation. It needs to do alot more work than Class B to keep track of open slots in the VDM so that it doesn’t end up stomping on top of another unit.
But given Moore’s law the costs should be coming down, at least for new Class B models.
I guess that one of the main reasons of cost difference between Class A and B is that A has to pay for SOTDMA license, and software on it whereas class B does not have to.
Salva
Class A’s cost a lot more than Class B’s because they are spec’ed harder. The RF specs are more stringent, the output power higher, both of which mean that you can’t use a Class B RF section in a Class A – hence it costs more.
Also Class A’s have a significant number of IO ports mandated by the standard, all of which require interfaces, protection and handling, which needs bigger processors, more power and more cost. In contrast, Class B has one NMEA 0183-HS interface.
And that’s in addition to the additional processing that the Class A has to do compared with Class B.
So, don’t be deceived – a Class A costs more to design, produce, test and approve than a Class B, and always will do.
And Jim – Class B’s only report id, position, speed etc, there is no option to select “at anchor” etc as in a Class A, so there is much less opportunity for reporting erroneous info on a Class B and little scope for an ignorant technician to get it wrong (but not zero, I grant you that…..)
I’d give BCL (Brian) and Del a lot of credence in this conversation as they are both technical guys working in the AIS field. I’m also aware of many Class A technical subtleties that I’ve never discussed here on Panbo, and I rarely see mentioned anywhere.
All that said, this would be a great time for a manufacturer to introduce a less expensive Class A transponder, if possible. If you read the USCG NPRM (Notice of Proposed Rule Making), you’ll see that the financial burden to the mandated AIS user seems to be a big factor in the deliberations.
Even if putting available Class A hardware on these New York Harbor Water Taxis would seem fairly trivial to their budgets, if there are similar class vessels (50-149 passengers, under 30 knots) working in tight financial conditions, they will be heard.
It seems incredible to me that those high speed water taxi’s you mention aren’t fitted with AIS at all – as you say, even an expensive Class A is peanuts to that sort of operation. Even the tatty old ferry going around Poole Harbour in the UK at 8knots has one!
Interestingly, there was a proposal for a sort of low-cost Class A a few years back, but it didn’t get enough support to be carried forward into a standard (oddly it was known as Class B SOTDMA – presumably because the definition of Class A was reserved for SOLAS???).
Seems to me that a suitable division would be if its commercial and does over (say) 14knots, it should have a Class A (so that it has the higher reporting rate, primarily). Sound reasonable?
ibsalin: The cost differential between 2 watts and 12 watts is nothing, 10 watts of RF power is very cheap. A 6w handheld VHF retails for less than $150.
McCorison: I haven’t seen those kind of errors. With a built in GPS I’m at a loss to understand how a tug tied to the dock can report a speed of 20 knots. How does this happen?
bcl/del/ben: No way! A furuno Class A AIS lists for $4,500 and Simrad is more like $7,000. They have less memory, storage, processing power, smaller display, etc. than a Furuno MFD8 which lists for only a touch over $3,000. An off the shelf laptop has tons more memory, storage, processing power, I/O, display, battery, etc. than is necessary for a Class A at about a quarter the price. The RF section of an AIS is less sophisitcated than a $130 handheld VHF and has only a 6 watts more power.
I don’t know if a Class A device requires a display, if so, that adds a smidgen of cost, but it’s nothing like the display of a $1,500 Dell laptop. The display that accompanies a Furuno FA150 is essentially the same display they provide on a $300 GP32, it’s not a big cost item!
An iPhone has two radios (3G GSM and WiFI), both of which are far more sophisticated protocols than VHF/AIS, and a color/touch display, plus far more processing power and storage (16GB) than is required by class A; it costs maybe $350 to build.
What about Class A-, with Class A power and update frequency, but without the display? The CG wants to see this stuff on a chart plotter which makes the display on a Class A redundant.
I reiterate that Class A transponders are grossly over priced.
Suppose that Class A cost $700 retail, and Class B cost $400 retail. Back to Ben and the CGs questions, what should be required on the various commercial vessels? Would you spend $300 to get A instead of B?
Class A costs more becuase of the IMO regs ect.
$4500 is cheap for your safety !
My point wasn’t that class A isn’t overpriced…it certainly seems to be and you make the point nicely, just that it DOES cost more than a Class B. The class A display requirement isn’t very stringent, but it must have some sort of display and it must have some specific output ports.
I would hope that class B units could come down to the $500 range soon and therefore by comparison, class A units should be able to come down to $1000 for a basic unit at most.
I understand initial prices being higher to cover development and regulatory costs, but eventually they should be based on price to manufacture which should be a lot less.
Then the decision of spending a few hundred bucks more for class A makes sense most of the time (unless you are a slow boat with small batteries in which case the power savings of class B would be valuable).
It strikes me as silly to compare Class A transponders to mass produced consumer electronics or even handheld VHF radios. I also question judgments about which protocols are more sophisticated unless the judge understands what all the tech talk on this page is about (I don’t):
http://tinyurl.com/6fh67c
That company, incidentally, is where “del” works, I think.
Here is a little AIS history, I think dating from sometime after Class A design, but before Class A manufacture, and before Class B:
Anyone who has gone to sea has experienced the frustration of staring into a radar screen and seeing a brunch of blips staring back at you. Some are bigger than others and you hope that the others can stare into their screens and see your blip. For years this staring has been going on.
In an effort to help the navigator identify some of the blips, various government and technical organizations working in concert with the International Maritime Organization (IMO) have come up with a system known as AIS or Automated Identification System.
So far so good.
There is much discussions around the fact that many times the problems with identification of target on radar is due to the fact that there are many small targets, such as fishing boats, tugs and pleasure craft.
These vessels are not considered SOLAS ships under the regulations.
These “blips” or smaller vessels often do not show up and there is a need to equip these boats with some kind of AIS equipment to allow them to be identified as well.
Ships normally do not collide with ships; they get into trouble because they have to take evasive action to avoid striking a smaller vessel. There are now proposals on the drawing board to establish a second class of AIS reporting known as AIS Class B who would have to carry basically the same equipment but only would transmit their signals every 15 minutes. Class A would be for SOLAS Vessels.
Makes sense, up to the 15 minutes comment!
Technology design goes so much faster than regulations and implementation, it’s almost a given that things are obsolete by the time we get them. So with hindsight, we know things could have been done better. But from experience we know that we’d better take advantage of what we do have now and not wait for perfection.
The above quotes are from
http://www.thedigitalship.com/DSmagazine/digital%20ship%20archive/digital%20ship%20nov/AIS.doc
For current AIS issues, this looks like a good link:
http://www.nautinst.org/ais/techFeedback.htm
quote:
“The RF section of an AIS is less sophisitcated than a $130 handheld VHF and has only a 6 watts more power.”
Sorry, I can’t let that go without a comment – as Ben says it is absolutely pointless trying to compare costs of equipment which has a limited market (SOLAS has 80k vessels at the last count) compared with the consumer market, which numbers in millions!
The RF section of a VHF handheld is relatively simple and has to meet basically the same specs (give or take) as much of the land mobile market and so designs and costs can be shared across both markets and the benefits of mass production apply to both.
AIS on the other hand consists of TWO parallel receivers to start with – which can’t be identical otherwise they interfere with each other and a transmitter which must ramp up and down VERY quicky without splattering across the whole spectrum – not an easy task, let me assure you! The Rx specs are also an awful lot tougher – the crowded nature of the marine environment and the self-organizing nature of the system mean that the interference and blocking performance of an AIS Rx is way more extreme that yer average VHF portable.
Class A’s will always be more expensive than Class B – after all, the IEC committee spent 5 years designing the Class B standard to make sure that Class B’s would be cheaper, however, I would point out that if I was designing a Class A from scratch today, I could do it a lot cheaper than 5 years ago…. however…. where’s the incentive to do so? The market for Class A is SOLAS, and all SOLAS ships already have a Class A installed, so it would be a very brave manufacturer to commit his (expensive) design resources to a project with a minuscule market. Hopefully, the initiative from the USCG will provide a new, expanded market for Class A (or Class B SOTDMA) and convince manufacturers to support it, and so develop new, cheaper units that will ultimately benefit the entire boating community.
I’ve had 30 years of designing radio equipment, both for land mobile, marine, data and trunked systems – trust me on this, AIS is NOT easy!
A vs. B cost: Dosn’t class A also need to connect to a dedicated flux gate compass, and maybe even rate gyro, to report heading ? That along with the additional interface port adds a bit of expense too ?
I’m very familiar with the protocols of GSM/Edge/3G on both the RF and base band processing. An iPhone has a 3G/GSM spread spectrum RF transceiver operating at four frequencies (800Mhz/900Mhz/1.8Ghz/1.9Ghz) and a WiFi transceiver operating at 2.5Ghz packed into a very small space, both interacting with other elements of the network that control their operating frequencies and power levels. An AIS transceiver has an enormous amount of space for shielding and RF isolation and operates almost autonomously.
As Del knows, silicon is essentially sold by the sq mm. The protocol processing is s/w in a baseband processor where the silicon can be shared across multiple markets and products.
There is nothing inherent in Class A AIS which can justify a retail price of $4K – $7K.
In any case, to answer Del’s core question of “what’s the incentive”; the incentive is that at a lower price the market is larger. This is marketing economics 101, otherwise known as price elasticity. Many vessels who purchase Class A today are not SOLAS, they’re larger private yachts. Lower the cost/price and you’ll find the market extending further and further down into smaller yachts, the fatter part of the pyramid where volumes rise dramatically. As a supplier to the systems builders, CML should be very interesting in increasing volume. It’s the systems builders that usually resist, not the component suppliers.
Over at http://nauticalinformation.blogspot.com/2008/05/ais-20-next-version.html they make a good case that:
“Data communications required to implement eNavigation are likely to overwhelm the capacity of the 10-year-old design of AIS. The next version of AIS should not only address this constraint, it should also act as back-up, both for AIS and for VHF voice communications, while reducing ship operators satellite telecommunications costs.”
If this is true, i.e. that the current version of AIS lacks sufficient bandwidth capacity to accommodate possible future growth, then perhaps this should be addressed prior to the CG mandating the installation of either class A or B devices?
Thanks, JonnyB, that’s an interesting blog I’ve never seen before. But that entry is quite speculative, the extra needs of eNavigation may not materialize for many years, and even this blogger suggests that AIS 2.0 will be introduced as a backup or parallel system.
So, no—in fact, hell no—boaters and the USCG should not hold up the expansion of AIS use while waiting for some ephemeral AIS 2.0.
To help decide which gear should be required on which new classes of mandated vessels, here’s my opinion. Start a Class A requirement for fast (>14 knots) commercial vessels with passengers in crowded areas and phase it in for other areas.
Ben pointed to the CML pages. Their Knowledge Base page on AIS in nicely geeky and informative:
http://www.cmlmicro.com/Press/briefs/ais.htm
Their comment on one of their Class B designs is:
Tests at various locations around the UK have demonstrated ranges in excess of 30 nautical miles even with a simple antenna mounted at deck level on a leisure sailing craft.
Ben, I am somewhat confused. According to the IMO Website
“The regulation requires AIS to be fitted aboard all ships of 300 gross tonnage and upwards engaged on international voyages, cargo ships of 500 gross tonnage and upwards not engaged on international voyages and all passenger ships irrespective of size. The requirement became effective for all ships by 31 December 2004.”
http://www.imo.org/Safety/mainframe.asp?topic_id=754
So what’s this about fitting of AIS not applying to passenger vessels?
What am I missing?
Regards
Northrop to build data exchange for vessel tracking
http://www.washingtontechnology.com/online/1_134139-1.html