The RANGE of AIS Signals; Class-A and Class-B Compared

VHF Marine Band radios, protocol, radio communication theory, practical advice; AIS; DSC; MMSI; EPIRB.
jimh
Posts: 11672
Joined: Fri Oct 09, 2015 12:25 pm
Location: Michigan, Lower Peninsula
Contact:

The RANGE of AIS Signals; Class-A and Class-B Compared

Postby jimh » Thu Nov 09, 2017 2:02 pm

Regarding range of AIS signals and a comparison between Class-A and Class-B AIS transceivers:

A recreational boat will typically have a CLASS-B AIS transceiver, and a large commercial vessel will have a CLASS-A AIS transceiver. In assessing the radio range (distance) between these two categories of vessels, consideration has to be given to the transmitter power. CLASS-A transmitters are 12.5-Watts, and CLASS-B transmitters are 2-Watts. This gives the CLASS-A signals an advantage of about 8-dB. The implication of this is as follows:

--assume everything else were the same, that is, the antenna height, the antenna gain, the transmission line loss, the receiver sensitivity;

--the CLASS-A vessel has an 8-dB stronger signal, which generally translates into an increase in radio range, that is, the CLASS-A signal can be heard at a greater distance than the CLASS-B signal.

The amount the radio range will increase by adding 8-dB to the signal power will depend on the nature of the path between the stations. For propagation over open water one can estimate the path loss will vary with distance (d) by

    dB = 40 x log(d)

Solving this for d when db=8 gives 1.58

This means the path distance can be 1.58-times longer for a signal with 8-dB more power. In simple terms, a CLASS-B vessel is going to hear a CLASS-A vessel's AIS signal for a long time before the CLASS-A vessel hears the CLASS-B, when the two vessels are far enough apart to be at the very limit of the radio communication path. That last part is the important part. If the vessels are only a few miles apart, the path loss is not going to put the signals at the threshold of being received. While the extra gain helps when the vessels are far apart, it won't have as much effect when they are closer.

Another factor is how often the vessels transmit. A CLASS-B vessel underway at less than 14-knots only transmits every 30-seconds, while a CLASS-A vessel at the same speed range transmits every 3.3-seconds if changing course and every 10-seconds if on steady course. This means the CLASS-B vessel is transmitting something like three-times to almost ten-times more often than the CLASS-B. The more often you transmit data the more likely it is to be received, so, again, the basic methods of AIS tend to favor the CLASS-A vessel being received and detected by the CLASS-B vessel before the other way around.

A further problem with AIS signals is the actual reception and accurate decoding of the signals. An AIS transmission is a broadcast transmission, that is, the transmitter is just sending out the signal, and it has no idea who might be receiving it. Many people misunderstand the nature of a digital transmission of data, and they anticipate that because the data is sent in a digital coding there can never be any error. This is far from the truth. Some digital transmission systems are bi-directional, and the sender and receiver switch roles frequently, with the receiver being able to interrupt and ask for a repeat transmission if there was some data loss. None of this happens with AIS. The transmitter just broadcasts a signal, out it goes, and who know what happens to it. Maybe another vessel hears it, but you can't really tell.

Even in a one-way broadcast transmission of digital data it is possible to incorporate a means of error detection. Indeed, in AIS signals the cyclical redundancy check (CRC) error checking method is employed. But this method generally only allows for the detection of errors. A number of very sophisticated methods have been proposed to allow error correction to occur by sophisticated signal analysis mathematics and use of the CRC data, but it seems very unlikely these complex computational methods would be available in an inexpensive ($200-class) AIS receiver. A typical recreational grade AIS receiver is just going to discard any data it received by AIS transmission if it detects an error in the data. On that basis, the more often the data is transmitted, as occurs with CLASS-A vessel, means a greater chance it will be received without error by other vessels.

If into this analysis we add the notion that the CLASS-B vessel has a much lower antenna height, probably a lower gain antenna, and is likely to have much more motion on the boat (causing the antenna radiation pattern to be skewed away from the horizon), we can find other influences that will tend to reduce the visibility of the CLASS-B vessel to the CLASS-A vessel.

There is a newer category of CLASS-B AIS transceivers called Self-Organizing, which will have increased power (5-Watts). This decreases the power difference to 4-dB, making the range factor 1.26-times, and helping to put CLASS-B SO transceiver on a more even footing with CLASS-A devices in terms of signal power. A closer look at the reporting intervals follows below.

jimh
Posts: 11672
Joined: Fri Oct 09, 2015 12:25 pm
Location: Michigan, Lower Peninsula
Contact:

Re: The RANGE of AIS Signals; Class-A and Class-B Compared

Postby jimh » Thu Nov 09, 2017 2:40 pm

The interval at which a vessel sending AIS data should transmit that data is defined in the AIS specifications. The source document is

Recommendation ITU-R M.1371-5
(02/2014)
Technical characteristics for an automatic
identification system using time division
multiple access in the VHF maritime
mobile frequency band


(You can obtain a free copy of this document from the ITU website.)

I have seen many representations of the reporting interval on other websites, but they often seem very hard to interpret and, in some cases, seem to have errors in them. Below I describe the reporting intervals as specified by the source document I mentioned above:

TABLE 1
Class A shipborne mobile equipment reporting intervals

Ship at anchor or moored and                 3 min
not moving faster than 3 knots
Ship at anchor or moored 10 s
and moving faster than 3 knots
Ship 0-14 knots 10 s
Ship 0-14 knots and changing course 3 1/3 s
Ship 14-23 knots 6 s
Ship 14-23 knots and changing course 2 s
Ship > 23 knots 2 s
Ship > 23 knots and changing course 2 s


TABLE 2
Reporting intervals for equipment
other than Class A shipborne mobile equipment

Class B “SO” shipborne mobile equipment      3 min
not moving faster than 2 knots
Class B “SO” shipborne mobile equipment 30 s
moving 2−14 knots
Class B “SO” shipborne mobile equipment 15 s
moving 14−23 knots
Class B “SO” shipborne mobile equipment 5 s
moving >23 knots
Class B “CS” shipborne mobile equipment 3 min
not moving faster than 2 knots
Class B “CS” shipborne mobile equipment 30 s
moving faster than 2 knots


The ITU specifications also provide for longer intervals between reports when the traffic load on the self-organized network can be detected by a transceiver to be near maximum capacity. If that situation occurs, then all vessels in certain speed categories will change to longer intervals between transmissions in order to decrease the loading on the self-organized network traffic. (See the specification for details.)

In normal conditions when all vessels in the AIS network are using the nominal reporting intervals, we can say that a ship moving at about 14-knots with Class-A transceiver will transmit at 10-second intervals unless changing course, when the reporting interval is shortened to 3.3-seconds. Class-B transceivers on vessels moving at about 14-knots will nominally transmit every 30-seconds.

jimh
Posts: 11672
Joined: Fri Oct 09, 2015 12:25 pm
Location: Michigan, Lower Peninsula
Contact:

Re: The RANGE of AIS Signals; Class-A and Class-B Compared

Postby jimh » Fri Nov 10, 2017 12:46 pm

If two vessels are sufficiently close to each to be within each other's radio range, it can be assumed that each will receive the AIS signals from the other vessel. To assume that reception of the signal implies perfect decoding of the information in the signal is, however, a bad assumption. We can look at some statistical data to see what an average rate of reception and successful decoding might be by using information from the U.S. government's N-AIS (Nationwide AIS) system.

According to publicly disclosed data from the government, N-AIS typically received AIS messages from about 14,000 unique vessels on an average day, and is able to decode 118,000,000 messages. This suggests that the average number of messages transmitted per day per vessel would be about

118,000,000-messages-per-day / 14,000-vessels = 8428.6-messages-per-day-per-vessel

In one day of 24-hours there are 86400-seconds, so the average rate of transmission is thus

8428.60-messages-per-day-per-vessel / 86400-seconds-per-day = 0.097556712962963-messages-per-second-per-vessel.

The inverse of that number is 10.25-seconds between reception of a message from the average vessel. In other words, the N-AIS is saying that as a daily average it gets an AIS message from a typical vessel every 10.25-seconds. This is really quite an astonishing number, considering the rate at which messages are transmitted. In order for a vessel to transmit messages at close to 10-seconds intervals, the vessel has to be in a particular category:

--must be a Class-A transceiver
--must be at anchor or moored and moving faster than 3-knots, or
--must be underway at 0 to 14 knots

If we assume that the overwhelming majority of AIS transmissions are probably from vessels that meet those three criteria, the we can make an estimate of the percentage of messages that were missed if we assume all average rate of all ships was a message every 10.0 seconds and N-AIS succeeded in receiving a valid message every 10.25 seconds. To send 100-message would take 1,000-seconds. Since N-AIS averages only one message every 10.25-seconds, in 1000-seconds they would receive only 97.5-messages. This suggests a remarkably low error rate of missed AIS messages by N-AIS.

jimh
Posts: 11672
Joined: Fri Oct 09, 2015 12:25 pm
Location: Michigan, Lower Peninsula
Contact:

Re: The RANGE of AIS Signals; Class-A and Class-B Compared

Postby jimh » Mon Jan 28, 2019 1:59 pm

Regarding Class-B AIS devices, there are now two categories of these devices on the market. The older or traditional Class-B devices used a carrier sense time division multiple access technique (Class B "CS" or CSDTMA) to know when it was polite to transmit, that is, they listened for a few milliseconds in the transmit frame they were about to use to make sure no other ship was actually transmitting in that frame. Within the past year or two, a new series of Class-B devices have come out that can participate in the self-organized time division multiple access (SOTDMA) network negotiations with Class-A ships. These Class-B "SO" or SOTDMA devices transmit at a power of 5-Watts. This means compared to 12.5-Watt Class-A transmitters, they are only at a disadvantage of 5/12.5 = -4 dB.

If we figure that the ultimate range of reception will vary as a function of path loss according to dB = 40 log(distance) and solve for dB = 4 we see the range for a 12.5-Watt AIS compared to 5-Watt AIS will increase by a factor of 1.25-times.

The assumption that path loss as a function of distance is well-described by the dB loss = 40 log (distance) relationship is just an estimate for a typical VHF signal path over uneven terrain. It is certainly possible that path loss could be less over certain conditions of open sea, calm seas, stable temperatures, and other atmospheric conditions. If path loss is lower with distance, the added 4 dB will improve the distance more than estimated above.

Again, the advantage of 4 dB additional signal only comes into play when the signals are at the margin of reception. Once signals are well above the AIS receiver's sensitivity threshold for very low bit error rate reception, the greater signal does not give a huge advantage. That said, it should be noted that many paths are unstable and subject to fading, so stronger signals have larger fade margins and will remain good copy if the path loss increases and signal levels fade, while the weaker signals may fade below the threshold of accurate copy.

A further advantage for the Class-B SOTDMA device is the frequency at which it can transmit data. While both Class-B CSTDMA and SOTDMA will be sending the same set of data, the Class-B CSTDMA device is allowed to transmit more often. Compare the rates as seen in TABLE 2 above; the Class B SOTDMA can transmit its data as often as every five seconds if moving rapidly. This gives the SOTDMA device a six-times more often transmission rate than a CSTDMA device.

jimh
Posts: 11672
Joined: Fri Oct 09, 2015 12:25 pm
Location: Michigan, Lower Peninsula
Contact:

Re: The RANGE of AIS Signals; Class-A and Class-B Compared

Postby jimh » Mon Jan 28, 2019 2:12 pm

Other factors can affect the range of Class-A and Class-B AIS signals, but those factors are not inherent in the limitations of the devices. For example, the antenna on a ship with a Class-A AIS may be more than 100-feet above the water, while the antenna on a recreational boat with a Class-B AIS may only be 10-feet above the water. It is well-known that for shorter VHF paths the antenna height has a very direct and substantial influence on the signal strength at distance. If starting at lower heights, a doubling of antenna height often doubles the signal at a distant receiver. On this basis, the added height of an AIS antenna on a large ship can very greatly increase the range at which other boats can detect and demodulate the Class-A AIS signal from the large ship.

Another factor may be transmission line loss. On large ships, it is common that all antennas are fed using very low-loss coaxial transmission lines, often with solid metal outer conductors. These transmission lines have minimal loss, while on a recreational boat an AIS antenna might be fed with small flexible coaxial transmission line cable with much greater loss characteristics. The difference in transmission line loss is likely at least a 1 dB advantage for the Class-A commercial ship over the Class-B recreational boat.

Finally, the AIS antenna on a commercial ship may have more gain than a smaller antenna on a recreational boat. A difference of perhaps 3 dB might occur.

If we include the factors of antenna height, transmission line loss, and antenna gain, the typical commercial ship with Class A AIS may have an additional advantage from these factors that could be a great as 5 to 10 dB. To that we add the 4 to 8 dB advantage in transmitter power, giving perhaps a 9 to 18 dB advantage to the Class A AIS signal from a large vessel. These advantages will, again, add to the distance the signals are able to be received with low bit error rate demodulation, and can account for reports of Class-A signals being received at ranges greater than 50-miles, while Class-B signals tend to be much more localized.