HELSINKI COMMISSION HELCOM AIS EWG 21/2010 Expert Working Group for Mutual Exchange and Deliveries of AIS data 21 st Meeting Gdynia, Poland, 27-28 October 2010 Agenda Item 6 Definition of AIS coverage Document code: 6/2/INF Date: 4.10.2010 Submitted by: Chairman DRAFT PROPOSAL FOR OPERATIONAL DEFINITIONS OF AIS COVERAGE Note by Secretariat: FOR REASONS OF ECONOMY, THE DELEGATES ARE KINDLY REQUESTED TO BRING THEIR OWN COPIES OF THE DOCUMENTS TO THE MEETING Page 1 of 1
EfficienSea Efficient, Safe and Sustainable Traffic at Sea Draft proposal for operational definitions of AIS coverage Summary This document proposes definitions for declaring coverage areas for shore based AIS networks, related to different operational needs. Background The introduction of shorebased AIS networks have proven useful for application in VTS centers, MRCC traffic monitoring and SAR coordination, incident investigations, and many other applications. In the GMDSS, firm declarations of service areas - like the A1 area are used to denote areas, where reliable VHF voice coverage can be expected. Similar definitions of reliable coverage areas for AIS are not declared, due to the lack of a good operational definition of what is meant by AIS coverage. IALA guidelines on AIS, Edition 1.1, Volume 1, Part II, section 16.1, refers to the basic principles of RF coverage based on the GMDSS Sea Area A1 definition for VHF, reused to a rough estimate of AIS radio coverage, based on the following formula, assuming omnidirectional antennas: The radius A is equal to the transmission distance between a ship s VHF antenna at a height (h) above sea level and the height (H) of the antenna of the VHF coast station which lies at the centre of the circle. The following formula should be used to calculate the range A in nautical miles A=2.5(Square root of H (in metres) + Square root of h (in metres)) The A1 area definition assumes vessel antenna heights of 4 meters, however several administrations have argued that an antenna height of 10 meters is more relevant for SOLAS vessels, and several studies have indicated that actual AIS coverage in terms of position tracking range is far greater than the predictions of the formula above. Also, there are obviously different operational considerations to take, if coverage is needed for providing DGNSS correction broadcasts at a certain availability level, or if AIS coverage is needed for providing a coastal Long Range Monitoring network for vessels sufficient for the purposes of, or equivalent to LRIT. Thus, the basic estimation of coverage range has proven insufficient to describe the coverage of AIS networks, and a more precise definition is sought.
Issues under consideration Radio coverage area at VHF frequencies, and AIS in particular, will be affected by: Antenna system and height Propagation path Weather Sea state (wave height) VHF datalink loading It is possible to use a propagation modelling tool based on Recommendation ITU-R P.1546-3 to estimate the range performance of AIS based on shore station antenna design, signal loss in feeder, assumed vessel antenna, etc., however this tool still does not take into account all the factors listed, and the coverage ranges estimated in this manner do not relate precisely to relevant operational usage of AIS information. Empirical studies performed by the Australian Maritime Safety Authority (AMSA) and by the Danish Maritime Safety Administration (DaMSA) have documented observations on reception quality, measured as the percentage of position reports received from class A AIS stations as a function of distance from shore station, compared to the nominal reporting intervals defined by the ITU-R M.1371, as a function of the vessel speed and rate of turn. These studies document, that the reception quality is seldom above 80-90%, provided that no terrain shadowing effects are present. The reception drops off almost linearly in the range 75-25%, and below ranges that provide a 25 percentile, the range corresponding to a given percentile becomes very sensitive to weather, or abnormal propagation phenomena. Illustration from study performed by AMSA:
Illustration from study performed by DaMSA: Based on the work of the National Institute of Telecommunications (Poland) and Gdynia Maritime University on simulating AIS signal propagation by Recommendation ITU-R P.1546-3, it is noted that commonly in radio engineering, areas where radio stations have a probability of signal reception of >50%, is considered to be within useful range, while areas with signal reception probabilities of >10% are considered to be within interference range i.e. stations within interference range are considered to be significantly affected by the transmitting station. Noting that the AIS is a 2-way communication data link, it is observed that there is likely to be a minor difference in uplink / downlink signal reception probability. Although transmission power and receiver sensitivity requirements are identical for AIS class A stations and AIS Base Stations, antenna installations at Base Stations are likely to be better and under stricter control, enabeling slightly better downlink probabilities than uplink probabilities, as the antenna quality of the transmitting station is the most significant contributor to high signal reception probability. While downlink propagation from Base Stations to vessels is easy to simulate, it is more difficult to
simulate uplink propagation, from a moving population of vessels with a highly variable arrangement of antenna installations towards Shore Stations. To simplify definitions, is is however assumed that the radiolink on average is symmetrical in performance. It is the intention of the EfficienSea project to perform further studies on the sensitivity of AIS coverage quality to the different factors, and attempt to correlate ITU-R P.1546-3 based radio propagation analysis with observed AIS reception quality as described above, to enable operational solutions for declaring AIS coverage areas relevant for certain operational criteria. The intention of this document is to propose an operational definition for AIS coverage, that relates to operational requirements of relevant services. Such Proposed definition of AIS coverage quality Based on the observations from the AMSA and DaMSA studies, and the content of ITU-R P.1546-3, we propose the following definition of AIS coverage quality: The AIS coverage quality (QAIS) of an area is expressed in percent, as the average ratio of position reports received to the number of position report transmissions. Number of received unique position reports 1 QAIS = ---------------------------------------------------------------------- * 100% Number of transmitted position reports 2 QAIS shall generally be defined for class A AIS vessels with an antenna height of 10 meters, or the average of those observed to populate a given area, but may, depending on operational considerations be derived separately for vessels with different characteristics for instance low power / low antenna height class B vessels (QAIS, class B), or class A vessels of a certain size / antenna height (QAIS, h=4m, QAIS,h=10m). Note that this definition is an approximation tied to single timeslot position reports, which are transmitted using the self organizing SOTDMA algorithm. It does thus not directly describe the probability of reception of multiple timeslot transmissions, or transmissions occurring in FATDMA proceted timeslots. Note also that QAIS is likely to be time dependant, due to sensitivity to Propagation path Weather Sea state (wave height) VHF datalink loading 1 By a certain AIS base station, or network of AIS stations. Note that position reports must be unique i.e. reception by more than one station does not provide quality levels above 100%. 2 Based on the Nominal Reporting Rates defined for mobile AIS equipment in ITU-R M.1371. The study performed by DaMSA under the EfficienSea project suggests, that this value can be difficult to determine in empirical analysis, as ROT and SOG values are unstable, and some slow moving AIS stations are observed to have unexpected high reporting rates.
For this reason, we differentiate between predicted operational AIS coverage quality (QAIS,Predicted) and actual AIS coverage quality (QAIS,Actual): Predicted Operational AIS Coverage Quality (QAIS,Predicted) may be estimated by radio propagation simulation assuming vessel antenna height 10m and/or validated by empirical analysis of received AIS signals, as an average over at least 30 days, and averaged over relevant vessels in the area, to take into account natural variations in propagation and vessel population. 3 Actual AIS Coverage Quality (QAIS,Actual) may be derived by empirical analysis of the last hour of AIS reception, and will reflect the currently available coverage, including the effect of full or partial system failures in an AIS network. Declarations of service areas for an AIS base station or network, should be based on QAIS,Predicted Proposal for defintions of AIS coverage Based on the definition of Predicted AIS coverage quality (for class A AIS vessels), the following definitions are proposed for declaring coverage areas for AIS services, specifying service levels relevant to particular operational situations: AIS coverage QAIS,Predicted 1) Good AIS coverage >75% 2) General AIS coverage >50% 3) Long Range AIS coverage >10% (interfering range) Rationale for definitions: 1) Good AIS coverage: Advanced broadcast services with high availability requirements, such as a DGNSS broadcast service, may require a high degree of signal reception probability. The 75% level seems to be where the reception quality curve has it s first significant bend, and is possible to determine empirically, however higher values than 75% have proven difficult to achieve. 4 2) General AIS coverage: The 50 % level seems to be roughly equivalent to the coverage ranges estimated by the formula available in the IALA guidelines, and is equivalent to what s commonly referred to as useful range in radio engineering. Most uses of AIS coverage which require a fairly detailed resolution in vessel movement monitoring, (including VTS operation, which relies more on RADAR for reliable position verification), will not suffer significantly, even if a few position reports are missed, as long as the likely distance covered by vessels between updated positions does not grow significantly. 3 In regions where propagation is known to vary significantly with seasonal changes, the minimum monthly value obtained for a year, should be considered the safe definition. 4 In radio engineering, percentiles above 50% are not used. Empirical studies of coverage quality as a function of distance from receiving station indicate, that percentiles above 80% can be difficult to determine, and thus 75% is selected as a possible operational definition of the best coverage quality that is commonly practically achievable.
3) Long Range AIS coverage (interfering range): In order to keep a last known position of vessels within a certain operational entity s area of responsibility updated, or in order to perform vessel transit statistics, longer reporting invervals are quite acceptable. Compared to LRIT requirements of 6 hour reporting intervals, with an option to poll or request for the minimum reporting interval of 15 minutes, a QAIS,Predicted of 10% will statistically result in the following performance for class A AIS equipped vessels 5 : QAIS,Expected = 10% Moving vessel Nominal reporting rate: 30 seconds (or less) Vessel moored or at anchor Nominal reporting rate: 3 min Probability of one or more position reports received every 6 hours Probability of one or more position reports received every 15 minutes Better than 99,99 % Better than 95 % Better than 99,8 % Better than 75 % Using coverage definitions in relation to operational services It is intended to use the coverage definitions proposed above in good practice recommendations for operational services. The coverage definitions are intended to define the relevant geographical extent of a given service. System reliability and network considerations such as redundant coverage or data sampling may further need to be taken into account, to describe whether availability requirements for a given operational service can be met. Operational service AIS DGNSS broadcast low risk area AIS DGNSS broadcast high risk area VTS Information service VTS Navigational assistance service VTS traffic organization service LRIT equivalent AIS monitoring To be continued Recommended Recommended Recommended Reference AIS coverage Availability calculation period Good AIS coverage >99,5% (signal in air) 2 years IALA R-121 Good AIS coverage >99,8% (signal in air) 2 years IALA R-121 General AIS coverage?? IALA V-119 / IALA VTS Manual? General AIS?? IALA V-119 / IALA coverage VTS Manual? Good AIS coverage?? IALA V-119 / IALA VTS Manual? Long Range AIS coverage??? 5 The extent of coverage areas with QAIS percentiles at 20% and below have however proven difficult to determine empirically, due to the extreme sensitivity to weather and abnormal propagation phenomena. Longer observation periods may be needed to determine the extent of an average 10% range, or the range must be derived by radio propagation simulation.