Can Too Much Data Sabotage Maritime Domain Awareness Tactics?

A unified approach to persistent surveillance can cut through the noise and lead to success.

Maritime surveillance in our tech-forward age should be extraordinary. And in many ways, it is. A sector that once relied on unwieldy technology and a lot of guesswork has evolved into a cutting-edge industry that takes advantage of multi-faceted data sources and reaches across oceans and atmospheres. Big data is now a critical component of effective coastal operations management.

How is Data Used In Coastal Surveillance Applications?

From vessel positioning to manifests and personnel lists, from voyage tracking to incident reports, Maritime Domain Awareness (MDA) analytics can help mitigate potential dangers and monitor transports through to safe delivery. Data scientists can use that information to predict future outcomes and frame decisions, helping countries safeguard their assets and waterways.

Navies, coast guards, and shipping companies have a variety of options when it comes to sourcing information and monitoring their waters within the Exclusive Economic Zone (EEZ). Many still use mainstay tools like limited-range microwave radar and visual surveillance, as well as reporting from coastal automatic identification systems (AIS) and, more recently, satellite identification systems (S-AIS) that can cover a greater surface area and transmit information more efficiently. These are often supplemented by commercially available, third-party data packages. The most effective systems add High Frequency Surface Wave Radar (HFSWR) and integrated machine learning to the mix.

Inconsistent monitoring and delayed reporting can send enforcement teams on a wild goose chase.

Acquiring these data sets can be cost-efficient, but filtering through them for pertinence can be labour intensive. Information of immediate relevance, such as ship positioning and vessel identification, can be buried within these bulky analytics that also include things like navigation status, route plans, cargo descriptions, ship speed, turning radius, and emissions. Moreover, there are significant limitations to S-AIS that are often overlooked. In a 2021 report to the Philippine Council for Foreign Relations (PCFR Journal), Maerospace CTO Brian Franklin, and Maerospace advisor Dr. Tony Ponsford highlight the myriad challenges attached to these identification tools, “While most users focus on latency of collected signals, more fundamental problems with the data are a) the measurably intermittent detection, b) ‘dark targets’ and c) ‘Doppelgangers.’”¹

Expected coastal AIS signal intervals range between 10-30 seconds, depending on the size and speed of the vessel, but can spread to as much as 3 minutes between messages. S-AIS can have hours or even days between messages! “Dark targets” are bad actors who deliberately deactivate their identification systems, effectively disappearing from screens in order to pass through certain zones undetected. “Doppelgangers” are the result of signal falsification or accidental duplication. In addition to these complications, traditional maritime data mining is subject to inconsistencies or inaccuracies in self-reported information, such as crew lists or ETAs. Data quality can vary by input source, and is dependent upon several external factors, including location, electronic noise, traffic density, and even human error.

It’s a lot to sift through.

Sometimes, the problems are unintentional, but a significant number are driven by tech-savvy criminals, as well, leaving transport companies and border protection agencies vulnerable to not just misinformation, but also cyber-attacks and hacking. Radiocommunication jamming devices, non-compliance with AIS regulations, and pirated (sometimes by actual pirates), unshielded satellite receivers and after-market AVR cards can all leak radio interference, leading to an overabundance of information, and ultimately inaccurate surveillance data and failed interceptions.

In other words, the best MDA solution is one that can tune out the static.

Leverage the data for continuous accuracy

The plethora of AIS and third-party information is still inherently valuable but needs to be collated without delay to be most beneficial. A system that automatically sorts through the superfluous data and works in collaboration with other software and systems offers the most advantageous approach to coastal surveillance.

Maerospace’s third generation TimeCaster™ information processing software can do just that. Aggregating the most relevant AIS, S-AIS, and multi-sensor data into a time-synchronized array updated every ten minutes, it offers a near real-time, comprehensive picture of shipping channels and EEZ waters. TimeCaster™ works in partnership with FutureCast™ machine learning predictive analytics software and the trailblazing 4th generation PASE™ high-frequency surface wave radar to provide the most precise and cohesive maritime security coverage available today.

PASE™ is a global leader in mid-range persistent radar coverage. Employing sense-and-adapt technology, this cutting-edge system minimizes bandwidth interference and maximizes vessel-detection accuracy, boasting a reach of 20 to 220 nm, optimized to the full expansion of the EEZ. TimeCaster™ and PASE™ are designed to integrate seamlessly with C2 and C4ISR technologies. The result? Multiple extremely effective collaborative surveillance processes, delivered in one comprehensive system.

Find the Right Frequency

Information alone cannot protect cargo or borders, prevent illegal fishing, thwart human trafficking nor improve global shipping traffic. Raw data is just one factor in a complex infrastructure that keeps each nation’s coastal operations functioning smoothly. Leading MDA systems utilize a combination of tools to perfectly calibrate the overwhelming amount of information, fine-tune the sources, and stay ahead of the fleet.

1. Brian Franklin, VP Engineering/CTO, Maerospace Corporation and Dr. A.M. Ponsford, PhD; Breakthrough Solution for the Security and Protection of the Maritime EEZ; PCFR Journal; 2021