European Commission logo
français français
CORDIS - Résultats de la recherche de l’UE
Contenu archivé le 2024-05-30

Trusted Vessel Information from Trusted On-board Instrumentation

Final Report Summary - TRITON (Trusted Vessel Information from Trusted On-board Instrumentation)

Executive Summary:
As a matter of fact, Ship Reporting Systems (SRS) are today the backbone of maritime surveillance and control, but their contents can be counterfeit by malevolent operators. Nowadays, intentional attacks to SRSs are increasing and the advances of mass-price technology make such attacks a potential and serious threat for the civilian and commercial maritime surveillance.

TRITON stands for TRusted vessel Information from Trusted On-board iNstrumentation and aimed at improving the intrinsic robustness of some on-board equipment used in SRSs, namely the Automatic Identification System (AIS) and the Global Navigation Satellite System (GNSS) receiver. The project wanted to increase the awareness on the problem of cyber-attacks and contributed to increase the overall trustworthiness of SRSs and, in turn, the security of the maritime domain as a whole. From a general point of view, the project focused on a twofold objective:

• the first concerned a robust GNSS receiver, able to mitigate intentional jamming and spoofing attacks (complementing GNSS authentication mechanisms foreseen by the next generation of Galileo signals), providing a “trusted” satellite-based source for Position, Navigation and Timing (PNT) data. The TRITON team worked on the design and development of proper signal processing algorithms for the detection of intentional interference over legacy signals. The use of Galileo signals added value to the demonstrations, that revealed the importance of a multi-GNSS scenario to intrinsically increase the level of security associated to some cyber-attacks;

• the second dealt with a secure communication link from the vessel to vessels and from vessel to shore base stations. The TRITON team worked on the design and development of an additional communication layer over the standard Very High Frequency (VHF) band used by AIS, exploiting the frequency diversity provided by the “white spaces” segment of the Ultra High Frequency (UHF) band. This improved the reliability of current communication links, ensuring the necessary Quality of Service.

The effort resulted in the realization of prototype, that serve to demonstrate the “proof of concept”. The performance of such nav/com prototype were assessed through a dedicate test campaign at the Joint Research Center (JRC) of the European Commission, in Ispra Italy.

The technical and scientific work was complemented by an early business assessment of the technical solution, through a Cost-Benefits Analysis (CBA). This served to identify key added values of the new solution compared to the system currently in use. Finally, a specific analysis was carried out to assess the existing corpus legis in the field of maritime surveillance and identify rules and legislative proposals at EU and international level related to the topic studied in TRITON. This analysis provided a set of recommendations in order to address the major results of the project to the most suitable working groups and regulatory committees involved in maritime security affairs.

Project Context and Objectives:
1.2 Context and project objectives
In the frame of the maritime domain awareness (MDA), the ship reporting systems are the main source of information. Vessel unique ID, date, time, position, course and speed are some of typical information automatically and periodically provided by these systems (e.g. each hour, at least 4 times a day, etc. depending on the applicable regulation).
Systems such as AIS (Automatic Identification System), LRIT (Long Range Identification and Tracking) and VMS (Vessel Monitoring System) are today the backbone of maritime surveillance, safety and security. Further non-cooperative systems exist as well (e.g. coastal radar), but data provided by ship reporting systems is taken for the most part at face value.
1.2.1 The role of GNSS
Services based on Global Navigation Satellite Systems (GNSS, including GPS) are currently in use for navigation, precise positioning and timing reference and synchronization in a large number of critical systems. In the maritime sector, GNSS-based services include ocean and costal navigation, vessel collision avoidance, fisheries monitoring and emergency management.
GNSS signal power levels are extremely low due to the large satellite-receiver distances. The use of GNSS in safety (e.g.: collision avoidance) and liability critical (e.g.: fishing monitoring) applications rises concerns about intentional interference, which attempts to deliberately disrupt nominal GNSS operations. In this context, attention must be paid to the threats posed not only by unintentional and intentional jamming, but also, even more, by spoofing. Spoofing refers to the transmission of counterfeit GNSS-like signals that force the victim receiver to compute erroneous positions. Spoofing’s objective is to convince the user that he is somewhere he is not.
1.2.2 The role of the communication segment
In principle, naval safety can be split into autonomous and cooperative. The former can be identified by all the autonomous actions of the vessels, such as their respective positioning and communications towards a central control station (also through satellites). This constitutes a first layer. However much more safety services can be designed by the cooperation among nodes. This, indeed, requires a secure and robust communication segment.
In fact an ideal communication segment for cooperation among vessels should provide the architecture with the following characteristics: a) Distributed approach; b) Scalability; c) Flexible authentication infrastructure; d) Robustness against DOS attacks and selective jamming interferences; e) Robustness against possible hidden terminal scenarios; f) Broadband transmission resources. Unfortunately, current AIS systems cannot fulfil all these requirements, and basically satisfy only the first two, namely a) and b). TRITON does not criticize current AIS solutions, but rather proposes to enforce them, based on techniques which have been developed in the meantime and to complement them with wireless resources which have been recently freed worldwide (white space bands).

1.2.3 The concept
The TRITON project aimed at improving the intrinsic robustness of the on-board equipment of ship reporting systems (e.g. AIS-transponder, VMS-equipment, LRIT-transponder), contributing to increase the overall trustworthiness of these cooperative systems and, in turn, the security of the maritime domain as a whole.

Figure 1: The role of the on- board unit of a ship reporting system.

As critical part of a cooperative reporting system, sketched Figure 1, the on-board unit typically:
1. gathers significant information concerning the navigational status of the ship, such as its ID, accurate position, course over ground (COG), speed over ground (SOG), heading, rate of turn (ROT) and other safety-related data. To do this, the on-board unit (the transceiver) relies on the data provided by both positioning systems and other navigation sensors (gyrocompass, ROT indicator, etc.);
2. transmits the information via a radio link to be received by other vessels and/or shore-based stations;
3. receives similar navigational data from other ships (basically for anti-collision avoidance) and/or data from shore-based stations (i.e. local information on weather, restriction to navigation, differential corrections, etc).
Taking into account this scenario, the TRITON project focused on a twofold objective: (i) the first concerned a GPS/Galileo receiver robust to intentional jamming and spoofing attacks; (ii) the second aimed at enhancing the robustness of the communication link from the vessel to vessels and from vessel to shore base stations. Hardening GNSS receivers against spoofing and jamming
Acknowledging the primary role of GNSS to support present ship reporting systems, the TRITON project focused on the GNSS-based positioning systems which interface with the on-board unit. The aim was to provide to the on-board unit a “trusted” GNSS-based source of positioning and timing information, robust to some intentional jamming and spoofing attacks, supporting the purposes of a robust ship reporting system. This is sketched in below.

Figure 2: The role of the robust GNSS receiver under development in the TRITON project.

Particularly, the TRITON project analysed, studied, designed and developed possible countermeasures against GNSS spoofing which can be implemented in GNSS receivers typically used in maritime applications. Enhancing communications
TRITON proposed to enforce AIS reliability and safety, by adding a new communication channel in the UHF band. The proposed approach is based on the introduction of a secure communication module on top of a standard AIS transceiver. This module exploited the “white spaces” freed by analog TV, offering a broadband channel enabling several services and enhancements to the current system.
As shown in Figure 3, the two modules will be connected via a cabled link, on which they will share different type of information: the AIS module sends a copy of its messages (for redundancy) as well as a snapshot of the current VHF channel allocation; the WSC module, after a computation based on the channel perception of his neighbours, sends back indications for an optimal slot allocation, implementing new mechanisms for guaranteeing a secure communication in the VHF segment.

Figure 3: Architecture of the White Space Communication layer

The main issues related to security concern the lack of mechanisms guaranteeing authentication, integrity and reliability of messages. In particular, the widely used techniques for securing other families of communication networks do not suit this specific case, mainly in terms of bandwidth.
Moreover, the current AIS systems use a limited and “closed” set of messages, resulting in a low flexibility of the message structure for what concerns possible extensions enabling security.
1.2.4 TRITON objectives
The main objective of the TRITON project was the improvement of the on-board unit of ship reporting systems, contributing to increase the overall trustworthiness of these cooperative systems and, consequently, the security of the maritime domain. Particularly, the objectives defined at the beginning of the project were:
▪ the implementation of GNSS anti-spoofing techniques for improving the robustness of on-board ship GNSS receiver;
▪ the use of GNSS jamming detectors to warn users in case of intentional attacks on legacy signals;
▪ to enhancement of communication standards, aiming at improving the security and reliability of ship-to-ship/ship-to-shore communications, through the introduction of additional services for security purposes;
▪ the analysis of “cooperative positioning” concept borrowed from car-to-car (road transportation) to ship-to-ship (maritime segment), aiming at improving the availability of ship positioning/timing service in case of GNSS outages;
▪ the design, implementation and validation of a “proof of concept” prototype platform;
▪ an early business assessment of the technical solutions proposed by the TRITON project, through a Cost-Benefits Analysis (CBA) to identify key added values of these solutions over the currently available technologies;
▪ the analysis of current EC policy and regulations to provide guidelines considering future maritime applications and advances brought by project results;
▪ the dissemination of the project results in a final workshop, inviting interested stakeholders.

Project Results:
1.3 Main scientific and technological results
This section summarizes the most important scientific and technological results achieved along the project. These are explained in a simple fashion and, as far as possible, in a non-technical language.

1.3.1 Results of the initial desk analysis
The first part of the project, from T0 to M4, was dedicated to an initial desk analysis. The WP2 established a common baseline on current SRSs depending on positioning data originating from GNSS or similar navigational tools. This WP was divided in 4 different tasks to carry out a complete analysis from different perspectives (i.e.: technological, business and legal/regulatory). This section reports only the major results related to technological aspects. Dependency of vessel monitoring systems on GNSS
The description of the state-of-the-art of SRSs reported in Section 3 of [AD. 3] demonstrates that many of them depend on positioning data originated from GNSS. Among all SRSs, the AIS is a fundamental piece of equipment for vessel traffic control through VHF data channels. AIS is a very common piece of equipment. According to IMO SOLAS (International Convention for the Safety of Life at Sea) (Chapter V, Regulation 19), functional AIS equipment is mandatory on 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. This requirement became effective for all ships by 31 December 2004.

AIS stations use an internal GNSS receiver to estimate the vessel positions and have a reliable time reference for synchronization purposes. AIS also accepts GNSS data from external devices. Figure 3 1 shows the AIS 300, manufactured by Kongsberg, that has been selected as a key component of the prototype developed in the rest of the project.

Figure 4: AIS 300 mobile station included in the TRITON prototype

Result of the analysis: Many commercial AISs, like that reported below, embed single frequency (i.e. 1575.42 MHz), mass-market GNSS receivers, generally employed in consumer grade devices. This type of GNSS receivers do not have any barriers against intentional interfering attacks on legacy signals. Vulnerabilities due to intentional interfering signals
GNSS receivers used in maritime transportation are vulnerable to different type of intentional interference. The classified Annex of [AD. 3] reports a detailed overview of possible attacks to GNSS receivers, ranging from jamming (i.e.: RF signals blockage, by deliberately emitting electromagnetic radiation to disrupt the receiver, through the reduction of the signal to noise level) to more sophisticated spoofing. This last refers to the transmission of false GNSS-like signals, with the intent to fool the receiver and produce false information).
The problem of intentional interfering signals against GNSS-based devices is an actual issue, as demonstrated in the classified Annex of [AD. 3] by the high number of references (i.e.: scientific papers, news, web links, etc.), showing the disruptive effect of interference in marine applications. It is worth of mentioning that a high number of articles and scientific publications on this topic were noticed even after the end of the desk analysis, and demonstrated once again that the research community is seriously considering the problem of cyber security. For sake of completeness, we report three different quotes of distinguished experts, that explain well why the problem of interfering signal is an actual concern.

• Authors of [AD. 4] reported that ‘Although GPS jamming incidents are relatively rare they can occur; and, when they do, their impact can be severe’.

• On February 13, 2014, the Financial Times, published an interview to Professor Bradford Parkinson [AD. 5], focused on the security of GPS: [...”We have to make the GPS system more robust...our cellphone towers are timed with GPS. If they lose that time, they lose sync and pretty soon they don’t operate. Our power grid is synchronized with GPS [and] our banking system’].

• [AD. 5] is concluded with a reference to the maritime sector, reporting a sentence of Professor David Last, a consultant to the UK’s General Lighthouse Authority: [...”When a ship loses GPS, it isn’t like a car satnav. Multiple systems fail simultaneously”]. Examples of systems on board of vessels that fail due to a GPS outage are: the AIS station, the ship’s gyro calibration system, the Electronic Chart Display & Information Systems (ECDIS).

The desk analysis on the vulnerabilities of SRSs, considering a list of known threats was carried out taking into account the level of feasibility (intended as the complexity required to accomplish the attack) and the residual risk left by countermeasures already in use or envisaged.

Result of the analysis: according to the analysis reported in the classified Annex of [AD. 3], the team concluded that jamming is currently the most dangerous attack, because it can be easily accomplished with (illegal) low cost devices available for purchasing on the web. Although it is detectable, it is hard to mitigate. Only back-up terrestrial technologies (i.e.: e-Loran) or sophisticated countermeasures, and still expensive, (i.e.: antenna arrays) can tackle jammers. Meaconing is potentially harmful, since it is the simplest form of attack (i.e.: reception, delay, and rebroadcast of radio-navigation signals), but can be detected and mitigated with appropriated methods. Spoofing seems not yet common and understood by most, even if some real attacks have been reported. These have been carried out by fraudulent fishermen to cheat monitoring systems. Spoofing is a growing concern, that can be detected and mitigated using methods not yet state of the art. More sophisticated spoofing attacks (e.g.: those involving high gain antenna) are considered unlikely, at least for civilian applications, even if they can be severe and difficult to detect.

The WP2 also investigated the vulnerabilities of the AIS-bases communication segment. The analysis confirmed that the data used to track maritime activity worldwide is increasingly being manipulated in order to disguise a ship’s identity, location or destination port. Vessels tracking, using AISs, is becoming subject to fraud, manipulation and spoofing, with severe implications for global trade and maritime security.

Result of the analysis: according to the analysis reported in the classified Annex of [AD. 3], the team concluded that jamming against AIS, S-AIS, VPCS and LMR is a highly feasible threat that can completely destroy communications. It is possible to detect it and some mitigation measures are feasible. All the forms of spoofing of AIS/S-AIS (with the exception of the unintentional form) are potentially harmful, because they are already accomplished at present, using low-cost equipment.

1.3.2 From users’ needs to new solutions
Part of the WP2 served to collect users’ requirements through dedicated interviews to selected stakeholders and experts. Such requirements, along with the results of the investigation on the cyber threats, were the inputs of the WP3 “Robust Ship Reporting System Specification”. Such a WP3 acted as the bridge between the desk analysis and the development phase of the TRITON prototype. The methodology followed in WP3 has been detailed in [AD. 6][AD. 7] and is sketched below.

Figure 5: High level diagram of the methodology used to convert users’ requirements into specifications

Such a methodology can be summarized in two main consecutive steps:

• The user requirements were critically reviewed, filtered and associated to the Trusted GNSS Module (TGM) or to the Robust Communication Module (RCM) – Task 3.1;
• The requirements were translated in specifications – Task 3.2.

Result of the analysis: at the end of the WP3, a set of clear and concise specifications was available either for the TGM and the RCM. Such specifications drove the design of the two prototypes.

1.3.3 Development of the TRITON prototype
The prototype developed in TRITON was composed by two components:
• The Trusted GNSS Module (TGM), featuring mitigation algorithms against intentional interference;
• The Robust Communication Module (RCM), featuring a secure UHF channel in addition to conventional AIS links.

The TGM was developed in the WP4, whereas the RCM was developed in the WP5. Both the WPs run in parallel for 12 months. In addition, the WP6 was dedicated to functional lab tests, prototype integration, software tuning, and test campaign at JRC. The Trusted GNSS Module
The algorithms implemented in the TGM were selected after a careful analysis of those introduced and explained in the scientific literature, considering their performance (in terms of an increased robustness and more accurate measurements in the presence of interference) and complexity. The TGM was developed in software radio technology, that allowed for the full design and implementation in a short period of time. In addition, the software radio development (i.e.: the receiver consists of software routines running on Personal Computer) introduced the level of flexibility required by the performance assessment of innovative algorithms. For instance, the receiver was tested under different configurations with a variety of possible settings to enable (disable) algorithms. The TGM, showed in Figure 6, was used as source of PNT data to the RCM.

(a) (b)
Figure 6: RF front end naked board (a) and TGM composed by the RF front end the real time GPS/Galileo software receiver running on a PC (b)

Result of the development: looking at the results of some lab tests during the WP4, we can conclude that the monitoring of the Automatic Gain Control (AGC) level, coupled with the observation of the signal statistics, is a valid method to detect unexpected signal power increments. Simulations and lab tests confirmed that the algorithm considered in TRITON outperforms conventional methods, because it is able to detect also the presence of weak interfering signals.
Some types of spoofing attacks induce a distortion of the received signal. This can be detected by monitoring the output of the correlation banks, with a negligible increment of receiver complexity.
The results stressed the importance to have an aggregated test at the receiver logic. Such test takes as input the results of all monitoring algorithms implemented along the receiving chain and provides a single measurement on the level of trust of the estimated position and timing data.

The level of trust (i.e.: warning in case of detection of interference) is an important piece of data generated by the TGM, that on board of vessels should be sent to other devices slaved to the GNSS receiver. A possibility is to dedicated one (or more) field of a specific message of the NMEA protocol. Currently, GNSS receivers lack in providing a measure of the level of trust of their estimates and no messages of the NMEA consider this option. The Robust Communication Module
The RCM featured a secondary channel, that enhanced conventional AIS communication, with respect to security-related issues and to the limited bandwidth. The most suitable technology, proposed as a candidate for the secondary channel, was selected after a preliminary analysis, taking into account the maritime environment, and comparing different systems on two main points: achievable data rate and maximum coverage area. Note that in parallel to integration and testing of the RCM prototype, an intense simulation campaign was performed to investigate some aspects related to the Cooperative Vessel Positioning concept (techniques for recovering position in absence of GNSS availability exploiting the WSC layer) and to the optimal allocation for time-slotted transmission protocols.

The RCM showed in Figure 7 was developed integrating two main components: a commercial AIS Class A module (AIS 300 produced by Kongsberg) and a White Space Communication (WSC) module, that is an embedded PC with wireless transmission capabilities on the 700 MHz TV White Space (TVWS) band.

Figure 7: the RCM prototype

Result of the development: from a technical perspective, the 700 MHz TVWS band demonstrated to be suitable for maritime communication to enhance current links.
In addition, the WSC was able to accept data coming from the AIS forward them through the high bandwidth wireless, encrypted channel. This implemented a backup communication segment, where transmitting stations could be authenticated. The integration has been carried out by means of standard interfaces, without modifying the existing, standard devices.
The activities performed in WP5 were invaluable to assess that the presence of a dual-frequency system, by design, strengthened the RCM against jamming on a single frequency. In addition, the WSC wireless link has plenty of bandwidth, that can be exploited for enabling multimedia services related to security (i.e. emergency calls among vessels). The simulations of CVP showed promising results, with a positioning accuracy less than 10 m for radars and of 15 m for some of the Time of Flight (ToF) techniques.

The complete prototype was successfully integrated and validated at the ISMB lab, during the Integration Readiness Review (IRR).

Figure 8: TRITON team working during the IRR meeting

1.3.4 Test campaign at the JRC and off-line analysis
In order to effectively validate the advantages of the TRITON prototypes in terms of security and reliability with respect to the state-of-the-art devices, a dedicated test campaign has been performed either on the TGM and the RCM. Some of the tests were conducted in the anechoic room of the Joint Research Centre (JRC), in Ispra (Italy), while others at the ISMB laboratories. During the tests, both the prototypes were stressed with intentional interfering signals to prove their enhanced performance with respect to commercial devices.

The results of the tests are included in the classified deliverable [AD. 14] and the main conclusions on the assessment of the two prototypes can be summarized as follows:
• Conclusions on TGM attack detection technology. The suite of attack detection tests included in the TGM allowed for robust detection of all attack types tested during the test campaign. In fact, the combination of several tests (e.g. Goodness of Fit, Ratio Test, PVT tests, AGC monitoring) was able to successfully detect, mitigate (when possible) and prevent large position errors in all of the attack scenarios considered in this project, which included jamming, spoofing and meaconing
• Conclusions on RCM assessment. The tests performed on the RCM underlined its main features and its key role in enforcing maritime communication security. Both device functionalities and new features have been tested. Specific tests have been set up in order to evaluate the performance in different scenarios. All in all, the main features of the WSC layer have been assessed, as well as possible enhancements enabled by this secure broadband link. Among those, message authentication, QoS management on the wireless link and VoIP emergency call availability are interesting features from the maritime security point of view.

1.3.5 Scientific and technological results versus initial objectives
The following table assesses the results of TRITON versus the original objectives defined in the Description of Work (see pages 16-17 of [AD. 8]).

Scope and objective of the topic Compliance How TRITON addresses the specific topic Results achieved
Ship reporting systems (mainly LRIT, VMS and AIS) are today the backbone of maritime surveillance, control, safety and security. However their contents can be faked (spoofed) by malevolent operators. ✓ TRITON will analyse the current reporting systems in order to recognize their weaknesses and to identify means to increase their robustness. TRITON dedicates the WP200 to such analysis. During the Task 2.3 “Security Threats of Ship Reporting Systems”, the team deeply reviewed the scientific literature.
The classified Annex to D2.1 contains 50 very recent references, reporting on the topic of SRSs vulnerabilities.
Research is needed into techniques to verify the sources of these messages [the messages provided by ship reporting systems (LRIT, VMS and AIS)] ✓ TRITON proposes a concept of trustable information transmitted from a trustable on-board equipment:
• The position, velocity and timing information provided by the on-board GNSS receiver can be trusted, because the GNSS equipment features jamming detectors and spoofing countermeasures;
• The reported information received from the radio link can be trusted, because the transmission protocol is made more robust;
• The reported information from a ship can be checked through algorithms based on cooperative positioning, intrinsically robust against single malevolent operators. TRITON developed, tested and demonstrated a nav/com prototype, enhancing the reliability of current devices:
• Some of the tests performed at JRC showed that when the TGM is the positioning data source for the AIS, it outputs correct data, even during an interference attack. When the AIS uses its internal receiver, it is vulnerable;
• Some of the test performed at JRC showed that the additional channel over the WSC band introduces redundancy and guarantees data authentication;
• Simulations performed in WP5 have shown that in case of GNSS outages, cooperative approaches using communication signals can provide a valid alternative for positioning.
This should also take into account the use of ship navigation radar for ship detection and tracking. ✓ The focus of TRITON is on the trustworthiness of the on-board equipment, being a fundamental part of the overall architecture of vessel monitoring systems. For this reason, on-board radar equipments may be one of the sources of information to be exploited in a cooperative vessel positioning approach. The results of the simulation of cooperative vessel positioning have been detailed in D6.2.
For cooperative approached, both Time of Flight and radar based methods can ensure a good position accuracy. Among the ToF-based methods, the iterative Least-square method is preferable and it achieves very close results to the radar-based method, whereas the ARPA radar-based method is the most performant since it combines range and bearing measurements.
The authorities in the EU [etc.] need to be ahead of this potential security hole by evolving counter-spoofing methods.[etc.]. Counter-measures should be developed at EU (and global) scale. ✓ Counter-spoofing methods are elaborated in the TRITON project at two levels:
1) inside the GNSS equipment, so as to make its information still trustable at face value, without the concern of a misleading information caused by a spoofer. This is an action impacting most on the equipment manufacturers.
2) at the transmission level, by exploiting the possibilities offered by an additional redundant channel in the AIS transmissions. This is an action impacting on a global scale, as it has consequences on the spectrum management and is related to recommendations. The deliverable D7.1 “Conclusions and Recommendations” summarizes the most important results achieved during the project, that are explained in a simple fashion and, as far as possible, in a non-technical language.
The section 4 of D7.1 reports a set of clear and concise recommendation from a technical perspective, touching GNSS and communication aspects.
The project would anticipate the need to further verify (double-check) ship reporting systems. ✓ The two main focuses of this project (GNSS equipment and communication transceiver) work together toward the goal of providing a trustable ship information to the monitoring systems. This way, the stringent necessity to resort to “literally” double checking the reported information can be partially relaxed.
On the other hand, the concept of cooperative vessel positioning can be seen as an additional method for double checking stand-alone ship information. In a nutshell, redundancy and frequency diversity are simple, but effective, means to tackle interfering signals. This was demonstrated and stressed either for the GNSS and the communication part. See the technical recommendations on:
• GNSS multi-constellation;
• GNSS dual frequency;
• AIS communication.
It should analyze possible different approaches and identify the most appropriate ones. ✓ Identify requirements and specifications applicable to the development of GNSS and communication modules.
Several possibilities to realize the trustable on-board equipment will be explored, for both the anti-spoofing GNSS equipment and the robust communication module. The project dedicated a specific WP (i.e.: WP3 Robust Ship Reporting System Specification) to the translation of users’ needs to technical specifications. These were the input for the design of the TGM and RCM.
Thanks to the software radio approach, the prototype offered a high level of flexibility and a number of settings and comparisons were possible during the tests and the off-line analysis.
The results of this project are expected to close a gap in the security of maritime domain. ✓ The final goal of the TRITON project can be described as:
• To harden the ship-board GNSS module in terms of robustness against spoofing attacks (achieving then a “trusted” source of position/timing information for ship-board unit of reporting systems) and ability to detect interfering signals (i.e.: intentional jamming and unintentional interference).
• To propose means to ensure/increase data integrity sent from vessels
• To globally increase the counter-spoofing capability and then the trustworthiness of the on-board segment of vessel monitoring systems, thus “contributing to close a gap in the security of the maritime domain”. Looking at the main scientific and technological results, we concluded that
• Intentional interfering signals are a growing security concern, but it is certainly possible to detect and mitigate them, with reasonably low probability of false alarms;
• Sophisticated algorithms at signal processing level can be combined with simple cross checks to enhance reliability
• More trusted and authentic data means more security for maritime transportation. Technical results need to be addressed to specific working groups focusing on cyber security and countermeasure.

Potential Impact:
1.4 The potential impact
This section summarises TRITON potential impact, presenting main socio and economic benefits, and main dissemination activities already described in [AD. 9] and in [AD. 10].

1.4.1 TRITON contribution to the maritime domain
The maritime domain provides a broad pathway for a wide spectrum of illegal operations (affecting maritime security) and accidents (impacting on maritime safety). A preliminary estimation of worldwide/ European impacts from illicit activities and ship accidents are presented below (for details and sources see [AD. 9])

Figure 9: Maritime security and safety: some facts and figures
As it can be noted, costs due to illegal operations and accidents each year are extremely high and are caused by a handful of events (such as in case of piracy or terrorist attacks). Therefore, even a single attack produces very high costs to various stakeholders, if we consider e.g.:
• Ship/ fleet owners that could be subjected among others to higher insurance premia, costs for ransoms, extra fuel measured in tons per day required to divert around affected areas, costs for deterrent equipment and personnel;
• Public authorities/ administrations that should face these activities through dedicated operations and naval forces; and
• Society that pays in terms of security and safety e.g. for passengers, higher price for final products/ goods transported, potential trade and price inflation, decrease in potential jobs (e.g. for legal fishermen) and available sea resources. This impact could, in turn, increase poverty and exacerbate conflicts.

Speaking about illegal operations, nowadays, they are perpetrated through even more sophisticated technologies than in the past. This could mean an overall increase in potentiality and severity of these actions in the future. Considering the high impact of each event, this increase could represent a primary risk in this domain. Experts, such as Dee Ann Divis in [AD. 11] and several newspapers (e.g. Financial Times, see also [AD. 12]) pointed out that cyber security is emerging as the hidden threat to shipping. Today, technologies, e.g. portable jammer, can be used easily for illegal activities, or like in the case of Chinese vessels, they count for 44% of worldwide GPS manipulation to allow them for illegal fishing.

This kind of GPS interference was analysed in the TRITON project. Together with them, ship-to-ship and ship-to-shore communication was also a key topic, considering its relevance to better face security issues. Moreover, failing communication networks could cause various types of accidents and endanger emergency response. In this context, the project has to be considered as a contribution to a more programmatic, wide and structured answer for some of main cyber security threats affecting this domain. In line with that, the developed solutions (i.e. the two embedded modules, the TGM and RCM) should not be intended as stand-alone technologies resolving the problems faced during the project, but as a part of on-board equipments enhanced by them. In particular, TRITON aims at improving current on-board Automatic Identification System transceivers (AISs). For this reason the related market should be considered as the key one.

At the same time, TRITON marketable components (i.e. AIS TRITON++ full optional or in one of its options – the TGM or RCM one – and TRITON intended as a test platform) could offer unique potentialities and features in the AIS market. Actually, according to our analyses, no comparable solution has been registered. Some projects on similar topics have been identified; however no commercial solution is commercially available or present in the market. Other technologies are used for similar purposes, such as radars or SatCom. Nevertheless, these systems are less common in the maritime domain than AISs. Moreover, they have higher costs to reach only few of the potentialities proposed by the two developed modules.

Thanks to its unique features, TRITON is expected to bring different types of benefits, in some way reflecting the major costs currently sustained by the maritime domain due to illegal activities and safety concerns. These benefits could be reached leveraging TRITON modules, together with (and not in place of) other technologies already adopted. In fact, a comprehensive approach is required to effectively face the analysed challenges. As a part of this approach, TRITON could impact the key maritime dimensions, bringing:
• A contribution to enhance maritime security, mainly anticipating a key future issue, i.e. potential cyber attacks;
• Even if to a smaller extent, a support to improve safety, through a better management of accidents/ collisions and a reduction of them; and
• Other potential benefits. TRTION impact on maritime security
TRITON contribution to maritime security enhancement is reflected in the support provided by the solution components for the reduction of occurrence and/ or the impact of main illegal operations, when linked to cyber techniques (in particular to jamming and spoofing perpetuated by third-parties). Moreover, some recent events (see also [AD. 12]) stress the importance of having a trustworthy positioning data for maritime security. This was, to some extent, one of the TRITON objectives.

Reducing the potentiality and the severity of illegal operations implies a decrease of the overall costs supported by the economies affected by them, with far-reaching worldwide benefits expected. Such benefits are split among different users and stakeholders, providing an overall socio-economic impact in terms of security aspects, as briefly summarised in the “impact table” presented below.

Contribution to enhance maritime security
Type of impact User and/or stakeholder Benefits, intended as a contribution to reduce:
Private-economic impact Private fleet owners • Insurance premia
• Costs for stolen and/or damaged ships
• Costs for potential ransoms
• Other direct or indirect costs
Public-economic impact
Authorities/ administrations • Costs for operations aimed at detecting, mitigating or hindering cyber attacks
• Costs of naval forces
Socio-economic impact
Society as a whole • Sea security risks (e.g. for passengers),
• Price for final products/ goods (lower costs would allow lower prices),
• Costs to regional economies (e.g. trade and price inflation, illegal fishing and related jobs, resource availability...)
• Related country’s poverty or conflicts
• Environmental damages and pollution.
Table 1: Impact on maritime security TRTION impact on maritime safety
Even if to a smaller extent (the project was primarily focused on security issues), TRITON components could be leveraged also in terms of safety improvements.
Thanks to more reliable, accurate and robust information (e.g. PND data), the possibility of leveraging a dual-frequency communication channel, and a corporative positioning, an increase in maritime safety in case of AIS TRITON++ adoption can be foreseen.
Here, apart from a better management of accidents/ collisions and a reduction of them, additional communication services and aids-to-navigation brought by the RCM could be expected, mainly by crew. The safety “impact table” shows the main benefits by impacted users and/ or stakeholders.

A support to improve safety
Type of impact User and/ or stakeholder Benefits, intended as a contribution to allow:
Socio-economic impact Main private and public maritime end-users and society • A reduction of number and/or severity of accidents and collisions
• An increase of accuracy on forecast-calculations made by the onshore and onboard systems when it comes to e.g. congestion prediction

Crew • VoIP emergency calls
• Port video surveillance
• Cooperative positioning
Search and Rescue (SAR) operators • Improvement of operations
Table 2: Impact on maritime safety Other potential impacts
In addition to those listed before, some other benefits could be brought by the system exploitation.
The scope of cyber attacks in future is expected to be even wider than that designed in this analysis. Though criminals will actively use these systems for illicit activities; it is also likely that governments, which have already used them, will continue to do that. As a potential consequence, countries seeking to establish strong cyber warfare capabilities are more and more likely to utilize jamming or spoofing for gaining considerable advantage in all of the main domains - not just within the maritime one. This capability could have several societal implications.

Finally, looking at TRITON as a test platform, TRITON could be used as a test tool for other receivers or systems, monitoring and verifying in this way their robustness towards cyber attacks. For this reason, TRITON could contribute to testing activities and the definition of minimum performance requirements for maritime navigation. Conclusions and fields of actions
The overall analysis of potential impacts brought by the TRITON solution allows some first conclusions. Costs supported by countries and economies for illegal operations and accidents are so important, that also a small contribution to their reduction could mean significant savings. This was demonstrated also during a simulated Costs Benefits Analysis (CBA) on security aspects. Assuming a minimum reduction of only three illegal activities (i.e. one piracy attack, a 0.1% of illegal fishing and one ship theft, all in one year), an annual saving of c. €51 m could be reached thanks to TRITON adoption. This potential impact could be achieved bearing only the marginal costs required to adopt the solution. At the same time, this saving could be translated into a financial benefit for ship owners and public authorities, with wider societal implications and effects (not considered in this preliminary estimation): affected countries and areas could reduce the weight that these illegal activities exercise on their economies and citizens.

Speaking of impact, it has to be mentioned that benefits are not only for system end-users and society as a whole, but also for system providers. In our Business Plan simulation, we considered an example of a small provider, part of the TRITON consortium, as part of an illustrative and possible business model. According to our results, the provider market share could grow up to ten times (from actual percentages) in the timeframe considered (ten years, from 2018 to 2028), becoming an important player in the market. Related financial figures reveal a small but relevant growth for a player currently selling c.50/ 100 new AISs each year.

The project pointed out that there is a strong attention on the topics promoted, at least from a research point of view. It was confirmed by recent newspaper headlines, as mentioned before (see also [AD. 11] and [AD. 12]). However, several difficulties have been encountered to exploit the solution potential from a commercial point of view. As of today, despite this interest, a commercial will of system providers and end-users seems not yet mature.

To foster this commercial interest, allowing reaping the benefits of an increased maritime security and safety, some actions seem necessary. In fact, only after these actions have been undertaken, a real improvement in technology used by end users can be expected. These actions can be grouped into different “fields of action”, i.e.:

• Promotion of a specific regulation. This is a fundamental driver in maritime domain for further technology improvement. Moreover, when it comes to cyber security regulations for the maritime community, several gaps seem to exist. A review and alignment of regulation could positively impact the overall domain, also from a commercial perspective. Specific tasks and activities during this project have been dedicated to this topic and a deliverable prepared. In particular, [AD. 13] was intended as a reference for the work of authorities and agencies regulating marine traffic and application;
• Demonstration of financial benefits. In this sense a project CBA and BP have been undertaken from a general and worldwide perspective. They tried to assess on one hand the potential project impact on end-users and society and, on the other, financial benefits for a potential solution provider. Moreover, considering the relevance of financial benefits in the adoption of new technologies, in our opinion it could be interesting to deepen the link between the adoption of an AIS TRITON++ (one module or full optional) or the use of TRITON test platform and some of the main costs sustained by end-users. In particular, the relation between more robust technologies and insurance premia presents a certain commercial potential and it should be analysed at a more mature stage of the technology development. If the developed components could contribute to a reduction of insurance premia and how they could contribute is an interesting topic (something similar has been studied in the road domain, where cars using GPS tracking where guaranteed lower premia, especially in countries with high theft ratios, like South Africa). These kinds of savings could revive the commercial interest of sheep owners (both public and private) towards the solutions due to the relevance of these costs; and
• Promotion of awareness on cyber security. Cyber security awareness in the maritime sector has been registered at a relatively low level. According to the first EU report on this challenge published by the European Network and Information Security Agency (ENISA) in 2011, cyber threats have grown steadily in the maritime sector, while awareness on them in the domain remained relatively low. In Germany and Norway, only recently, concerns start coming out from the authorities and the potential risk of cyber attacks is now recognised by the maritime industry. We assess that there is a medium level of awareness on cyber attacks among key actors in these countries, but it is mainly in terms of general considerations regarding future trends (more than current adoption of related systems). In fact, these topics are not considered as priorities or perceived as concrete risks at the moment. Therefore, TRITON dissemination and promotion activities were intended as a mean to enhance the overall awareness on threats faced during the whole project, as better described in the paragraph below.

1.4.2 TRITON dissemination, communication and exploitations
The figure below summarises the overall TRITON “promotion system” among main target groups, as defined in [AD. 10].

Figure 10: TRITON “promotion system”

During the project three different types of promotion activities have been undertaken:
• Dissemination actions. They were thought mainly for primary (i.e. end-users) and secondary target groups (e.g. regulators, experts, main maritime associations...) considering their impacts on core project issues and relevant direct and indirect benefits they could receive;
• Exploitation actions. Also in this case, primary and secondary target groups were considered key actors; and
• Communication actions. The majority of users were professionals and security was a concern. Therefore, specific dissemination and exploitation actions, more than a communication strategy for a wide audience, were identified as the most appropriate means since the beginning of the project.

Dissemination, exploitation, and communication were strictly connected and, in some cases, the same channels/ measures (e.g. website) have been used to undertake the activities, even if with a high level of customization (dedicated website sections) in relation to target audience and objectives.

Dissemination, exploitation, and communication activities have been undertaken both at consortium and partners’ level, as a part of an overall strategy composed by four types of actions in relation to their objectives. In this context, first of all an improvement of awareness concerning the main maritime security threats has been considered as a key point. In line with that, promotion activities mainly aimed at overcoming this gap. At end of the project, this assumption on which we built our promotion strategy was confirmed also by the “Business and Exploitation” analyses.

The table below summarises the promotion actions undertaken, presented by type of action, their objectives and a brief description. Moreover, possible actions identified at the beginning of the project vs those concretely undertaken are reported. Section 1.4 reports further details on main actions (e.g. description of the events and submitted papers).
Type of action Objective and Description Activities foreseen in the TRITON proposal Status of the action at the end of the project
Dissemination actions for awareness
(most relevant activities for TRITON project) Objective: set of activities aimed at improving awareness mainly among users on project topics and related technologies.
Description: the majority of these actions start immediately after preliminary project results and after the conclusion of first WPs. It is strictly connected with project outcomes. Logo ✓
Website ✓
Website prepared and regularly updated
Project leaflet/ brochure ✓
3 brochures and
2 posters
Articles and papers ✓
4 papers and
1 project advertorial
International conferences and events ✓
Participation to 7 events
Dissemination for user and stakeholder involvement Objective: engagement and involvement of relevant users and stakeholders in different phases of the project in relation to different objectives and activities.
Description: this type of action starts early in the project and could last sometimes the project timeframe. It is strictly connected with specific WP and/or Task objectives (e.g. identification of user needs) Contact database

2 different databases prepared and regularly updated
Set of interviews for identification of user needs ✓
2 set of interviews
Other potential actions for user and stakeholder involvement, if necessary and possible ✓
A dedicated final WS organised
Exploitation Objective: activities aimed at the market uptake of proposed solution
Description: this type of action is linked to the last part of project activities aimed at commercial exploitation of results Project Business Plan ✓
Similar and new projects monitoring and interface

Interface with European Institutions ✓
Event participation and a dedicated final WS organised
Communication Objective: additional communication actions (with regard to dissemination activities) to communicate project results not only to the main stakeholders, end-users or scientific community (primary and secondary target groups) but also to the society at large (third and other target groups)
Description: few potential communication actions (e.g. usage of specific web 2.0 social media) could be undertaken through the TRITON project, according to the security of the project Website ✓
Website used also to inform community on main TRITON events
Specific web 2.0 social media ✓
Even if only through project members’ personnel accounts, Twitter was partially used
Other communication tools ✓
Ad hoc usage of a specific contact database for UW promotion

Table 3: Overview of the dissemination, exploitation and communication types of actions
Concluding, a general and theoretical interest has been shown by different domain actors during the events where TRITON participated, through social media, when used, or with specific requests for more information related to the project. This interest confirms also the project findings coming from the business analysis.

1.4.3 Results achieved versus expected impacts
Expected impacts Description Results achieved
Increase the level of consciousness on the vulnerability of GNSS-based ship reporting systems As GNSS is continuing to be modernized (i.e.: new constellations, signals and services) and adopted in critical infrastructures, few recognize that severe system failures can be caused by intentional interfering signals.
The TRITON project will provide a detail analysis of jamming and spoofing threats for ship reporting systems, providing an assessment on the level of complexity associated to the attack and the corresponding risk. The work will provide clear recommendations for the improvement of real systems. In [AD. 1], a detailed overview of possible attacks to GNSS receivers, ranging from jamming to more sophisticated spoofing, has been provided. The desk analysis on the vulnerabilities of SRSs was carried out taking into account the level of feasibility (intended as the complexity required to accomplish the attack) and the residual risk left by countermeasures already in use or envisaged. A dedicated deliverable ([AD. 13]) for conclusions and recommendations for the improvement of real systems have been provided.
Hardening GNSS module used by on-board unit • By implementing jamming detection solutions (i.e.: the receiver will be able to recognized if it is being jammed or working in an interfered scenario, providing appropriate warning to the user)
• By implementing spoofing mitigation solutions (i.e.: the receiver will be able to detect and mitigate some spoofing attacks)
• By exploiting new methods for cooperative vessel positioning, exploiting technologies different from GNSS. • Through the TGM, featuring mitigation algorithms against intentional interference
• Through the RCM, featuring a secure UHF channel in addition to conventional AIS links and allowing cooperative vessel positioning
Improve the ability of the vessel’s crew to react to GNSS failures, promptly reverting to traditional means of navigation Some vessels have integrated systems that enable automatic execution of a passage plan on autopilot. If this system is operating when jamming occurs, the vessel’s course and heading may change without informing the crew, potentially leading to hazardous consequences.
The use of a “trusted” GNSS receiver (as that developed in TRITON project) able to detect jamming and rise warnings in case of temporary GNSS outages, helps crew to quickly react and take the most appropriated decision. In turn, this improves safety. As described in paragraph 1.4.1 the two modules together effectively contribute to enhance security and improve safety. In particular, for crew:
• The TGM can support the detection of GPS interferences, so consequent prompt reactions in these circumstances; and
• The RCM allows a second communication channel in case of jamming or spoofing of the first one (dual frequency) an a set of additional communication services and aids-to-navigation
Enhancing ship-to-ship and ship-to-shore communications By extend bandwidth or available communication channels as well as enhancing information integrity Thanks to the RCM, TRITON allows the introduction of a secure communication module, by exploiting the “white spaces” freed by analogue TV and offering a broadband channel enabling several services and enhancements to the current system
Development of a robust system prototype on top of current systems The prototype will be developed on top of equipment currently used on board of vessels. The test will demonstrated (and assess) how it is possible to improve the robustness of current systems, implementing low cost methods and alternative technologies. Both the two modules have been developed on top of a standard AIS transceiver. In particular, Kongsberg Seatex AIS 300 has been included in the TRITON prototype. At the same time, current AIS market was considered in the business assessment. The WP6 was dedicated to lab tests and test campaign at JRC aimed at assessing the robustness of current systems, implementing low cost methods and alternative technologies.
Improvement of collaboration between industry partners and research institutes across Europe By building up a scientific/technical team where industrial players, SMEs and research institutes (from different European countries and with different expertises) will work together for 2 years towards a common objective. The Consortium know-how and network was strongly exploited thanks to this 2 years collaboration. The mix of scientific and commercial backgrounds was fundamental for the in-depth understanding of user needs, from a technical and business perspective. At the same time, the technical support has been a key driver for the overall commercial exploitation.
Contribute from a legal/regulation perspective Together with the scientific analysis and the development of the robust system prototype, there will be the review of current EC policy and regulations. On the basis of the major outcome, the project will also provide recommendations from a regulatory perspective. While [AD. 2] provides a wide a comprehensive overview on current regulation, [AD. 13], i.e. project conclusions and recommendations, were intended as a reference for the work of authorities and agencies regulating marine traffic and application.

List of Websites:
1.5 Useful links and contacts
This section provides key contact details, presents some dissemination materials (i.e. logo and website) and provides useful links.

1.5.1 Project contact
Marco Pini – Project Coordinator
Head of the Navigation Technologies Research Area
Istituto Superiore Mario Boella (ISMB)
via P.C. Boggio 61, 10138 Torino (Italy)
office: +39 011 2276 436 | mobile: +39 335 6443351

1.5.2 TRITON logo and website
The TRITON logo aims at providing an immediate and visual indication of the forming blocks of the project: the maritime domain and the ship reporting systems. It is used in all materials produced within the project.
Figure 11: TRITON logo

The TRITON website, has been used as a key channel to disseminate the overall project and all relevant results achieved. It has been regularly updated during the whole project duration.

Figure 12: TRITON website
1.5.3 Useful links