The existing train control systems in Europe are neither compatible nor interoperable. Before the advent of high-speed international trains this fact was not a major problem. International traffic did not make up a large proportion of rail traffic and international trains had to change their locomotive at borders anyway. Where regular cross-border traffic existed, country specific locomotives or trainsets equipped with the specific catenary and train control systems needed operated by national drivers provided bi-laterally agreed specific solutions in the past, such as the Benelux trains and the Amsterdam-Brussels-Paris TEE trains. The introduction of the high speed trains in Europe opened up a new market for high speed passenger traffic and allowed the European High Speed networks to be created. With the introduction of multiple unit high speed trains designed specifically for this market segment, such as the TGVs, ICEs and the Eurostars, the ATP/ATC situation became more complicated. These trains have to be multiple equipped with all train control systems of the countries they are planned to travel to when they are to be admitted on the networks.
The treaty of Rome, founding the European Community, mandates the European Commission to develop integrated international transport networks. The European transport policy calls for a revival of Rail transport, both Passenger and Freight. This is to be achieved by opening up the national rail networks to new entrants, a separation of the responsibilities for infrastructure and operation and a system of technical harmonisation removing barriers to cross-border traffic. The existing national ATP systems, most of which are not only incompatible, but also proprietary to a specific railway and/or supplier, was identified as an important obstacle to achieving these goals. The picture shows the diversity of ATP/ATC systems in Europe and was often used to illustrate the need for a harmonised solution for Europe.
Following the decision taken by the European Transport ministers in December 1989, the EC embarked upon a project to analyse the problems relating to signalling and train control. At the end of 1990, the European rail research Institute (ERRI) created a group of railway experts (ERRI A200) to develop the specification of requirements of ETCS. In June 1991, Industry ( Eurosig ) and Railways ( UIC, ERRI A200) agreed the principles of tight co-operation in order to consider the requirement specifications as the base for industrial development. The project framework included a new on-board equipment based on open computer architecture (EUROCAB), a new intermittent system for data transmission, (EUROBALISE) and a new continuous transmission system (EURORADIO). At the end of 1993, the EU council issued an Interoperability Directive and a decision was taken to create a structure to define the Technical Specification for Interoperability.
The High-Speed interoperability directive 96/48 mandates a system of interface specifications (TSI’s) between subsystems making up a high speed railway, of which ERTMS/ETCS is the command and control subsystem. This principle was subsequently extended to the Conventional Rail international network by directive 2001/16.
In 1995 the EC defined a global strategy for the further development of ERTMS with the aim to prepare its future implementation on the European Rail Network. The global strategy described in the "Master Plan of Activities" included the development and validation phase. The objective of the validation phase was to perform full scale tests on sites located in different countries (France, Germany and Italy). In the summer of 1998, Unisig, comprising the European Signalling companies was formed to finalise the specifications. The ERTMS specification, Class 1, was accepted on 25th April 2000, Interoperability tests were carried out on the Madrid-Seville (EMSET) test track and the Vienna-Budapest trials. The Test Track in Italy carried out trials in 2001 and trials in several other countries, such as Germany (Jüterborg-Halle-Leipzig), France and the Netherlands followed. The results of all these trials were incorporated in the consolidated specifications SRS 2.2.2C, which have been approved in 2005 and are the basis for the revision of the Command Control and Signalling Technical Specifications for High-Speed and Conventional Rail. There are a number of commercial projects at varying stages like the HSL-Zuid, Switzerland, Berlin-Halle-Leipzig, Athens and Madrid – Lleida. The Rome-Naples line was the first ERTMS/ETCS level 2 line to enter commercial service in January 2006.
The two major elements of ERTMS at present are the train control system (ETCS) and the common train radio system for voice and data communication GSM–R.
The European Railway Agency has been charged with the task of acting as the ERTMS/ETCS system authority and Change Control Board from the 1st of January 2006.
An example of a train especially built high-speed cross border traffic is the Thalys train. It has various train control systems on-board, in its PBKA (Paris-Brussels-Cologne-Amsterdam) and Thalys versions these would be the systems of France (TVM and KVB), of Belgium (Crocodile and TBL), of Germany (LZB and Indusi) and of Holland (ATB). Such a juxtaposition does not only cause problems with mounting all the necessary antennae, speed sensors etc., but also with the space restriction in the driver’s cab for all the controls and indications. In this case, displays were combined to save space and avoid confusion of the driver.
14 ATP / ATC Class B systems which were declared interoperable on the European High Speed Network in a transition phase until the unified European system is introduced in all member states.
Levels of application
Being a harmonised European system, ERTMS/ETCS has to comply with the very diverging requirements for ATP/ATC systems in Europe. The common system shall cover the wide spectrum from low traffic secondary lines to high speed lines carrying dense traffic. The technical solutions have to offer an economically optimised system. On the other hand, there should be no restrictions in the use of vehicles in the various railways; this means that the on-board equipment should be able to interpret all equipment levels installed lineside.
The system was designed to be applied on the following levels:
Level 1 is provides an intermittent ATP system, using spot or semi continuous transmission only. Conventional lineside signalling is the base, but the system can be applied without lineside colour light signals, using block-markers and indication lights, as practiced on the Dutch and Belgian sections of the Amsterdam-Brussels-Paris high speed corridor. On that line it provides a fall-back option from the normal ETCS level 2 operation.
Level 2 is also based on lineside signalling, but can be operated without lineside signals. The system is based on continuous transmission using the GSM-R radio link. Balises are required only to provide position references and fixed information such as line geometry. This level is suited for main and high-speed lines.
Level 3 systems of ERTMS/ETCS are intended to operate without lineside signals and track based train detection systems are made redundant by active reporting of the train position. The train reports its present position back to the control centre, which calculates the train end position by taking the train length into account. The separation between trains is the task allocated to the ATC centre called Radio Block Centre (RBC). Level 3 systems can operate in fixed or moving block mode. The absence of conventional train detection systems in level 3 requires trains to continuously monitor and report on their integrity, i.e. the fact that the train is still intact and therefore no cars are still occupying sections of track about to be included in a following train’s movement authority. Until now however unfortunately no proven technical solutions for train integrity monitoring exist, especially for locomotive hauled freight trains and this, combined with the complexity of arriving at harmonised specifications for a level 3 system, have led Eurosig to postpone the development work on Level 3.
For spot transmission, the Eurobalise was standardised. The technology is similar to the equipment used in Sweden (Ebicab) and France (KVB), but has improved performance. The trackside equipment is based on passive transponder technology, which means that it needs no external power supply for data transmission to the train. The coupling between the trackside balise and the balise transmission module underneath the train is inductive. The train constantly radiates a magnetic field with a frequency of 27.095 MHz, which is picked up by the wayside balise when the train is passing. The received power is used to feed the transmitter in the wayside balise, which generates a serial data stream back to the trainborne receiver. For the data transmission, a frequency range between 4 MHz and 5 MHz was chosen, the data rate is 600 kbit/s.
There are two applications of the Eurobalise, depending on the level of ERTMS/ETCS. In level 1, the Eurobalise transmits signal information, which is variable information. Other parts of the data telegram can be fixed, such as distance to the next balise, maximum line speed or line gradients. These balises are connected to a lineside electronic unit (LEU), similar to the one described for the SELCAB system. The other application is mainly in level 2 or level 3 systems for positioning purposes. As the train cannot derive information on its position from the radio transmission system, the balises carrying fixed information are used to allow the train to evaluate its position when passing a balise. The interpolation between balises is performed by an on-board Speed and Distance Monitoring Unit (SDMU). For this function, a selection of sensors, such as tachogenerators, radar speed sensors or others can be used, depending on the supplier’s choice.
For semi-continuous data transmission, that can be applied to transmit infill information in level 1 systems, a technology using a leaky cable is the preferred solution. The leaky cable in principle is a coaxial cable with the outer conductor carrying slots, so that a small portion of the transmitted power can leak out of the cable and be received by an on-board receiver, which can use the same antenna as the Eurobalise transmission unit. This so-called Euroloop is always carrying variable signal information, it has to be actively powered from the wayside, its length can be up to several hundred meters.
To study the application of radio for ATC purposes, a research program named DIBMOF (services integrating railway mobile radio) was performed. It confirmed the concept of using the same radio system for voice communication between driver and central and data transmission. The base for radio communication was the GSM (Global System for Mobile communication) system, which was specified in Europe and has become a world-wide standard. Most of the public cellular networks in Europe are compatible with the GSM standard.
The International Union of Railways took a majority decision to introduce GSM as the base for voice and data transmission between train and ground. For railway applications, an exclusive transmission frequency range below the frequency of the public GSM band was recommended and granted. A survey was made to collect all the requirements from the member railways for such a radio system. The evaluation led to a system called GSM-R, where R stands for railways. This set of requirements was passed on to ETSI (European Telecommunication Standards Institute), who generated, based also on the requirements of others, the set of specifications GSM 2+. Most of the additional features requested by UIC and exceeding the GSM 2+ standard are only related to voice services. There is, however, one future important feature which permits short headways by reducing the time for establishing a connection to a train. This service is based on Internet protocols and is called GPRS (General Package Radio System).
A specific problem in transmitting vital signalling data over an open radio channel is that of safety and security. European standards define threats against which suitable defences have to be established:
In the same way as in conventional systems, a protection against random errors caused by lightening or similar effects has to be in place. This is normally done by adding code bits to the message.
Data are passed through a number of devices with unknown behaviour from transmitter to receiver. Therefore, the transmission path is frequently called a grey channel, which means that it is not totally transparent for the user. Coding strategies have to be used which allow the receiver to detect and reject corrupted messages.
There may be an intrusion from external users of the open network which is common to public users and railway users. Therefore, protective coding has to be used to avoid unauthorised access to the data transmission between train and wayside.
Selection of the appropriate level of operation
The same criteria as for existing systems apply to ERTMS/ETCS. For the introduction or replacement of a simple ATP system, level 1 based on balises only is the most cost effective solution. There are, however, two aspects to be considered:
As with any intermittent ATP system, there has to be sufficient overlap in the track layout to allow a train to stop within the overlap when it passes the signal at danger with the release speed of the system. AS discussed in chapters 3.1.1 and 3.1.2, this release speed is a trade-off between line throughput and travel time on one hand and safety on the other hand.
As said above, the line throughput suffers from a system based only on spot transmission. In case of higher capacity needs, Euroloop has to be added to provide infill transmission or a level 2 system has to be chosen.
Level 2 would be the normal choice for main lines, being similar in performance to the LZB and TVM type of systems. As the GSM-R radio is in most cases also used for voice transmission between driver and dispatcher, the cost of the radio system can be considered part of the communication system and therefore does not need to be justified solely by the ATP/ATC system. This can make ERTMS/ETCS level 2 systems more cost effective than many of the existing continuous ATC systems in Europe. The number of lineside signals can be significantly reduced and be replaced by cab signalling. There is, however, the required availability and the fallback strategy in case of a failure to be considered.
As Level 3 systems of ERTMS/ETCS operate without lineside signals and track based train detection systems are made redundant by active reporting of the train position. The train integrity supervision is therefore a vital function, which up to now does not have convincing technical solutions for freight trains. If this technical problem can be overcome in the future, level 3 offers an implementation which should be very effective in terms of investment and maintenance cost, but also in terms of Occupational Health and Safety as it reduces the amount of equipment mounted close to or even inside the fouling gauge.
The mode of operation of a level 3 system can be either fixed block, where line capacity is not critical or moving block for minimised train headway.
To facilitate the migration to ERTMS/ETCS whilst the required maintaining investment at an affordable level, some railways, most notably the Swiss, advocate a Limited Supervision level of application to be added to the standard which would require only a limited number of signals to be equipped and the UIC still pursues a regional ERTMS project.
As the existing national systems in Europe cannot be replaced by ERTMS/ETCS within a short time, migration strategies were elaborated. One strategy is to dual-equip lines already carrying national systems with the European system as well, which allows trains with either type of on-board equipment to run on the line. The alternative strategy then is to first dual equip trains and then convert the network. Most likely a compromise between both, taking into account fleet circulation patterns and age distribution of trains and lineside equipment will be chosen.
Meanwhile High-Speed trains still need to travel over parts of the conventional networks to gain access to and from their terminal stations, when diverted etc. They must therefore maintain a backwards compatibility, to be able to operate with the conventional systems, identified as Class B systems in the TSIs, for the foreseeable future. This can be achieved through installation of Specific Transmission Modules (STMs) which interface to the European Vital Computer implementing the ERTMS/ETCS on-board equipment. These specific transmission modules (STM) on-board to translate the national ATP- “language” into the ERTMS/ETCS- “language”, so that the information can be processed by the ERTMS/ETCS on-board equipment. As there are 14 national ATP/ATC systems in use in the EU, theoretically this number of STMs would have to be installed on a train running through all countries in addition to the ETCS equipment itself. It is up to the individual railways and train operators, which mixture of both strategies described they choose to apply for their network or trains. In practice it is likely that STMs will be stripped versions of existing platforms as the market for them is likely to be small.