Method for data transmission via data networks

ES 2 348 618 T3

-1 METHOD FOR DATA TRANSMISSION THROUGH DATA NETWORKS

DESCRIPTION

The present invention relates to real-time communication systems and, in particular, to data networks based on Ethernet. In known state-of-the-art real-time communication systems, there is a problem that standard Ethernet participants cannot be directly integrated into the communication system or, for example, standard Ethernet participants cannot be coupled to a real-time communication system. The present invention is particularly described in the field of Ethernet-based communication systems, for which a simple implementation of a switching unit is presented.

A real-time communications system is essentially made up of a plurality of participants. In a special embodiment that is typically used in automation applications, the system has main participants (so-called masters) and secondary participants (so-called slaves).

Within this real-time communications system, data is exchanged in the form of messages in real time. In these real-time messages, your exact position is critical to the correct gathering of information. In addition to these messages in real time, there are also data exchange mechanisms that do not take place in real time. These mechanisms are used, for example, for commissioning and visualization or for diagnostic purposes with reduced time requirements.

By real-time communications is meant communications in which predetermined activities are executed with approximately no time delay, that is, generally with a maximum guaranteed delay previously known from other systems.

In this class of communication systems, there is often a need to allow access to each participant of the communication system. For this purpose, they are used in common computers that do not have a

ES 2 348 618 T3

-2direct coupling to the communication system in real time. These computers must be coupled to the system by switching units. In this way, the connection of computers to the communication system is effected, for example, by network participants. It is also common to connect computers to the Ethernet (Office). The coupling function is also received by each real-time communications participant.

Patent US 2002/0064157 A1 evidences a method according to the general concept of claim 1.

The object of the present invention is to make available a method for data transmission that allows the coupling of systems or devices that cannot be directly integrated into a real-time communication system. Said object is achieved by a method according to claim 1, a communication device according to claim 13 and a switching unit according to claim 17. The forms of execution and improvements that are beneficial are the subject of the claims below.

In the method, according to the present invention, for data transmission via data networks and particularly via Ethernet-based data networks with at least two participants, where the data networks have at least one topological domain in real time and one non-real-time topological domain, and data messages sent over data networks present at least a portion of real-time data and at least a portion of non-real-time data, data messages from the non-real-time domain, arriving within the real-time data portion, are temporarily delayed and transmitted to the non-real-time data portion by a switching unit according to the present invention.

The present invention relates in particular to data networks with repetition, at least partially. With this, data networks can be considered both exclusively with repetition, as well as with combined repetition and / or with switching. However, the repeat functionality is preferably mandatory.

A real-time topological domain is understood to be a domain that can be found within a data network. For a portion of real-time data and a portion of

ES 2 348 618 T3

-3 non-real-time data is understood, in particular, not only temporary portions. Each piece of real-time or non-real-time data can, however, also be defined by a position within a communication cycle.

By temporarily delaying data messages from the non-real-time domain and transmitting to the non-real-time data portion, conflict situations in critical real-time communications can be avoided. Furthermore, said portion of non-real-time data is particularly relevant in the real-time domain, that is, that domain to which so-called critical data is transmitted in real time.

Preferably they are also taken by removing the data messages from the non-real-time data portion (within the real-time topological domain) and transmitted to the non-real-time domain. This procedure represents the opposite direction of transmission with respect to the transmission of data messages mentioned above from the non-real-time domain to the real-time domain. Furthermore, in those data messages that are transmitted from the domain in non-real time to the domain in real time and those data messages that are transmitted from the domain in real time to the domain in non-real time, it is preferably treated of different messages of data demonstrating that data messages can be transmitted bi-directionally.

Preferably they are also taken by removing the data messages from the real-time data portion (within the real-time domain) and transmitted to the non-real-time domain. As a result, in this case too there is a transmission of data messages from the domain in real time to the domain in non-real time.

The data is preferably transmitted in the form of communication cycles, whereby said communication cycles have, particularly preferably, predetermined time intervals. Communications in real-time systems occurs here in communications cycles, where a communication cycle is essentially made up of two parts, namely the real-time communications part and the non-real-time communications part. The real-time communications part or the real-time data portion (also indicated below

ES 2 348 618 T3

-4as RT channel (in real time)) is usually managed by the main participant and the latter grants transmission authorizations. The data messages on said RT channel are preferably repeated by the participants.

Furthermore, there is a portion called in non-real time (hereinafter also indicated as IP channel) in which it is a time domain during which transmission authorizations are not predetermined. In this time domain, each participant can occupy the communication system. To avoid collisions on this IP channel, or to regulate it, it can be changed from the repeat function to a switched function for the IP channel time slot (switching of data messages on the IP channel). Within the communication cycles, both real-time temporal portions and non-real-time temporal portions appear.

The non-real-time data portion is preferably alterable set within one communication cycle.

The switching unit preferably determines parameters that are characteristic of the position of at least the non-real-time data portion within a communication cycle. Better to say, these parameters are learned. The parameters are preferably selected from a group of parameters comprising the start times of the non-real-time data portion, the end times of the non-real-time data portion, the time length of the non-real-time data portion non-real, the temporal length of the real-time data portion and the like. In particular, the values ​​for the start time of the non-real-time data portion and for the end time of the non-real-time data portion are required in order to determine the position of the data portion in not real time.

Furthermore, the temporal length of the non-real-time data portion is preferably determined. According to these parameters, that is, according to the reception time of the messages in non-real time, or of the messages in real time, the switching unit can conclude in the time domain of the IP channel, while said unit calculates the supported domain for the IP channel using the early or late reception time and the respective message lengths. Since the message identifier of the real-time messages

ES 2 348 618 T3

-5 (RT channel) differs from those present in the IP channel, the switching unit can differentiate the messages in the time domain of the IP channel from those present in the time domain of the RT channel. In other words, the position of the non-real-time data portion is determined by evaluating the reception times of the received messages.

No switching units are known from the state of the art that operate with a combined repeat and switch function. The switching units, according to the current state, must have parameters and, for example, must be parameterized by a main participant, in such a way that said parameters know the position of the IP channel. This means that the switching units known from the state of the art must respectively be independent network nodes that can be activated from the central station. That is, for example, a network address, special node management and node address setting are required. Thus, the method according to the present invention presents a drastic simplification of the known methods of the state of the art.

The position of the non-real-time data portion is preferably also determined by evaluating the reception times of the received data messages and, especially preferably, of the non-real-time messages.

The present invention is further directed to the use of a method of the kind described above for Ethernet-based communication systems. Furthermore, the present invention is directed to a use of the method described above for real-time communication systems in general. In particular, the present invention is intended to use a method of the kind described above for at least one repeating network, that is, a network in which it is not only switched (combined), but at least also repeats . Thus, the invention is also applicable to exclusively repeating networks. The processes presented here are developed, in particular, in the lower stages of the OSI reference model.

The present invention is further intended for a communications device for data transmission, where the communications device has at least one first participant and at least one second

ES 2 348 618 T3

-6participant, as well as a switching unit for the connection of other participants to the communications device. Furthermore, the communication device has at least one real-time topological domain and one non-real-time topological domain, and the data messages transmitted by the communication device have at least, in particular, a real-time temporal portion and, in particular, a temporary portion in non-real time.

In accordance with the present invention, the switching unit has a controller that achieves that data messages from the non-real-time domain arriving within the real-time data portion in the switching unit are temporarily delayed and transmitted. to the non-real-time data portion of a communication cycle. In this way, an interfering collision with critical data can be avoided in real time.

The controller preferably achieves that data messages are taken by removing them from a portion of data in non-real time and transmitted to the domain in non-real time. This means that data messages are also transmitted here from the real-time domain to the non-real-time domain. In addition, the controller also achieves that data messages are taken by removing them from a portion of data in real time and transmitted to the domain in non- real time . The concept of data messages, within the framework of the present invention, can be described both as individual data messages, as well as a plurality of data messages respectively.

The switching unit in particular independently determines parameters that are characteristic of the position of the non-real-time data portion within a communication cycle.

The present invention is further intended for a switching unit for a communications device for data transmission, in particular, via data networks based on Ethernet with a plurality of participants presenting at least one domain in real time and at the same time. minus one domain in non-real time. Thus, the data messages transmitted by the communication device comprise at least a real-time data portion and a non-real-time data portion.

ES 2 348 618 T3

-7 According to the invention, the switching unit has a controller that achieves that the data messages from the non-real-time domain, arriving within the real-time time portion in the switching unit, are temporarily delayed and incorporate into the non-real-time data portion.

The present invention is further intended to use a switching unit described above for a network at least also with repetition.

Other beneficial embodiments are apparent from the accompanying drawings. They show:

Fig. 1 a block diagram for exemplifying the structure of a real-time communication system;

FIG. 2 the block diagram of FIG. 1 for the exemplification of data streams;

Fig. 3 couplings, known from the state of the art, of participants of

Standard Ethernet;

Fig. 4 a representation of a communication cycle;

FIG. 5 a representation for the exemplification of the time parameters of a communication cycle;

Fig. 6 an illustration of a communication structure with a router participant;

Fig. 7 a representation of a data field;

Fig. 8 a detailed representation of the data field of fig. 7;

Fig. 9 an example for a field of a command instruction (data word field for control).

Figure 1 shows a block diagram of the structure of a real-time communication system 1, as it comes into play especially in automation applications. This system has a main participant 3 and a plurality of secondary participants 4, 4a, 4b, 4c. Between said main participant and the secondary participants, the data is exchanged in the form of messages in real time. With this, the reference sign 6 refers to a real-time domain of the communication system. The main participant presents two lines P1 and P2 for the corresponding data lines. In the case of the secondary participant 4b, in this embodiment, it is an interface of

ES 2 348 618 T3

-8E / S. Reference sign 21 refers to a non-real-time (topological) domain coupling, for example in the form of Ethernet (Office), and reference sign 5 refers to a data line.

Figure 2 shows the data flows between each of the participants. Furthermore, between the main participant 3 and the secondary participants 4, 4a, 4b and 4c, critical data is exchanged in real time, which is exemplified by the real time channel 13 (RT channel). In addition, critical data is also exchanged in non-real time via an IP channel 14 between each of the participants. This is the case, for example, when it is desired to access each participant of the communication system, for example, by computers.

In the embodiment shown in figure 3, a situation is evident in which computers 27 and 28 are used that do not have any direct connection to the real-time communication system. These computers must be coupled to the system by switching units. Furthermore, it is known from the state of the art that non-real-time participants, indicated with 27 and 28, are coupled to the communication system through network participants. It is also common to connect a computer to the Ethernet (Office). In addition, each real-time communications participant receives the coupling function. In the representation shown in figure 3, the computers are coupled via input P1 or the last secondary participant 4c. For coupling, it is necessary in these systems that each participant knows the time division of the corresponding communication cycles or the positions of the IP channel and the RT.

The last participant 4c shown in fig. 3, use its internal bridge function. With this, the RT channel is prevented from being interfered with by IP data of the messages. The main participant 3 supports the transmission of IP data via the P1 input.

Figure 4 shows a representation of a communication cycle 16. In the preceding embodiment, the communication cycle 16 presents a real-time data portion 13 and a non-real-time data portion 14, where within these portions data messages are arranged in each case 11. In the

ES 2 348 618 T3

In the lower representation, the communication cycle 16 presents a non-real-time data portion that is arranged between two real-time portions 13.

Within the real-time data portion, there are data fields (HDR / MST) for synchronization and management.

As shown in Figures 4 and 5, the position of the IP channel or non-real time data portion within communication cycle 16 is defined by a predetermined constant time domain. This time domain is typically predetermined by the primary participant. Furthermore, the position of this IP channel can be transmitted, for example, within a startup phase, from the main participant to the secondary participant.

In the state of the art, the position of the main participant's IP channel is explicitly transmitted to the secondary participants as so-called parameters. Therefore, the secondary participants must possess, as mentioned above, a parameter storage medium that can be accessed, which can be defined by the central station. In contrast to this, the times t6 and t7 in Figure 5, according to the present invention, are those parameters that describe the position of the IP channel within the communication cycle, since time t6 represents the start time of the portion non-real-time data 14 and time t7 represents the completion time of the non-real-time data portion 14. Time tscyc represents the total communication cycle time 16.

To avoid interference in real-time communications, data messages from the non-real-time domain, indicated in Figure 3 with the reference sign 7, are coupled by the switching units to the real-time domain, in such a way that real-time communications are not interfered with. This means that the data must be entered on the IP channel. Furthermore, data messages from the IP channel and optionally also from the RT channel can also be entered into the domain in non-real time 7, as mentioned above.

FIG. 6 shows a communication device according to the invention with a switching unit 10. In contrast to the known switching units of the state of the art, a separate switching unit is described according to the invention (also indicated as IP T connector) that up to

ES 2 348 618 T3

The time was not known for the communication systems present in this invention, which comprise a combination of repeater function and switched bus communications.

In contrast to the switching units known from the state of the art, the switching unit according to the present invention does not have network participants with their own network addresses. Consequently, said unit is also not visible as a node in the communications network, but only transparently routes the domain data in non-real time (represented in figure 6 as Ethernet switch 4a with three participants 8, 9 and 12) to the IP channel in the domain in real time. This switching unit 10, moreover, does not include parameters to be configured, but independently determines the necessary parameters, which are treated at least of the parameters t6 and t7, as indicated above. Beneficially, the value for the time interval tscyc is also determined. Thus, the switching unit 10, according to the present invention, is a passive element with respect to real-time communications. Said switching unit processes only the IP channel, therefore, as mentioned above, the times t6 and t7 must be evaluated. In addition, the MDT 0 message must also be evaluated, which is represented in figures 4 and 5. An IP message should be entered only in one channel (P or S), due to the time offset (call delay) of the P channel. and S, to be able to carry out the complete arrangement in a simpler way.

Figures 7 and 8 describe a self-adaptive switching unit that uses the evaluation / analysis / interpretation of the content of the messages. If the information is transmitted in data messages within a cyclic channel through the position of the IP channel to a position known to the switching unit, the switching unit can determine the temporal position of the IP channel by said evaluation of the content of the IP channel. the messages.

As shown in figure 7, it could be used for this purpose, for example, the so-called hot-plug field that exists in SERCOS III (Serial Communications System in Real Time). In SERCOS III, in addition to data divisions such as the source address, the destination address, the initial synchronizer (preamble), the

ES 2 348 618 T3

- 11 frame check sequence data, also the MDT data field. This MDT data field consists of a MDT hot plug data field, which is generally used to admit additional new participants, the MDT Service Channel Field of devices Service Channel Field, and the MDT Real Time Data Field of devices. devices.

As FIG. 8 shows, the so-called MDT hot-plug field is further composed of subfields intended for the device address, a data word for control and an information field.

In a preferred embodiment, it is recommended that the last hot-plug information field be populated with times t6 and t7. This and other data are transmitted to the field preferably multiplexed, as long as a new participant does not have to be admitted. In this way, a switching unit can determine the times during this time by evaluating the entire data message.

Fig. 8 shows for multiplexing a possible synchronization for the aforementioned data word for hot plug control. Furthermore, both encodings for t6 and t7, which are evaluated by the switching unit, are of particular importance, where in this example only the data in the case MHP-ADR = 0xFF is relevant, which means that neither should be supported. new participant. Furthermore, the values ​​for tscyc and t1 can also be transmitted.

In this embodiment, information is preferably multiplexed within a predetermined data field.

All the features evidenced in the present application are claimed as essential of the present invention, insofar as they are novel individually or in combination compared to the art technique.

List of main participant real-time communication system reference symbols

4, 4a, 4b, 4c secondary participants data line

ES 2 348 618 T3 real-time domain non-real-time domain

8, 9, 12 participants switching unit data messages real-time data portion non-real-time data portion communication cycle non-real-time domain coupling

27, 28 computers tscyc total communication cycle time t6 non-real-time data portion start time t7 non-real-time data portion end time