DE102006036562B4 - Method and system for transmitting data streams related to one another and / or for synchronizing data streams related to one another - Google Patents

Abstract

A method of receiving, with a first data stream having at least two successive subsections and with a second data stream, comprising the steps of: - receiving the first data stream (E); - Receiving the second data stream (P) with a useful portion (W) and a first identification feature (Q, U) which is derived from a first section of the at least two successive subsections of the first data stream (E) and can be used to identify the first section is; - Delaying at least one of the first and second data streams; - Determining the first identification feature (Q, U) in the second data stream (P); - Determining the first subsection in the first data stream (E) as a function of the first identification feature (Q, U); - Outputting the useful portion (W, Y) of the second data stream (P) temporally using the first identification feature, coordinated with an output of the first subsection.

Description

The invention relates to a method for receiving a first data stream and a second data stream. The invention also relates to a method for transmitting and to a method for transmitting a first and second data stream from a transmitter to a receiver. The invention further relates to a transmitting arrangement, a receiving arrangement and a system comprising a transmitting and a receiving arrangement.

The invention relates generally to a method and a system for transmitting data streams related to one another, and to a method for synchronizing data streams related to one another, as well as during the implementation of the method or in the construction of usable system components. With regard to the synchronization of data streams related to each other, it is primarily thought that they are data streams that are to be synchronized in a communication environment, possibly via different transmission paths or in different ways. This is not mandatory. It may also be other data streams that occur in any technical environment or may even be created for the purpose of synchronizing related information or data.

In the following, the invention is explained on the basis of data streams occurring in a communication environment, since this is indeed a particularly important application of the invention proposals.

Modern communication links, for example for the transmission of voice and audio signals, often use protocols in networks, which connect the respective transmitting or receiving station via different network nodes. Such a connection is also referred to as a circuit switched connection. Data or signal transmission takes place via the network nodes, it being possible for different types of data transmission to be provided, depending on the protocol and network used. In a circuit-switched connection, the transmission path is generally predetermined and kept clear for all signals and signal components. As a result, a substantially constant delay time between the transmitter and the receiver occurs with such a communication connection. The duration of the delay is determined inter alia by the switching times of the different components of the various nodes over which the communication link is established.

The transmission formats are often designed such that useful signals without time information contained therein are transported over the medium. A typical example is an ordinary, z. B. analog, telephone connection between a transmitter and a receiver, are received in the voice or audio signals transmitter side of a microphone and sent via the communication link to the receiver. Once available at the receiver, they will be output through a speaker without additional delay. In digital communication links, the audio signals are often divided into sections of certain lengths, coded, and then sent to the receiver. Even digitally encoded audio signals usually contain no additional time information. In general, audio signals in an end-to-end communication link using circuit switched technology can be expected to have no additional time information.

In addition, it is possible to transmit user data of different types or media, for example voice and image, in a synchronized manner between sender and receiver. Such synchronized transmission of a combination of two or more media signals or data streams is also referred to as English. "Session". A typical example of this is the so-called video conference, in which in addition to a pure audio signal or audio data stream and video or video signals are transmitted. There are various possibilities for the synchronization between audio and video signals. In a circuit-switched approach, audio and video signals are combined into a circuit-switched data stream, typically referred to as "multiplexing", so that one and the same transmission path is used for both streams or signals and thus the synchronization is conveniently provided by appropriate arrangement in the data stream, by all signals are played immediately after their decoding. In another, z. B. packet-switched approach different media or every single packet of each medium can be transmitted in principle via different transmission paths. In this case, for correct playout and synchronization, additional information in the form of time stamps, eg. B. in each media package provided.

However, there may be a need to add additional second user data to an already existing communication connection, wherein the existing communication connection thereby is continued. This is the case, for example, if in addition to video or audio signal transmission video signals are to be transmitted. The question of a common output arises, inter alia, when the initially mentioned "circuit switched" technique is used for the first connection.

A simple possibility is to transmit the second media signal, in this case the video signal, via an additional communication link. In such a case, it is quite possible that the second communication connection uses other areas of the communication network. If both media signals or user data are output on the receiver side, the established session is not synchronized and there is generally a time offset between the different media signals or data streams.

One way to prevent mismatched output would be to terminate the first communication connection and rebuild a completely new "session" including the additional second media signal. Synchronization can then take place by the simultaneous transmission of the two media signals or data streams over the same connection. Alternatively, a further media signal, for example a video conference, can be provided for the connection. In this solution, however, the already existing communication connection is terminated, an operation that is not desired by a user as a rule. In addition, even if one of the two data streams is changed, the existing connection must be terminated. A flexible addition of an additional media signal or data stream with, for example, different bit rate or content to an already existing communication connection is not possible in this embodiment.

The WO 2004/052020 A1 discloses a synchronization method for two signals in which both signals are initially synchronized on the source side and each provided with a time stamp, the signals being synchronized again on the receiving side on the basis of the time stamp.

The WO 2005/011281 A1 discloses a method for the synchronization of data streams, in which a fingerprint is derived on the source side from both data streams, ie a feature by means of which a segment of the corresponding data stream can be identified. The fingerprints are transmitted independently of the data streams and separately from them to the receiver, which identifies the respective segments of the data streams on the basis of the fingerprints and uses these to synchronize the data streams.

The WO 02/51155 A1 discloses a synchronization method that uses timestamps inserted into data streams.

The GB 2366 110 A discloses a method for eliminating synchronization errors based on auditory and visual speech recognition techniques.

The object of the invention is to optimize a coordinated temporal output of first and second data streams and in particular to be able to dispense with the use of known from the prior art time stamp.

To solve these objects, the objects are provided according to the independent claims. Embodiments and further aspects of the invention are the subject of the dependent claims.

According to the principle of the invention, it is provided to provide additional identification features in the second data stream which are derived from the first data stream and thus allow a coordinated output of the first and second data streams. As a result, an already existing communication connection with a transmission of the first data stream can be continued.

The term data stream means any type of signal that is suitable for the transmission of information. The data stream can therefore contain a useful signal, but also represent the useful signal itself. Likewise, it may be an analog or a digital signal or include such.

According to the proposed principle, a second communication connection for the transmission of the second data stream is established for the first communication connection for a transmission of the first data stream. The second data stream contains identification features which enable an assignment of sections of the second data stream to corresponding sections of the first data stream. Likewise, identification features can be contained in the second data stream, which can be assigned allow sections of a useful signal contained in the second data stream to corresponding sections of the first data stream or to corresponding sections of a useful signal in the first data stream. On the receiver side, a coordinated output of the first data stream and of the second data stream can thus take place.

Expediently, the identification features in the second data stream are derived from the first data stream or from payload data in the first data stream. It is even possible that the identification features comprise essential parts of the first data stream or the payload of the first data stream.

For example, the first data stream may contain audio signals or audio data in at least two of temporally successive subsections. The term audio signal or audio data is understood to mean any form of audible signals, ie speech, sounds, or even music. An identification feature is extracted from each of the subsections, which enables identification of the respective subsection of the first data stream. This identification feature is assigned to a subsection of a second data stream so that a later coordinated or even synchronous output of the respective subsections is possible. A synchronous output is understood to mean a balanced playing, in order to have the best possible synchronicity under other conditions. It is expedient if the features used for the identification of the subsections are particularly robust against artifacts or disturbances during the data transmission of the first data stream. For example, perturbations or signal variations may arise from the encoding and compression of signals due to irrelevance reduction and quantization or from distortion or noise during transmission or any other effects. In this way, a synchronous output of the first and second data stream can take place on the receiver side even after one or more of these disturbances during the data transmission.

In one aspect of the invention, the method comprises the steps of receiving and delaying the first data stream having a plurality, typically a plurality of successive subsections. Likewise, a second data stream is received. From this, at least one first identification feature is determined, which is derived from at least a first subsection of the successive subsections of the first data stream. In addition, the method comprises the step of analyzing the first delayed data stream or the individual subsections in the first data stream. A first subsection of the first data stream is determined as a function of the at least one first identification feature, and subsequently this first subsection is output together with the second data stream.

For delaying the first data stream, it is sufficient if this is temporarily stored. Measures can also be provided for the processing of the first data stream, which temporally delay an output of the subsections of the first data stream, so that sufficient time remains for the further method steps. A time offset between corresponding portions of the first and second data streams can be corrected.

Alternatively, the second data stream can also be delayed and, if necessary, buffered for this purpose. In that regard, the second data stream could then be considered as above regarding the first data stream and the first data stream could be considered as previously discussed with respect to the second data stream. Both data streams could also be delayed and buffered.

For a coordinated output of the first and second data stream, it is therefore expedient, in addition to the extraction of the first identification feature in the second data stream, to determine a corresponding identification feature in the first data stream. For this purpose, the subsections of the first data stream are analyzed and an identification feature is derived from each subsection. This should be of the same nature as the first identification feature, so that after a comparison, a decision can be made as to whether the first identification feature contained in the second data stream matches or is at least sufficiently similar to the derived identification features of the first data stream.

In the present invention, an additional second data stream can thus be added to an already existing communication connection with a data transmission without having to interrupt the already existing communication connection. It is only necessary to buffer at least one of the data streams, in particular the first data stream, before output in order to enable the steps of determining identification features and subsequently synchronizing them.

As has already been explained above, it is primarily thought that the first data stream and the second data stream each have a plurality or multiplicity of successive subsections, possibly data frames, with several or all subsections of the second data stream pointing to a respective assigned subsection associated with the first data stream, contained in the second data stream identification feature is assigned and for several of these identification features of the associated subsection of the first data stream is determined and over the respective identification feature mutually associated portions of the first and the second data stream are output timed. In this context, it may be expedient if, for the assignment of a subsection of the first data stream to a subsection of the second data stream, a plurality of identification features that refer to different subsections of the first data stream are taken into account.

In one embodiment of the invention, a first useful signal is provided for generating the first data stream, which comprises a plurality of successive subsections. The first payload signal may be any type of signal that includes audio data. From the individual subsections of the useful signal identification features are derived, which identify the respective subsection with a high probability. As a result, an assignment to the respective subsection can be determined with a high degree of probability on the basis of the identification feature. The derived identification features may be used to generate a second data stream by associating subsections in the second data stream intended for concurrent output with the corresponding subsections of the first data stream.

Conveniently, the derived identification features are particularly robust against artifacts in signal processing, compression or interference during data transmission. For example, the identification features can be formed from LTP (Long Term Prediction) coefficients of a subsection. Alternatively, the identification feature can be generated by analyzing frequency information of a content of the respective subsection of the first data stream or of the first useful signal. It is also possible to determine various sequence information, for example HASH sequences or checksums from subsections. The latter is useful if the first useful signal contains coded information according to a coding method. But it is also possible that the identification features encoded portions of the first data stream or the first useful signal.

In another embodiment of the invention, the method additionally comprises the steps of providing a first and second useful signal, which each have a plurality of temporally successive subsections. Corresponding subsections are assigned to each other for a later coordinated output. Identification features are determined from subsections of the first useful signal and assigned to the corresponding subsections of the second user signal. Subsequently, first and second useful signals are further processed and transmitted as the first or second data stream. After a reception, the two data streams are buffered. The identification features present in the second data stream are extracted. At the same time, the subsections of the first useful signal are obtained from the first data stream and identification features are determined therefrom. These are compared with the identification features present in the second data stream. If there is a match, the corresponding sections are output so that a coordinated output is ensured.

It is additionally possible to discard or issue unused sections after a certain period of time. This can be useful, in particular, if parts have been lost during the data transmission of the first or second data stream or can not be reconstructed due to errors during the data transmission.

• – Analoge Audiosignale bzw. Audiodaten;
• – GSM kodierte Audiosignale;
• – G.711 kodierte Audiosignale;
• – PCM kodierte Audiosignale;
• – AMR oder AMR-WB kodierte Audiosignale;
• – nach einem der G.72x Verfahren des ITU-T Standards kodierte Audiosignale, z. B. G.723.1, G.727, G.729.

The first data stream may contain at least one of the following useful signals or be formed as such:

• - analog audio signals or audio data;
• - GSM encoded audio signals;
• G.711 encoded audio signals;
• - PCM encoded audio signals;
• AMR or AMR-WB encoded audio signals;
• - According to one of the G.72x method of the ITU-T standard encoded audio signals, eg. BG723.1, G.727, G.729.

Identification features for subsections can be determined from each of the mentioned audio signals, which with high probability enable identification of the respective subsection. It is possibly useful to use identification features that are generated anyway during a signal processing or coding of the audio signal or the first useful signal. This reduces the complexity within a transmission arrangement. By using existing coding or communication standards, the existing technologies and transmission methods can continue to be used.

With the method according to the invention, it is possible to add any additional data stream to the already existing data stream without interrupting it. For example, the additional data stream may contain video data, in particular in MPG, MPG2 or H.26X format. Likewise, additional image data, in particular in the JPG format or even data for a computer-assisted slide show can be transmitted. Another possibility is the parallel transmission of further voice or audio signals or audio data, for example for the transmission of a stereo signal. Likewise, text can be transmitted as a useful signal in the second data stream tuned to the first data stream. In principle, any type of information that requires synchronization to the first data stream or to the user data present in the first data stream is suitable.

A transmission of the first or the second data stream can be done in part by transmitting signals according to a mobile communication standard or for example by an Internet connection between at least two different nodes. It is likewise possible to carry out a transmission at least partially through a wired communication connection.

Accordingly, a transmission arrangement according to the proposed principle comprises a first device for detecting a first useful signal and a second device for detecting a second useful signal. Furthermore, an analysis device is provided, which is designed to determine at least one identification feature for an identification of the first or of the respective subsection from a first subsection or subsections of the first useful signal. The analysis device is connected to a coding device which assigns the identification feature provided by the analysis device or a respective identification feature to a first subsection or to a respective subsection of the second useful signal. The transmission arrangement also comprises a first transmission stage, which is coupled to the first device for transmission of the first useful signal. A second transmission stage for a transmission of the second useful signal is connected to the coding device.

With the transmission arrangement according to the invention, a first and a second useful signal can be detected. While the first useful signal is transmitted by the first transmission stage, the transmission arrangement comprises different means for further processing of the second useful signal. It is provided to assign the second useful signal identification features that allow synchronization with the first useful signal. For this purpose, the transmission arrangement according to the invention contains an analysis device for an analysis of subsections of the first useful signal.

The analysis device can be designed to determine from these signals identification features that are both robust against interference of any kind in the voice or audio signal and on the other hand allow identification of the respective subsection of the signal. In addition to the use of purely analog signals, the identification features can also be determined from coded digital useful signals. Likewise, the identification features may include portions of the audio signal or encoded audio data.

In one embodiment of the invention, therefore, the first device for detecting the first useful signal for coding according to a Kodierungsart is executed. For example, this is a μ-law coding, a GSM coding, a PCM coding, an AMR or AMR-WB coding a G.711 coding or a G.72x coding, whereby x can assume different values. The latter encodings refer to various methods of the ITU-T standard for voice compression. The various coding methods are particularly suitable for the coding of speech or audio signals.

For a coordinated output of the first and second useful signals, it may be expedient if the coding device additionally comprises a synchronization device. This is suitable for assigning an identification feature of a subsection of the first useful signal to a subsection of the second wanted signal, this assignment corresponding to a predetermined time ratio of the subsection of the second wanted signal to the subsection of the first useful signal. In addition, the coding device can be designed for the generation of so-called time stamps. Thus, the identification features and the subsections can be applied in the second useful signal, whereby an assignment is determined.

In another embodiment of the invention, the first transmission stage is designed for transmission on a first transmission channel and the second transmission stage for transmission onto a second transmission channel. These can each represent different transmission channels but also the same transmission channel. Likewise, it is possible for the first and the second transmission stage to form a common transmission stage and to transmit first and second user data in a common transmission channel.

Another aspect of the invention relates to a receiving arrangement for a coordinated playback of payloads of a first and second data stream. This may include a first buffer for storing a first useful signal or a delay unit. An analysis device in the second receiving arrangement is coupled to the first intermediate memory or the delay unit and designed to determine and provide at least one identification feature for the identification of at least one first subsection of the first useful signal from the first useful signal.

The receiving arrangement contains a synchronizer for a concerted output of the first and second useful signal based on a comparison of stored in the second payload identification features with the at least one provided by the analysis device identification feature.

The processing of the first data stream and the extraction of identification features eliminates the need to terminate an already existing communication connection and transmission of the first useful signal. Rather, it is possible to set up a new communication connection for a transmission of a second data stream, to receive this and to determine a second useful signal and the identification features, which are provided for a synchronization of the second useful signal with the first useful signal. As a result, existing communication standards or technologies can continue to be used.

In one aspect of the invention, the receiving arrangement includes a comparison device configured to output a control signal in response to a comparison of the identification features stored in the second data stream with the identification features provided by the analyzer. This control signal is used to tune the output of the first and second useful signal.

The comparison device is embodied in one aspect of the invention for analyzing a similarity between the deposited identification features and the at least one provided identification feature. By analyzing a similarity, it becomes possible to enable identification and assignment for a synchronous output even during a slight disturbance of the first or second useful signal during a data transmission.

In another aspect of the invention, the comparison device is designed with a generation of the first useful signal from the identification features. This is useful, for example, if the identification features in the second data stream represent an encoded form of the first useful signal. In this embodiment, it is possible to make a decision as to which first useful signal is to be output, which is determined from the first data stream or generated from the identification features contained in the second data stream.

In general, because the transmitted signals are encoded to the receiving device in accordance with a communications standard, in one aspect of the invention, the receiving device comprises a first encoder. The coding device is designed to generate the first useful signal from a first signal received by the receiving device. Furthermore, a second decoding device may be provided which is designed to generate the second useful signal from a second received signal. It is possible that the receiving arrangement is designed for receiving the first and the second signal in a common transmission channel or in different transmission channels.

In the system according to the invention is a transmitting device, which is designed for the delivery of a first and a second signal, wherein the second signal comprises identification features, which are derived from the first signal or from information contained therein. The receiving arrangement is designed to determine identification features from the first signal and includes means for a coordinated, in particular to a synchronous output derived from the first and second signals. The output is dependent on a comparison of the determined identification features with the identification feature contained in the second signal. Thereby, the output of any content of a second signal can be timed to an output of the first signal. The second signal may include, for example, audio signals, video signals or computer-aided slide shows.

According to a second main aspect, the invention provides a method for synchronizing a plurality of data streams, each having a plurality of successive subsections, possibly data frames, wherein at least a second of the data streams in association with subsections of the second data stream refer to subsections of at least a first of the data streams Contains identification features and the identification features associated with each other sections of the data streams are determined to synchronize the data streams. The data streams to be synchronized may be transmitted data streams, possibly transmitted via different communication channels. This is not mandatory.

In a further development, it is proposed that a plurality of identification features that refer to different subsections of the first data stream be taken into account for the assignment of a respective subsection of the first data stream to a respective subsection of the second data stream. In this way, the synchronization is particularly good and possible with a certain error tolerance, which is particularly useful in the case of possibly disturbed communication channels. Disturbances of a respective communication channel could lead to a falsification of the identification features or the first data stream contained in the second data stream, which is less problematic if synchronization takes account of identification identifiers referring to different subsections of the first data stream.

In order to enable unambiguous identification of subsections of the data streams allocated to one another, it is expedient for a respective subsection of the first data stream to contain a set of several identification features that refer to this subsection in the second data stream and to be taken into account in the determination of mutually allocated subsections of the data streams.

The first data stream may include the identification features corresponding or complementary to the identification features transmitted in the second data stream. This is not mandatory. It could also be necessary to determine appropriate identification features, for example, to calculate. In general, it is proposed that in each case at least one identification feature, possibly a set of identification features, is determined for subsections of the first data stream and compared with identification features or identification feature sets containing the second data stream. The identification of identification features also includes, in particular, that the identification features are still being determined, for example, calculated from data contained in the first data stream or otherwise determined, as already mentioned. It is further included that the identification features represent partial data of the first data stream.

In a further development, the comparison is made by applying at least one comparison criterion, possibly a distance criterion, preferably by determining at least one comparison measure from the comparison criterion or at least one distance measure from the distance criterion in an abstract feature space containing the identification features.

According to an expedient embodiment, it is provided that this by multiple application of the comparison criterion between on the one hand for a Aufpunkt portion of the first data stream determined identification feature or identifier feature set and on the other hand, to different sections of the first data stream, with the second data stream transmitted identification features or Identification feature sets on the one hand on the / the identification feature or identification feature set of the Aufpunkt subsection and on the other hand on a / a specific / determined, contained in the second data / contained identification feature or Identification feature set related comparison measures are determined, which are each associated with the Aufpunkt subsection and a particular subsection of the second data stream and / or a subsection distance between the Aufpunkt subsection and the particular subsection of the second data stream.

Alternatively and additionally, it can advantageously be provided that the identification features or identification feature sets determined by multiple application of the comparison criterion between, on the one hand, an assignment point to a sub-point subsection of the second data stream contained in this identification feature or identifier feature set and on the other hand, for different sections of the first data stream several on the one hand to the / the Aufpunkt subsection associated (n) identification feature or identification feature set and on the other hand on a / for a particular subsection of the first data flow determined / determined identification feature or identification feature set related comparison measures are determined, each of the Aufpunkt subsection and a certain subsection of the first data stream are assigned and / or a subsection distance between the Aufpunkt- section and the determined Part of the first data stream.

In a further development, it is proposed that, on the basis of the determined comparison measures, if appropriate by determining a minimum or maximum comparison measure, mutually allocated sections of the first and the second data flow are determined. It is specifically thought that on the basis of the determined comparison measures, if necessary by determining a minimum or maximum comparison measure, a synchronization measure, possibly a subsection distance corresponding shift measure is determined, which at least a subsection or subsections of the first data stream at least assigns a subsection or subsections of the second data stream in pairs.

According to a particularly preferred embodiment of the method according to the invention, it is provided that a plurality of sets of comparative measures are determined in each case for a plurality of different sub-point subsections of the first or second data stream by repeated application of the comparison criterion. In this context, it is further proposed to be particularly preferred because it is particularly fault tolerant that it is proposed that comparative dimensions of the sets of comparative dimensions assigned to a same partial section distance be combined with one another for a respective combination comparison measure. Preferably, then, the determination of subsections of the first and the second data stream allocated to one another takes place on the basis of the determined combination comparison measures, possibly by determining a minimum or maximum combination comparison measure.

Often it will be useful if, based on the determined combination comparison measures, if necessary, by determining a minimum or maximum combination-Vergleichmaßes, a synchronization measure is determined, for example, a subsection distance corresponding shift measure. The synchronization measure or the shift measure assigns at least one subsection or subsections of the first data stream to at least one subsection or subsections of the second data stream in pairs.

Advantageously, the respective combination comparison measure of an optionally scaled sum or an optionally scaled product or another mathematical combination corresponding to the same subsection distance associated comparison measures that increases at least monotonically with a respective incoming comparative measure or at least monotonically decreases.

A particularly expedient embodiment of the method provides for the combination comparison measures to be iteratively calculated from an initial set of combination comparison measures or a set of comparison measures by successively taking into account further subsections of the first data stream and / or the second data stream or be determined by taking into account further sets of comparative measures. Furthermore, it can be provided that a respective set of comparative measures is supplemented by further comparison measures by taking into account further subsections of the first or second data stream.

It is especially (but not exclusively) thought of synchronizing media streams. For example, at least one first data stream comprising an audio data stream or voice data stream can be synchronized with at least one second data stream comprising at least one further audio data stream or voice data stream and / or comprising at least one video data stream.

According to a preferred embodiment, which is particularly relevant for practice, it is provided that the first data stream is a codec-coded data stream, wherein parameters or data output by a coding device based on a codec are recorded in the first data stream in order to be reconstructed from the first data stream underlying audio or voice information by decoding the first data stream, and at least partially included in the second data stream to enable synchronization of the first with the second data stream. For example, in this context, it is contemplated that the codec is a codec based on prediction analysis, preferably linear prediction analysis, of the underlying audio or speech information, in particular an LP or Linear Predictive Coding Codec or a CELP or Code Excited Linear Predictive Coding codec. Specifically, the codec may be a voice codec and based on a language model.

Also with regard to the second data stream, it is thought that this is a codec-coded data stream, wherein parameters or data output by a coding device based on a codec are recorded in the second data stream in order to reconstruct the audio or speech information underlying the second data stream To enable video information by decoding the second data stream.

One can directly or directly synchronize the first and the second data stream by performing the synchronization with respect to these data streams themselves. Alternatively, however, it is also possible for the first and the second data stream to be synchronized indirectly or indirectly by synchronizing successive data streams obtained by decoding or recoding the first and the second data stream.

In general, with regard to the method according to the invention and the various design possibilities mentioned, it is further proposed that the first data stream and the second data stream are generated by a first instance and transmitted to a second instance, and that the first and the second data stream are passed through the second instance or an intermediate instance mediating the transmission between the first and the second instance is synchronized, wherein in the case of the intermediate instance this combines the first and the second data stream, if desired, to form a combined data stream and transmits it to the second instance, preferably instead of the first and the second data stream.

In a communication environment, possibly mobile telecommunications environment, at least one first user terminal or the first instance and at least one second user terminal can form the or a second instance. If the synchronization is performed by an intermediate instance, then at least one infrastructure device of the communication environment can form the or an intermediate instance, in particular, for example, a device of the mobile telecommunications system comprising the base station, etc.

In the case of bidirectional or multidirectional communication, the first user terminal may also have a second instance with respect to at least one further first data stream and at least one further second data stream, and the second user terminal may refer to the further first data stream and the further second At the same time, the data stream also forms a first instance.

In a further development, it is proposed that the first and the second data stream or the further first and the further second data stream are transmitted via different transmission channels and / or over between the instances on the basis of various transmission standards. In this case, it may be expedient or occur in certain situations that one of the first and second data stream or one of the further first or second data stream via a circuit-switched or circuit-switched connection, preferably at least partially by radio according to a mobile telephony standard is transmitted and the other is transmitted from the first and second data streams and the further first or second data streams over a packet-switched connection, preferably according to a network or Internet transmission standard. In this connection, it is primarily thought that the first or the further first data stream is transmitted via the circuit or circuit-switched connection and the second or the further second data stream is transmitted via the packet-switched connection.

The synchronization of the data streams according to the invention is particularly expedient if one of the first and second data stream or one of the further first or second data stream is delayed with respect to the other from the first and second data stream or from the further first or second data stream in the transmission, if necessary with variable, time-varying delay time.

According to an embodiment of the method which is particularly important for practice, it is provided that first the generation and transmission of the first or further first data stream takes place and then later the generation and transmission of the second or further second data stream and then the synchronization of the data streams. Thus, for example, a normal telephone conversation can first be recorded and then transferred to a video telephony call or a telephone conference with several subscribers without the original telephone conversation being interrupted. It should not be ruled out that the original telephone conversation is continued only until a seamless and uninterrupted transition to a telephone conference or a video call is reached. If the second data stream also contains the voice information of the first data stream, after the synchronization has taken place, the communication can take place solely by means of the second data stream and the first data stream can then be interrupted.

The inventive method according to the second aspect may be carried out expedient in connection with the implementation of the method according to the first main aspect or as a partial method of the method according to the first main aspect. From the invention and further training proposals for the method according to the first main aspect, there are possible further process features of the method according to the second main aspect.

The invention further provides, according to the second main aspect, a system for carrying out the method according to the second main aspect, comprising at least a first system component configured to generate and transmit the first data stream and the second data stream to a second system component second system component or at least one intermediate component of the system which mediates the transmission between the first and the second system component is designed to synchronize the first and the second data stream. It is especially, but not exclusively, contemplated that the system is a communication system, possibly a mobile telecommunications system, or comprises a mobile telecommunication system, wherein at least a first user terminal the or a first system component and at least one second user Terminal which forms or a second system component. The first and second system components may represent the first and second instances of the method according to the invention.

If the synchronization is effected by an intermediate component of the system, then an infrastructure device of the communication system can form the or an intermediate component. The intermediate component can represent the intermediate instance of the method according to the invention.

It is preferably provided that the system is designed for bidirectional or multidirectional communication, the first user terminal also having a second system component and the second user terminal with respect to at least one further first data stream and at least one further second data stream the further first data stream and the further second data stream at the same time also forms a first system component.

The system may be distinguished by further device features which result from the explanations concerning the system and the devices according to the first main aspect of the invention.

Above all, a user terminal, possibly a mobile telephone, for a system according to the invention is also provided, wherein the user terminal is designed to generate the first and second data streams in carrying out the method according to the second main aspect of the invention, in particular on the basis input or acquired audio or voice information and video information, and to send the data streams, possibly via different communication channels or based on different transmission standards, or / and wherein the user terminal is designed for, in the implementation of the method according to the second main aspect of the invention to receive and synchronize the first and second data stream and the further first and further second data stream, in particular for a temporally synchronous output by means of the data streams of transmitted audio or voice information and video information.

The user terminal may comprise a transmission arrangement and / or a reception arrangement according to the first main aspect of the invention, as mentioned above.

The system according to the invention may comprise a plurality of user terminals as mentioned as the first system component and / or as the second system component.

In the following the invention will be explained with reference to various embodiments with the aid of the drawing in detail. Show it:

1 a first form of representation of a system with a transmitter and a receiver according to the inventive principle,

2 a second representation of a system for a synchronous transmission between a transmitter and a receiver according to the principle of the invention,

3 a third representation of a system for a synchronous transmission between a transmitter and a receiver according to the inventive principle,

4 a schematic representation with first and second useful signal,

5 a first embodiment of the method according to the invention,

6 a second embodiment of the method,

7 a third embodiment of the method.

8th serves with the schematic data flow representation of the subfigures 8a) and 8b) for explaining a further embodiment of the method according to the invention and of the system according to the invention and with the data flow representation of the subfigure 8c) to explain an embodiment of the method and the system.

9 serves with the subfigures 9a) and 9b) for explaining a preferred way of synchronizing two data streams, in the present case a video stream and an audio stream, in particular corresponding to a speech-coded audio stream and a video stream in principle 8th ,

As a rule, a system for communication comprises the following components. On the one hand, a transmitter is provided, which records the payload data to be transmitted and processes it according to a communication standard. The processed and possibly processed data are passed on to a communication network. As a rule, this contains several individual nodes which pass on the useful signal. Depending on the selected communication connection, it may happen that the useful signal is processed again or converted into another data format. At the end of the system, the network is in turn coupled to the receiving unit, which converts the transmitted data back into the payload.

A typical example of this is a connection of a mobile communication device in a fixed network and, for example, to an ISDN telephone line. In this case, the useful signals to be transmitted in the form of voice or audio signals are converted by the transmitter, the mobile communication device into signals according to a communication standard and transmitted to a base station. The base station forms part of the network and further processes the received signals. For example, it extracts the useful signal in the form of the speech signal. This is coded according to the PCM coding method and transmitted to the receiver via a landline line. It may well happen that depending on the distance between the transmitting device and receiving device within the network several different nodes are provided, which reprocess the signal during data transmission or otherwise process.

For the data transmission in the present case, an end-to-end connection is used. Here, the communication link between the communication device and the telephone is established, and the line remains independent of a voice transmission. The connection is interrupted only after a complete termination. This results in the transmission of the language to a substantially constant delay time that does not change the individual nodes of the communication link. In general, such a communication connection is also referred to as a circuit-switched connection.

In contrast, in the case of a packet-oriented transmission, the individual packets can be transmitted to the recipient via different routes. Consequently, it can also come to different terms here. A typical example of such a packet-oriented communication connection is in Internet-based transmissions.

1 shows a system of a transmitting unit 1 , a receiving unit 3 and a network in between 2 for a communication between transmitting unit 1 and receiving unit 3 ,

The transmitting unit is designed in the present case as a mobile communication device. It contains a microphone 131 for recording an analog voice or audio signal, for example from a user. This is done by a coding device 13 converted into a mobile data compliant data stream B. The coding device contains this 13 For example, a signal processor that converts the analog audio signals into digital signals. These are filtered low-pass, and coded according to a method provided for the mobile radio standard. Subsequently, the thus encoded and processed analog audio signals as data stream B via an antenna 19 to a communication network 23 issued. The term data stream is to be understood in the following as a generic term for all types of signals for the transmission of information.

In the present case, the communication network includes 23 a base station and other landline connections with multiple landline nodes. Since a transmission according to a mobile standard between the landline nodes is usually not provided, the base station generates in the network 23 from the data stream B another signal. For example, this can now take place in the form of a PCM-coded variant of the analog audio signal A present in the data stream B.

Finally, a data stream C, which in the ideal case contains the same user data (ie the audio signal A) as the stream B to the base station 24 issued. This generates the signal D according to a predetermined mobile radio standard and sends it to the receiving unit 3 ,

From the receiving unit 3 the signal D is received and decoded. The decoding includes, for example, the conversion and extraction of the encoded audio data. These are used as data stream E in the audio decoder 31 fed, which converts them into the analog audio signal F. On the output side is the audio decoder with a loudspeaker 37 coupled to the output of the audio signal.

According to the invention, it is provided to establish a further communication connection to an already existing communication connection for a transmission of voice or audio signals. With the second connection video signals to the receiving unit 3 be transmitted.

The transmitter unit contains this 1 another entrance 132 For example, to a video camera 15 for detecting video signals I is connected. The analog image information I is from the video camera 15 converted into a digital video signal J and sent to the video encoder 11 transmitted. For example, the digital video signal J may include YUV encoded image information at a frame rate of 30 frames per second. The video encoder 11 generates from it a compressed digital video signal M. A compression is particularly useful if only a limited bandwidth is available for the data transmission of the video signal. For example, digital video information can be very well compressed and encoded using the MPEG or MPEG2 method. A compression and coding with the video encoder 11 also has the advantage that the video encoder 11 additionally time stamp or general time information K in the compressed video signal M can store. For example, the time information can be transmitted by means of RTP timestamps.

The total signal N is a data stream over the interface 18 to a second network 22 transmitted. In the present case, this represents the Internet, which forms a packet-oriented network. For transmission of packets of the data stream N, for example, a UDP / IP protocol can be used. The individual packets are distributed over several nodes within the Internet 22 finally as a data stream 0 to the base station 24 transmitted to the receiving unit as a signal D using a mobile communication standard 3 transmitted.

The transmitted signal D is decoded and the video information therein in the form of the signal R to a video decoder 33 passed on. This decodes the video information contained in the signal R and gives it optionally via a buffer 35 to a screen 38 from.

Because of the network 22 used UDP / IP protocol may lead to a faulty data transmission of the video information or the data stream N and O respectively. For example, individual packets may be lost. It is also possible that the different packets of the stream N at different times at the base station 24 and thus in the receiver 3 arrive. Another problem is that there is no direct temporal relationship between the first communication connection for the transmission of the audio signals A in the first data stream D and the second communication connection for the transmission of the video signals I. As a result, image or audio signals can be output at different times. A synchronization between sound and picture is no longer guaranteed.

For a coordinated output of the signals A and I in the receiver is an analysis device in the transmission arrangement according to the invention 12 and an encoder 11a intended. The analyzer 12 is in the present example with the input 131 for the audio signals A coupled. It generates from the detected audio signal A identification features L, with the help of which sections or areas of the audio signal can be identified. The identification features may be associated with the encoded video signal M so as to achieve a determination of which portions of the audio signal to Sections of the video signal. For this purpose, the analysis device decomposes in the exemplary embodiment 12 the audio signal in sections of the same length.

Each subsection of the audio signal is selected so that its output takes place in a temporally defined relationship to a representation of one or more images of the detected video signal. The analyzer 12 extracts from each subsection an identification feature L, a so-called fingerprint. This identifies the respective subsection of the audio signal A uniquely. The uniqueness is through the synchronization device 17 ensures that adds a time stamp K both the identification feature L and the corresponding section of the video signal. In other words, the identification feature L with the help of the synchronization device 17 a portion of the video signal J in the video encoder 11 assigned.

The identification feature is chosen so that on the one hand it is particularly robust against possible interference during data transmission. On the other hand, it is also robust against artifacts that occur due to signal processing or compression during the transmission of the first data stream B. These artifacts can affect the useful signal in the first data stream B.

The compressed video signal M and the identification feature are in the encoder 11a summarized to the data stream N.

4 shows a schematic representation to illustrate this assignment. The audio signal A, represented as a simple sinusoidal oscillation, is subdivided into different subsections TA1, TA2, TA3, each of equal length. In the embodiment, the length of each subsection TA1 to TA3 is correlated with the digital image information VB1 to VB3 of the video signal J. In the present case, subdividing the audio signal A into subsections to each second video image VB1, VB2 and VB3 of the digital video signal J. Da Video signal consists essentially of a number of temporally successive frames, it is easily possible to achieve a synchronization in that a particular section of the audio signal A is played back to a video image.

For each of the subsections TA1, TA2 and TA3, a fingerprint, a unique identification feature FP1, FP2 or FP3 is now determined. This fingerprint may be determined, for example, by LTP or LPC coefficients. Such coefficients are generated when encoding the speech signal A and thus can be easily used without additional effort. Another possibility is to determine particularly robust frequency information for each subsection. After generation of the respective identification features, these are assigned to the corresponding video image of the digital video signal.

For an assignment, the video images VB1 to VB3 as well as the generated identification features FP1 to FP3 are now provided with a time stamp K on the one hand. This time stamp K allows the assignment of the identification features L to the coded video signal M with the coded video images.

In the coding device 11a to 1 the identification features L and the coded video signal M are then combined. This can be done, for example, by multiplexing the two data streams L and M. The data stream N generated by multiplexing in the present example is designed as a packet-oriented data stream whose packets contain sections of the coded video signal M and the associated identification feature. Each packet contains the subsection VB1 to VB3 of the video signal J, which is assigned to the corresponding identification feature FP1 to FP3.

Because of the time stamps K contained in the identification features L and the coded video data stream M, an allocation is ensured even if a different package form is selected. Thus, video data stream M and the identification features L can also be transmitted in different packets or even different data streams.

Another possibility is, within the encoded video data stream M, the identification features L as additional information, engl. "Supplementary enhancement information". One video standard that allows this is the H.264 / AVC standard. An additional multiplexing can then be omitted, the coded and provided with the features L video stream M forms in such a case the data stream N. Basically, only an association between the identification features in the stream L and the video signal J should be guaranteed.

Present includes 4 a packet of the multiplexed data stream N a header portion RTP, are stored in the header information for each data format and time stamp. Connected to it is a second portion in which the identification feature FP1 is stored. The video information part DVB1 contains the image information of the digital video signal J starting from the video image VB1 to the video image VB2. In a corresponding manner, the coding device multiplexes 11 Likewise, it adds the identification features FP2, FP3 from the feature stream L accordingly.

The thus prepared second data stream N is again through the second communication link via the network 22 to the recipient 3 transfer. The received data stream P comprises not only the coded video signal R but also the additional identification features Q.

The feature stream Q becomes the device 36 fed. This again determines the identification features and the timestamps that have been supplied to the identification features on the transmission side. In addition, the receiver contains 3 a synchronizer 34 , the input side to the output of the audio decoder 31 for supplying the decoded speech or audio signal F is connected. In addition, he is at the facility 36 connected to the signal U, which includes the identification features. The synchronizer 34 serves to, from the sections of the speech signal F of the audio decoder 31 To generate identification features. That from the synchronizer 34 The method used for generating the identification features essentially corresponds to the analysis apparatus 12 within the transmission arrangement 1 , The generated features will then match those of the facility 36 compared transmitted characteristics. If there is a match or a high similarity, it is assumed that a corresponding section of the video signal is present for the subsection of the speech signal F.

To ensure a concerted output is the audio decoder 31 in addition to a buffer device 32 connected for buffering the audio signals. Another buffer device 35 is used for buffering the decoded video signal W. They are used to correct a corresponding time offset due to different transmission times or due to the different signal processing. Both buffer devices 23 and 35 be from the synchronizer 34 controlled by the control signals V and X. Depending on the drive, the buffers supply the sub-sections G, Y of the speech signal F and the video signal W selected by the control signal to the output devices 37 and 38 from.

The synchronizer is with the video decoder 33 and the facility 36 for supplying the time stamp information S, T coupled. Together with the comparison of the identified identification features with the identification features of the device 36 and the timestamp information is generated by the synchronizer 34 the control signals V and X. It is useful, in addition to the audio buffer 32 also the video buffer 35 in the event that the video signal before the corresponding audio signal, the receiver arrangement 3 reached or the signal processing in the synchronizer takes longer than a decoding of the video stream R.

The additional buffering also makes it possible, in the case of lost data, to emit the corresponding signals of the respective other data stream. For example, areas of the audio signal F can also be output if the video signal W corresponding thereto is lost or not reconstructable during the transmission. Due to different delays in the two networks 22 . 23 it is expedient to choose a suitable buffer size. In addition, it may be useful, if the identifiers in the first or second data stream match, to not immediately output the corresponding sections, but to buffer them again instead. This has the advantage that later output can be largely continuous without any later delays due to longer signal propagation times in the network 2 ,

A buffering or delaying of an output of the speech signal can also be effected by a special processing of the speech signal. Instead of the buffer, a signal processor may be provided, which performs a time-based modification of the speech signal. So the speech signal could be dynamically distorted, eg. For example, vocal sounds may be dynamically lengthened or shortened at the output, thereby achieving a delay. This "time scale modification" method is particularly suitable, since a continuous output of the speech signal without interruptions made, but still time for the processing of the second data stream and the vote of an output is obtained.

During the transmission of the first data stream in the network 23 and the second data stream in the network 22 it can lead to signal interference or signal changes. So in the first network 23 the encoded audio signals are otherwise processed. This changes the frequency spectrum, z. B. phase information, the first data stream B and its timing. It is therefore appropriate to choose identification features that are particularly robust against such interference. For example, the so-called LTP or LPC coefficients for the speech signal (Long-Term Prediction, Linear Predictive Coding).

2 shows a further embodiment of a system according to the invention with a transmission arrangement 1 , a receiving arrangement 3 and a network connected in between 2 , The transmitter 1 has a first detection device 130 for detecting the first useful signal, for example of speech signals. Another arrangement 150 serves here for generating and displaying a computer-aided presentation J '. The presentation J 'may include audio and additional video data as well as still pictures.

The first device 130 encodes the detected first useful signal and generates from the encoded first useful signal identification features L, which allow identification of subsections of the first useful signal. The identification features L are a coding device 110 transmitted. The coding device 110 is also on the second device 150 connected. It generates the second data stream N ', in which the identification features L are combined with the information about the computer-assisted slide show J'.

The second data stream is sent to the second network 220 while the encoded audio signals are transmitted through an antenna 190 to the first network 230 be sent. Both networks 220 . 230 are each with a base station 240 . 250 for a mobile communication link with the receiving device 3 coupled.

In the present case, the transmission of the first data stream, that is to say of the speech signal, and of the second data stream takes place via two different transmission channels. For example, the encoded audio data D in the form of a GSM signal to the receiver arrangement 3 be sent. The second data stream, on the other hand, is sent to the receiving arrangement via a second transmission channel D1 3 transmitted. The receiving arrangement 3 includes two receiver stages 310 . 330 which process the two signals in the different transmission channels. They contain a decoder for decoding the audio signal or the screen presentations. In the synchronizer 340 An assignment of subsections of the decoded audio signal to the corresponding sections of the computer-aided screen presentation using the identification features.

Illustrates another aspect of the invention 3 , Elements of the transmitter 1 or the network 2 with the same function or effect as elements in the design 1 bear the same reference numbers. In the illustrated embodiment, the analysis device for generating coded and provided with time stamps K audio signals L is executed. These serve as identification features and are combined with the coded and time stamped video stream M to the second data stream N. About the first network 23 Furthermore, the first data stream B is transmitted, which contains a coded speech signal. About the second network 22 the transmission of the data stream N. the receiving arrangement 3 ' is configured to decompose the received second data stream P back into the current Q and the current R. Furthermore, it contains the device 36 , which is embodied here for decoding the feature stream Q into speech signals U and time stamps S. Together with the existing time stamps S and T. which were added on the transmitter side as a time stamp K, a time-coordinated output of the speech signal U and the signal W is possible.

The synchronizer 34a is designed to, on the one hand, determine the voice quality and, in the case of an erroneous reception of the second data stream P, to switch over to the still existing first communication connection for the output of the voice signal F. This is useful, for example, in a transmission of the first and second data stream with different mobile radio standards. If, for example, the first data stream is transmitted as a GSM mobile radio signal and the second data stream is transmitted using the standard UMTS, a continuation of the circuit-switched communication connection is possible even if the second data stream fails.

With the elements of the illustrated systems according to the invention can be in an existing communication connection for transmitting a first data stream, a second communication link for transmitting a second data stream without interrupting the first communication link Add. It is likewise possible to set up a new communication connection for transmitting the first and the second data stream. A timed output is provided by the identification features added to the second data stream derived from the first data stream. The additional identification features in the second data stream are added on the transmission side. This is possible, for example, in the form of a fingerprint or additional information within the transmission protocol used for the second data stream. In addition, further information can be transmitted, for example, timestamps. By using a suitable transmission protocol, such as the Real Time Protocol, the addition of timestamps to data or identification features is still formalized and their use for synchronization simplified.

In addition, it is possible to transmit the same identification features by a further transmission channel to the receiver. It merely has to be ensured that the identification features in the further transmission channel can be assigned to the individual sections of the second data stream.

5 shows an embodiment of the method according to the invention. In a first step S1, a receiving arrangement receives a first data stream in the form of a first signal. The data stream is processed and the useful signal contained therein is buffered in a buffer. Parallel or substantially simultaneously for this purpose, a second data stream is received in step S2. In addition to second payload data, this also contains identification features which are suitable for assignment to the payload signal in the first data stream. The identification features are derived from the first useful signal.

In step S3, the identification features contained in the second data stream are extracted. Subsections of the useful signal in the second data stream and the identification features of the subsections of the first data stream assigned to the subsections are then temporarily stored in step S4. During steps S3 and S4, that is to say the determination of the identification features in the second data stream, it is possible to analyze the subsections of the buffered useful signal. This process takes place in step S5. In this case, identification features are determined from the individual subsections, which correspond in type and structure to the identification features of the second data stream. If, for example, the identification features in the second data stream represent LTP coefficients (Long Term Prediction), the LTP coefficients are also determined in step S5 from the individual subsections of the stored first data stream.

In step S6, the identification features determined from the subsections of the useful signal are then compared with the one or more identification features of the second data stream. For example, each identification feature of the subsections of the useful signal in the first data stream can be compared with each of the identification features determined from the second data stream. For this purpose, different comparison methods and methods are available in step S7. The comparison can be designed such that a positive result is generated only if there is an exact match of an identification feature of a subsection of the first data stream with an identification feature of the second data stream.

However, since errors may have occurred within the two data streams during a transmission, or distortions or other changes may have occurred due to signal processing or compression, the identifiers may also be slightly different. Therefore, it may be appropriate to provide for some freedom of choice when making a decision. The match is thus determined only with a probability. If this exceeds a predetermined value, then a match is accepted. Another possibility would be to perform a correlation analysis for the identification features of the first and second data streams.

If the comparison process in steps S6 and S7 reveals that the identification features of the subsections of the first data stream do not match the identification feature (s) in the second data stream, the comparison is repeated with new identification features of the first or second data stream. In the case of a match, on the other hand, in step S8, the subsections of the first and second data streams assigned to the two identification features are loaded from the buffer. In step S9, the sections are finally output timed to each other.

By buffering the first data stream and buffering the second data stream, a time delay during a receive operation of the first or second data stream can be corrected. By analyzing the useful signals in the first data stream and determining the Identification features eliminates the need to insert in advance timestamps in both streams for later output.

Rather, any useful signals may be provided in subsections in the second data stream, each of the subsections being assigned identification features if these are to be output synchronously to the subsections of the useful signal in the first data stream. A special choice of the second useful signal is not necessary. For example, by the presented method, a video signal as a useful signal of the second data stream can be synchronized with the aid of the identification features to an already existing transmission of a speech signal. It is not necessary to transmit the first and second data stream over the same transmission channel to a receiver.

6 shows an extension of the inventive method, which is taken into account when parts of the first or second data stream are lost or damaged during transmission. In this embodiment, it is assumed that the first data stream forms an analog or digital telephone signal for the transmission of audio signals. This is delivered from the transmitter to the receiver via several nodes.

To the already existing communication connection, a second data stream is now to be added, which contains a second useful signal, which is synchronized to the telephone signal. For this purpose, in step S10, the first and the second data stream are received substantially simultaneously via a common transmission channel. This transmission channel can represent a mobile connection according to the UMTS standard, for example. Another possibility would be to receive, for example, the first data stream via a telephone connection and the second data stream packet-oriented over a DSL connection. In such a case, the transmission of the two data streams takes place via the same physical line.

In step S11, the common received signal is decoded or separated into the first and the second data stream. The first data stream as a telephone signal is divided into different sections of equal length in step S12. However, these sections are not arbitrary, but follow a predefined pattern that results from the first data stream. In step S16, these sections are cached for later processing.

For processing the first data stream in steps S12 and S16, the second data stream with the second useful signal is also processed. The second useful signal also includes a plurality of subsections, which are separated in step S13. Certain subsections are each assigned an identification feature that is extracted from the second data stream and stored in a memory.

Although the reception of the first and second data streams is generally simultaneous, this does not mean that the corresponding sections of the first and second data signals have a specific temporal relationship to one another. For example, due to different transmission paths between the transmitter and the receiver, a time offset between corresponding subsections of the first data stream and subsections of the useful signal in the second data stream occur. In other words, subsections of the first data stream may arrive earlier or later at the receiver than the corresponding subsections of the second useful signal. If this time offset now exceeds a predetermined value, it is no longer expedient to buffer late arriving partial sections. This is especially meaningless if, for example, previously arrived sections of the first data stream have been further processed and possibly already output.

In the present case, it is assumed that the second useful signal is delayed with respect to the telephone signal. In step S15, it is therefore checked whether the identification features assigned to the subsections of the second data stream have already been output. This can be done by a brief comparison of the extracted identification features of the subsections of the second useful signal with identification features of already output subsections of the first data stream. Another possibility for such a check is when individual subsections of the second useful signal in the second data stream are assigned time stamp information. Thus, all the subsections can be discarded, which include a timestamp, which are before the last time stamp issued a subsection of the second useful signal.

If such a comparison shows a match in step S18, then the corresponding subsection of the second useful signal is discarded in step S19, since it is synchronous with this subsection to be outputted subsection of the first data stream has already been issued. If necessary, the subsection of the second useful signal and the associated identification feature are stored in step S20.

In step S17, an analysis and determination of the identification features of the subsections of the first data stream is undertaken.

In step S21, the identification features of the subsections of the first data stream are compared with the identification features from the second data stream. If such a comparison results in a match in step S22, this essentially means that the subsections of the first and second data streams are to be output in a coordinated manner. These are loaded from the respective buffer in step S23 and output synchronously in step S24.

If, on the other hand, a subsection of the second useful signal to the corresponding subsections of the first data stream has not yet been received and processed, the comparison procedure in steps S21 and S22 results in no match. In such a case, it is assumed that the corresponding subsection has not yet been received in the second data stream. Therefore, a counter is incremented in step S25. The counter indicates whether the subsection in the telephone signal should be output in the second useful signal despite the missing subsection. For this purpose, the counter is compared with a predetermined value in step S26. If the counter is smaller than this value, the method is continued with the further analysis of newly received subsections of the first data stream.

Otherwise, if the counter is greater than the predetermined value in step S25, the portion of the telephone signal to be outputted next is determined in step S27 and output in step S28. The identification feature belonging to the output subsection is now marked as already issued and serves to identify a subsection of the second useful signal with the same assigned identification feature. Since the associated subsection has already been output in the first data stream, the subsection in the second data stream can be discarded.

In the embodiment of the method, it is expedient to take into account both a delay of the subsections of the first data stream relative to the subsections of the second data stream and the reverse possibility.

7 shows a further embodiment of the method according to the proposed principle. In this exemplary embodiment, besides the method steps for receiving the first and second data streams, the method steps during the transmission process are also shown.

In step S30, a communication link for transmitting a first useful signal or a first data stream is set up between a transmitter and a receiver. The structure is circuit-switched and remains until the end. The first data stream contains the useful signal, for example a voice signal. During a period of time that follows, in step 31 a continuous communication. Voice is transmitted from the sender to the receiver via the circuit-switched connection.

This communication is also continued in step S32 over the already existing connection. In addition, a second communication connection should now be established to the already existing first communication connection, via which a second data stream is transmitted. The useful signal of the first and second data stream should be synchronized with each other. For this purpose, the first and the second useful signal are each divided into subsections.

During the continuation of the first communication connection and the transmission of the first user data in step S32, a second communication connection is set up via a further transmission channel in step S33. This transmission channel can be constructed differently with respect to the transmission channel for the first communication connection. For example, while the first transmission channel for the first communication connection comprises a fixed network connection, the second transmission channel includes a packet-oriented Internet connection.

In step S34, a voice or audio signal in the form of several subsections is now detected as the first useful signal. Essentially at the same time and synchronously, a corresponding video signal is detected in step S35, which likewise has several subsections. In step S36, identification features are determined from the detected audio signal. These features can be determined both from the detected analog audio signal, as well as from a digital audio signal or an already coded audio signal. The term encoded audio signal is understood to mean processed audio signals that have been encoded according to a method. Examples of such coding methods are the AMR or the AMR-WB coding method, the GSM EFR coding method, G.72x coding method of the ITU-T standard for speech compression or combinations thereof. These are particularly suitable for speech coding.

The identification features that were determined in step S36 allow a unique identification of the respective subsection of the speech signal. In a coding according to one of the above-mentioned methods, it is also conceivable to use features which are generated during the coding as identification features. Thus, the LPC coefficients generated in the AMR coding can also serve as identification features. In addition, the generated identification features should be particularly robust against disturbances or artifacts occurring during the transmission of the first data stream.

The identification features determined in step S36 are added to the second data stream during the encoding of the subsections of the video signal in step S37. This is done such that an identification feature of a subsection of the speech signal is assigned to a subsection of the video signal that is synchronous with the corresponding subsection of the speech signal. This is achieved by timestamps or a suitable combination of the video signal and the identification features.

After coding the video signal in step S37, the second data stream and the first data stream are transmitted in steps S38 and S39. The transmission may include various transmission channels as well as a common transmission channel. By a common transmission channel is meant a transmission channel on the same frequency, using the same communication standard or using the same network nodes. During steps S38 and S39, however, a time shift or a time delay of the first and second data streams may occur. This results in a reception of the first data stream in step S40 and a reception of the second data stream in step S41 at different times.

In step S42, the received first data stream is decoded and the speech or audio signal contained therein is extracted. This is subdivided into several sections, each of which is analyzed and an identifier is extracted. In parallel, in step S43, the received second data stream is further processed. The identification features added in the second data stream of the subsections of the video signal are extracted. These identification features are then compared in step S44 with the generated identification features of the subsections of the buffered audio signal. If the comparison gives a match in step S46, the respective subsections of the audio signal and the video signal are loaded in step S45 and output synchronously in step S47. Otherwise, in step S48, the audio signal may be output without a corresponding video signal.

Since the sequence of the subsections of the audio signal is detected, it is thus also possible to output subsections of the audio signal that are temporally ahead of a subsection of the audio signal to which a corresponding subsection of a video signal was found. As a result, a continuous output of the audio signal can be ensured, especially in the case of error-prone transmission of the second data stream.

The illustrated method has the advantage that it can be applied in a simple manner to already existing mobile radio standards or communication connections. In particular, it is possible to continue an already existing communication connection without much additional effort. It merely has to be ensured that the received first circuit-switched useful signal is buffered so that synchronization with a possibly later received second data stream is ensured. It is not necessary that each subsection of the first useful signal is analyzed. This is rather dependent on the requirements of the second data stream. For example, when transmitting video signals at thirty frames per second, it may be convenient to break the voice or audio signal into 1/30 second increments and extract a corresponding identifier from each of these subsections. However, it is also possible to associate only every xth image of the video signal with an identification feature of a corresponding subsection of the audio signal. It is conceivable in the method according to the invention that the receiver does not know the exact assignment. Nevertheless, a synchronous output can be made if the subsections of the audio signal to which no subsection of the video signal is assigned are also output continuously.

In addition to the synchronization of a video signal to a corresponding audio signal, the method according to the invention as well as the arrangements can be applied to all data streams which contain useful data that can be timed to one another. For example, it is also possible to transmit a common video signal to a receiver via a plurality of transmission channels and multiple data streams. It is also conceivable a voice or audio transmission in multiple streams (or signals), for example, to allow a stereo signal. A synchronization of the data streams can be done as indicated. Also, the method and the arrangement is not limited to the synchronization of two data streams, but can be extended to multiple data streams (more than two), provided that each user data to be synchronized allow the generation of unique identifiable identification features.

It should be noted that in the case of user data contained in data streams in coded form, the synchronization with respect to the data streams before their decoding or alternatively with respect to the user data streams obtained from the data streams after the decoding can take place. Furthermore, it is also possible to carry out the synchronization before or after a recoding of the data streams.

An interesting possibility is further that the synchronization is not carried out by functionalities of a terminal but by functionalities of an infrastructure device such as a mobile telephony operator, which then synchronizes, for example, a first, about voice data stream and a second, for example, containing video data stream to a with respect to the voice data and the video data synchronous sequence data stream combined, possibly based on a decoding of the coded voice data containing first data stream and the coded video data containing second data stream and coding the synchronized, still uncoded voice and video data stream to a coded voice and video data stream.

In the following, an embodiment for a particularly expedient manner of synchronization of multiple media streams, especially an audio stream, special speech data audio stream, on the one hand, and a video data stream, on the other hand, will be given. It will be on the 8th and 9 Reference is made in the for analog data streams reference numerals corresponding to the reference numerals of 1 to 7 are registered.

It should be noted that coded voice data generated by a voice codec depending on the voice codec and the underlying principles of synchronization with another media stream, in particular video stream, particularly easy, since the content of the voice data stream B or certain data or parameter itself can serve as an identification feature or identification features (can also be labeled as a fingerprint). In this case, consideration is primarily given to data and parameters based on prediction analysis, in particular linear prediction analysis, such as LPC parameters, LSP parameters, pitch information, codebook gain, for example, of the given codebook, and combinations of these parameters.

According to the schematic representation of 8th is a user terminal 100 with a microphone 102 and a video camera 104 Running and includes facilities that have a voice codec 106 and a video codec 108 realize. The voice codec 106 encodes that from the microphone 102 obtained speech payload data stream and generates the coded voice data contained first data stream B, for example, formed of speech frames with a duration of 10 msec or 20 msec and based on a linear prediction analysis by the speech codec based on data frame-wise, such as LPC parameters , LSP parameters, pitch information (P), gain or gain G with respect to a fixed codebook, gain or gain aG with respect to an adaptive codebook, etc. The video codec 108 generated from one of the camera 104 incoming video payload a coded video stream M. The voice codec 106 represents selected data of the coded voice data, such as an LPC parameter and an LSP parameter, as fingerprint information L of a combiner 110 available, which adds the encoded video stream M to the fingerprint information, thus generating the data stream N. Preferably, there is a frame-wise correspondence between the audio data stream B and the video data stream M and, accordingly, the supplemented video data stream N, such that respective frames correspond to the same length of time with respect to the audio and video payload, for example 10 msec or 20 in length msec, so every voice data frame 112 exactly a video data frame 114 of the video codec 108 given video data stream M and exactly one data frame 116 of the video data stream N supplemented in each data frame by the fingerprint information L and Q, respectively.

It should be noted that it does not matter whether the fingerprint information is included in separate data frames or is included with the video data in common data frames.

The first data stream B and the data stream N contained in the video and fingerprint information to be identified as the second data stream are transmitted via different transmission channels and based on various transmission standards and protocols. This is in 8th on the one hand by the arrows 114 and 118 representing a circuit-switched combination connection and, on the other hand, a symbolically represented router network 117 , which stands for a packet-switched communication connection, for example via the Internet.

As a result of the different transmission channels and transmission standards, the synchronization between the first data stream B and the second data stream N on the receiver side, which is initially present in accordance with the exemplary embodiment, is canceled. The received, the coded voice data containing first data stream E and the received, the coded video data R and the fingerprint information Q containing second data stream P must first be synchronized, which can be done appropriately before the decoding.

A synchronizer 120 receives the data streams and extracts from the audio data frames 112 those parameters and data corresponding to the fingerprint information carried in the second data stream, in the present case therefore the LPC and LSP parameters. Since voice data encoded as fingerprint information itself is used, an extraction device 122 simply to the relevant data in the respective voice data frame 112 access. In a corresponding manner extracts an extraction device 124 framewise the fingerprint information carried in the second data stream, in this case the one from the speech codec 106 output LPC and LSP parameters carried in the second data stream.

A comparison device 126 compares the parameters extracted from the first data stream and those extracted from the second data stream, thus allowing a pairwise association between speech data frames 112 the first data stream and data frames 116 or data frames 114 determine the second data stream. Suitably, the extraction device 124 not just the fingerprint information to the comparator 126 but provide from the second data stream P contained in the frame of the fingerprint information frame a second data stream R containing only the coded video data.

In the embodiment of 8a) and 8b) It has been assumed that, due to the packet-switched transmission of the second data stream, the latter is delayed with respect to the first data stream, so that the synchronization device 120 just a cache 128 for the data frames of the first data stream. The data frames are buffered in this buffer and under the control of the comparison device 126 frame-wise in synchronism with the video data stream R output. These frames are then passed through a receiver-side voice codec 130 and a receiver-side video codec 132 decodes and the resulting speech payload and the resulting video payload are timed by a loudspeaker 134 or a screen 136 played.

As already indicated, this could be done by the Snychronisiereinrichtung 120 Synchronization carried out by a device of a network infrastructure or the like, which then, for example, a combined audio-video data stream in which the voice and video information is synchronous to each other, provides and via a device leading to the transmission channel, for example via a Air interface, this combined data stream to the receiver terminal transmits. A corresponding embodiment variant is in 8c) illustrated. The synchronization device 120 ' is here with a combination device 138 executed, which brings together the synchronized audio and video streams to a combined data stream K, the one through the arrow 140 symbolized, for example, circuit-switched transmission channel is transmitted to a terminal, in which then split this data stream in the voice data on the one hand and the video data on the other hand and then this data by means of the voice codec 130 or the video codec 132 decoded and by the playback devices 134 . 136 be issued. In this embodiment, the synchronization is 120 ' additionally with a cache 129 for the video data frames to also be able to compensate for delays of the audio data stream relative to the video data stream. Also the synchronizer 120 the embodiment according to 8b) can advantageously with such, the video data associated buffer 129 be executed.

With regard to the synchronization of the media data streams, it should be noted that distortions could occur due to disruptions of the respective transmission channel and that ambivalences of the mutual assignment of the data frames of the first and second data stream, possibly even more than two data streams, could also occur. The danger of ambivalences of the kind that no unambiguous assignment between respective data frames of the first and second data stream is possible, can in principle by appropriately extensive fingerprint data is virtually eliminated. However, it may be desirable to manage with a minimum of fingerprint data in terms of data transfer rates.

A particularly secure association between the data frames of the first and second data stream, which is very fault-tolerant with regard to interference from the transmission channels or other types of interference and moreover avoids ambiguities with good reliability, is based on the fact that for the assignment the fingerprint information of a plurality of data frames of the second data stream, Here the video data stream is compared with several data frames of the first data stream, here the audio / voice data stream. It is assumed below that parameters par_1 to par_N (in particular LPC parameters, LSP parameters, etc. as mentioned above) form the fingerprint information and for each voice data frame of the first data stream associated with a respective video data frame in the video data stream (second Data stream), preferably within the respective video data frame. In the embodiment of a voice and video communication in a telecommunications environment which is the basis here, this fingerprint information is extracted on the transmitter side from the coded voice data frame or provided by the voice codec without changing the voice data. These fingerprint data are then transmitted together with the associated video frame or in the associated video frame.

ermittelt werden, wobei DM eine Funktion in der Art eines Abstandsmaßes oder eines reziproken Abstandsmaßes ist, damit e(i, k) einem Fehlermaß entspricht oder eine Funktion ist, die dem reziproken eines Abstandsmaßes entspricht, also einem Übereinstimmungsmaß. Es wird hier im Folgenden der Einfachheit davon ausgegangen, dass DM ein Abstandsmaß ist in der Art einer quadratischen Norm oder Ähnliches, beispielsweise Euklidischer Abstand, absolute Differenz, Itakura-Saito, logarithmischer Abstand und dergleichen.For a respective video frame of the second data stream, the associated i-th voice frame of the first data stream may be identified based on the similarity between the transmitted fingerprint data and corresponding fingerprint data determined for the voice data frames. From the received sequence of voice data frames, the fingerprint information is determined or extracted from the receiver side (or alternatively in the infrastructure device) for each data frame of the first data stream. It may then for a respective data frame i of the first data stream, an error or agreement measure

where DM is a function in the form of a distance measure or a reciprocal distance measure so that e (i, k) corresponds to an error measure or is a function that corresponds to the reciprocal of a distance measure, that is, a measure of agreement. Hereinafter, the simplicity is assumed that DM is a pitch measure such as a quadratic norm or the like, for example, Euclidean distance, absolute difference, Itakura-Saito, logarithmic distance, and the like.

In this formula k represents an assignment parameter in the form of a search distance or shift measure, which corresponds to the i-th data frame of the first data stream, exactly the fingerprint data determined therefrom, the (i-k) -th data frame of the second data stream, more precisely the one ( i - k) -ten data frame of the second data stream associated, referring to a specific data frame of the first data stream fingerprint data in the second data stream assigns.

Thus, for several data frames i of the first data stream, the variable e (i, k) is determined for different values of k in order to perform the synchronization on this basis.

If ten LPC parameters (N = 10) are transmitted as the fingerprint information with the second data stream and a quadratic norm is used for the determination of the error measure, the following results are obtained for the speech data frame i and the search distance k: e (i, k) = [LPC1 (i) -LPC1 (i-k)] ^ 2 + [LPC2 (i) -LPC2 (i-k)] ^ 2 + ... + [LPC10 (i) -LPC10 (i - k)] ^ 2 e (i-k) is calculated for all values of k of interest, ie within a certain search range around the speech data frame i to be assigned to or to an associated video data frame, for example for the k values k = 1, 2, ... , 50. Usually, an order of 10 is sufficient for k, ie for a frame length of 20 msec, a search range of 200 msec.

oder

It is quite appropriate to perform the synchronization on the basis of the error measures (or matching measures) e (i, k) alone. In contrast, however, it is preferable to consider a plurality of dimensions e (i, k) determined for the same assignment distance k, for example in the form of a combination comparison measure

or

This combination measure can be determined for each value k, the sum going over all data frames of the first data stream to be considered. Unless a fixed but a varying number of variables e (i, k) are included in the combination measure E (k), the sum should be nominated by the quantity M indicating the number of summands.

The combination comparison measure E (k), which in particular corresponds to a summary or average error measure or a summary or average match measure, is much more error-tolerant and less susceptible to ambivalences than the underlying measures e (i, k), since a summation or Mittlang over the time or over a plurality of data frame pairs of a respective data frame of the first and a respective data frame of the second data stream is included. Other combinations of the dimensions e (i, k) are also conceivable, for example a product formation.

Depending on the definition of the incoming function DM, a k value can be determined as a synchronization measure by determining a maximum or a minimum of the combination comparison measure E (k). In the case of a measure DM realized as a distance measure, the measures e (i, k) and accordingly also the combination measures E (k) for better match between the fingerprint information of the second data stream on the one hand and the fingerprint information of a respective data frame of the first data stream on the other hand decrease, so that a minimum of E (k) can be found. For example, if a minimum of E (k) is obtained for k = 3, it means that the video and voice data streams have diverge by three data frames, that is, the video stream is delayed by three frames from the audio stream. This makes it possible to synchronize the video stream (second data stream) with the first data stream (audio or speech data stream) by artificially introducing a corresponding delay for the audio data stream by storing it approximately in the buffer 128 ,

Advantageously, the combination comparison measures E (k) can be calculated iteratively. First, we calculate for each value of k: E [1] (k) = e (1, k)

It can then be determined as a first approximation k as a minimum of E [1] (k).

Once the next speech data frame is present, then E [2] (k) = e (1, k) + e (2, k) = E [1] (k) + e (2, k) which allows the determination of k as a minimum of E [2] (k).

The third iteration stage can then be k E [3] (k) = e (1, k) + e (2, k) + e (3, k) = E [2] (k) + e (3, k) be determined. This can be continued accordingly: E [n + 1] (k) = E [n] (k) + e (n + 1, k)

As long as only comparison measures E [p] (k) are taken into account for the minimum or maximum search, which contain the same number of summands, normalization with the number of summands is not necessary, but quite expedient. Taking into account a normalization quantity M indicating the number of summands, the following approximate combination comparison measures for the first to third approximation stage result: E [1] (k) = e (1, k) E [2] (k) = [e (1, k) + e (2, k)] / 2 E [3] (k) = [e (1, k) + e (2, k) + e (3, k)] / 3

Even in a recursive or iterative calculation, the correct normalization factor, if needed at all, can be easily taken into account.

The recursive or iterative calculation is based on averaging or offsetting over time. For each additionally considered speech data frame, a better approximation for the combination measure E (k) is obtained.

During the synchronization, it can be determined for the synchronous output of the two data streams or for a synchronous merging of the two data streams in each case for which value of k the combination comparison measure E (k) is minimal (or alternatively also maximum). In each case, buffered data frames of the first data stream and buffer memories assigned to the second data stream of the second data stream can be taken into account in each buffer, whereby additional data frames of the first or second data stream can be taken into account successively.

The manner of synchronization explained above in general will be better illustrated below by means of a greatly simplified example. It will open 9 Referenced. It is assumed that in a second data stream Q, a video data stream, fingerprint data in the form of the first data stream E, a speech data stream, referring parameters par_0 and par_1 are included, which are included as the payload data of the first data stream also in this parameter , In order to distinguish the parameters contained in the first data stream as payload data from the parameters recorded as fingerprint data in the second data stream, the latter are in 9 marked with an asterisk: par_0 *, par_1 *. This also takes into account that distortions may occur due to disturbances of a transmission channel or the like, so that the corresponding parameters in a data frame of the first data stream may differ from the parameters referring to this frame in the second data stream.

According to 9a are from the audio or voice data stream frames i - 1, i, i + 1, i + 2, i + 3, which arrived in this order and are recorded in an audio buffer. The assigned video data stream Q is delayed by one data frame, and therefore there are in a video buffer the data frames i-1, i, i + 1 and i + 2, which arrived in this order and are stored in the buffer.

For the sake of simplicity, the payload parameters par_0 and par_1 of the first data stream have integer values as follows: 6, 7 for the data frame i-1; 4, 5 for the data frame i; 2, 3 for the data frame i + 1; 0, 1 for the data frame i + 2 and 8, 9 for the data frame i + 3. In the second data stream, as fingerprints, the integer values 6, 7 for the data frame i-1 addressed as par_0 * and par_1 * are stored in the frame. 4, 5 for the data frame i; 2, 0 for the data frame i + 1 and 0, 1 for the data frame i + 2 recorded. In the case of the video data frame i + 1, the fingerprint data 2, 0 deviate from the corresponding parameters 2, 3 contained in the frame i + 2 of the first data stream in order to represent a deviation caused, for example, by a disturbance. Without interference, the video frame i + 2 would have to contain the fingerprint data 2, 3 or the audio frame i + 1 would have to contain the integer values 2, 0 for the specified parameters.

9a shows the error measures e (i, k) resulting for the illustrated data frames of the first and second data streams, the effects of the - here assumed - incorrect fingerprint data in the video frame i + 1 being characterized by "!?". In the case of the error measure e (i + 1, k = 1) results from the application of the distance measure in the form of an absolute difference, the error measure 3 instead of 0. Other error measures e (i, k) for other pairings of data frames of the first and second data stream are in 9a) and 9b) as well as the resulting combination comparative measures E (-1), E (0), E (1) and E (3). The combination measure E (k) is minimal for k = 1, which means that the video stream is delayed by one frame of data from the audio stream and accordingly the audio stream has to be delayed by one frame for a video stream synchronous playback.

If one compares a specific data frame of the one data stream, for example the frame i-1 of the speech data stream E, in pairs with different data frames of the other data stream, ie in the present example with the data frames i-1, i, i + 1 and i + 2 of the video data stream Q, the relevant data frame of the first-mentioned data stream-in the example the data frame i-1 of the data stream E-can be addressed as a "point-to-point" data frame, which is compared with different data frames of the other data stream. Alternatively, one could have a particular "look-up" data frame of the other data stream, ie, for example, the data frame i of the video data stream Q, in pairs with different data frames of the former data stream, ie, for example, with the data frames i-1, i, i + 1, i + Compare 2 and i + 3 of the voice data stream E. Alone on the basis of such a comparison can be done in principle the synchronization of the data streams. Preferably, however, the synchronization is based on the combination comparison measures E (k).

Proposed is, inter alia, a method and a system for synchronizing a plurality of data streams, each having a plurality of successive subsections, possibly data frames, wherein at least a second of the data streams in association with subsections of the second data stream on subsections of at least one of the first data streams reference identification features contains and the identification features associated with each other sections of the data streams are determined to synchronize the data streams. Devices and system components are also provided on the basis of which the method can be carried out or the system can be set up.

Claims (110)

A method of receiving, comprising a first data stream having at least two consecutive subsections and a second data stream comprising the steps: - receiving the first data stream (E); - receiving the second data stream (P) with a useful portion (W) and a first identification feature (Q, U) derived from a first subsection of the at least two successive subsections of the first data stream (E) and usable for an identification of the first subsection is; Delaying at least one of the first and second data streams; - Determining the first identification feature (Q, U) in the second data stream (P); - Determining the first section in the first data stream (E) in dependence of the first identification feature (Q, U); - Outputting the useful portion (W, Y) of the second data stream (P) in time using the first identification feature tuned to an output of the first subsection. A method according to claim 1, characterized in that the first data stream (E) and the second data stream (P) each have a plurality or plurality of successive subsections, wherein several or all subsections of the second data stream refer to a respective associated subsection of the first data stream , is assigned in the second data stream contained identification feature and for several of these identification features of the associated subsection of the first data stream is determined and output via the respective identification feature mutually associated portions of the first and the second data stream timed. A method according to claim 2, characterized in that for the assignment of a subsection of the first data stream (E) to a subsection of the second data stream (P) several, to different sections of the first data stream referring identification features are taken into account. Method according to one of Claims 1 to 3, in which the step of determining the respective first subsection in the first data stream (E) comprises the steps: Deriving an identification feature from at least one of the subsections in the first data stream (E); - Comparing the determined first identification feature (Q, U) with the derived identification feature of the at least one subsection of the first data stream (E). Method according to one of claims 1 to 4, wherein the step of receiving the first data stream comprises the steps of: - providing a first useful signal (A) comprising at least two successive subsections; Deriving an identification feature (L) from at least one of the successive subsections, which is usable for an identification of the at least one subsection of the successive subsections; - generating the first data stream (B) from the first useful signal; Transmitting the first data stream (B, C) over a network ( 23 . 24 ); Receiving the first data stream (D, E) from the network ( 23 . 24 ). Method according to one of Claims 4 or 5, in which the step of deriving an identification feature (L) comprises at least one of the following steps: Generating coefficients from the first subsection that are suitable for long term prediction; Generating coefficients from the first subsection that are suitable for linear predictive coding; Generating frequency information derived from a content of the first subsection; Generating HASH sequences from the first subsection; - generating a checksum from the first subsection; - Generating a coded subsection from the first subsection. Method according to one of Claims 1 to 6, in which the useful portion (W, J, M) of the second data stream comprises at least two subsections, and a subsection (Y) of the at least two subsections is assigned the first identification feature, and the output of the first subsection (G) of the first data stream and the subsection (Y) of the second data stream is timed to each other. The method of any one of claims 1 to 7, wherein the step of receiving the second data stream comprises the steps of: - Providing a second useful signal (J, M); - providing at least one first identification feature (L); - generating the second data stream (N, O, P) by assigning the at least one first identification feature (L) to the second useful signal (M); - Transferring the second data stream. The method of claim 8, wherein the step of providing the second payload signal (J, M) comprises dividing the second payload signal (J, M) into at least two consecutive subsections and in the step of generating the second data stream (N, P) at least one first identification feature (L) is associated with one of the successive subsections of the second useful signal (J, M). The method of any one of claims 1 to 9, wherein the step of receiving the second data stream comprises the steps of: - Disassembling the second data stream (P) in the second useful signal (R) and in the first identification feature (L); - delaying or buffering the second useful signals. The method of claim 9, wherein the step of outputting the second data stream comprises outputting the buffered second payload signals timed to output the first subsection. Method according to one of claims 1 to 11, in which the first data stream (B, E) comprises at least one of the following useful signals (A): - analog audio signals; - digital audio data or digital audio signals - encoded audio signals; - μ-law encoded audio data or μ-law encoded audio signals - GSM EFR encoded audio data or GSM EFR encoded audio signals; PCM encoded audio data or PCM encoded audio signals; AMR encoded audio data or AMR encoded audio signals; - AMR-WB encoded audio data or AMR encoded audio signals; G.729 encoded audio data or G.729 encoded audio signals; - encoded audio data or encoded audio signals according to one of the G.72x methods of the ITU-T standard for voice compression. The method of claim 12, wherein the first identification feature (L) comprises one of the useful signals of the first data stream or the useful signal of the first data stream. The method of claim 13, wherein the output of the first sub-section is suppressed in a fault-free presence of the identification feature in the second data stream. Method according to one of claims 1 to 14, wherein the step of outputting the first and / or second data stream comprises the step of converting the first and / or second data stream into a speech signal or an audio signal and / or a video signal. Method according to one of Claims 1 to 15, in which the second data stream (N, P) comprises at least one of the following useful signals (J, R): A video signal; - images or image data; - data for a computer-assisted slide show; - audio data or audio signals; - time-based text, scrolling text. Method according to one of claims 1 to 16, wherein a transmission of the first and / or second data stream takes place at least partially by transmitting signals according to a mobile communication standard. Method according to one of claims 1 to 17, wherein a transmission of the first and / or second data stream at least partially by an Internet connection ( 22 ) between at least two nodes. Method according to one of claims 1 to 18, wherein a transmission of the first and / or second data stream (B, N) at least partially by a wired communication link ( 22 ) he follows. Method according to one of claims 1 to 19, wherein a transmission of the first and / or second data stream (B, N) at least partially via a circuit-switched communication link (B, N). 23 ) he follows. A method of transmitting a first data stream having at least two consecutive subsections and a second data stream comprising the steps of: - Detecting a first useful signal (A) and a second useful signal (J, M); - Determining at least one identification feature (L) from the first useful signal (A), which allows identification of the first useful signal; - assigning the at least one identification feature (L) to the second useful signal (J, M); - Generating the first data stream (B) from the first useful signal and generating the second data stream (N) from the second useful signal (M, J) and the at least one identification feature (L). A method according to claim 21, characterized in that the first data stream (B) and the second data stream (N) are respectively generated with a plurality or plurality of successive subsections, wherein a plurality or all subsections of the second data stream to a respective associated subsection of the first Data stream referencing, contained in the second data stream identifier is assigned. The method of claim 21 or 22, wherein the step Assigning the steps includes: - Providing a time stamp (K) with a unique time information; - Assigning the time stamp (K) to the identification feature (L) and a portion of the second useful signal (J, M). A method according to any one of claims 21 to 23, wherein the Step of detecting the first useful signal (A) and the second useful signal (J) comprises at least one of the following steps: - detecting a speech signal; - detecting an audio signal; - detecting a video signal; - encoding an audio signal, a speech signal, or a video signal; - capturing a computerized slide show; - capture current text; - decomposing the detected useful signal into sections. Method according to one of claims 21 to 24, wherein the step of determining at least one identification feature comprises at least one of the following steps: - generating a coded signal from the detected first useful signal; Generating coefficients from the detected first useful signal or from subsections of the first useful signal based on a long term prediction; Generating coefficients from the detected first payload signal or from subsections of the first payload signal based on a linear predictive coding; Generating frequency information from a content of the first payload signal or from a content of partial portions of the first payload signal; Generating HASH sequences from the first payload signal or from subsections of the first payload signal; - Generating checksums from the first useful signal or from subsections of the first useful signal. Method for transmitting a first and a second data stream from a transmitter to a receiver, comprising the steps: - establishing a first communication connection between transmitters ( 1 ) and receiver ( 3 ) for transmitting the first data stream; - transmitting the first data stream (B) over the first communication link to the receiver ( 3 ); - Generating a second data stream (N) with a useful signal (M), wherein the second data stream (N) from the first data stream derived identification features (L), which are associated with the useful signal (M) of the second data stream (N); - transmitting the second data stream via a second communication link to the receiver; Receiving the first and receiving the second data stream (E, P); - determining the identification features in the second data stream; - Timed mutually outputting the first data stream and the useful signals in dependence of the determined identification features. A method according to claim 26, characterized in that the first data stream (B) and the second data stream (N) are each transmitted with a plurality or plurality of successive subsections, wherein several or all subsections of the second data stream to a respective associated subsection of the first Data stream referencing, contained in the second data stream identifier is assigned. The method of claim 26 or 27, wherein the step of transmitting the first data stream comprises the step of: - detecting a first useful signal (A); - decomposing the first useful signal into sections and - Determining at least one first identification feature (L) for identifying a first subsection of the first useful signal (A); - Generating the first data stream from the subsections. The method of claim 28, wherein the step of generating of the first data stream (B) or the step of detecting a first useful signal (A) comprises the step of coding according to at least one of the following methods: - GSM EFR; - AMR; - AMR-WB; - G.729; G.72x according to the ITU-T standard; - μ-law The method of any one of claims 26 to 29, wherein the step of generating a second data stream comprises the steps of: - detecting a second useful signal (J) having a plurality of subsections; - Generating a second data stream (N) from the second useful signal (J, M) and derived from the first data stream identification features (L), each subsection of the second data stream is assigned to one of the first data stream derived identification features for later timed output. The method of any one of claims 26 to 30, wherein the step of timing in synchronism comprises the steps of: - determining identification features of the first data stream; - comparing the identified identification features of the first data stream with the identification features of the second data stream; - Outputting the first data stream and the payload signals of the second data stream assigned to the identification features as a function of the comparison. The method of any one of claims 26 to 31, wherein the step of receiving the first data stream comprises the step of buffering the first data stream. The method of any one of claims 26 to 32, wherein in the steps of transmitting, the first communication connection and the second communication connection at least partially comprise a shared signal path. The method of any one of claims 26 to 33, wherein the step of establishing a communication link further comprises the step of transmitting a third data stream for communication between the transmitter and receiver prior to transmitting the first data stream. Method according to one of Claims 26 to 34, in which the first data stream comprises at least one of the following useful signals: - analog audio data or audio signals; - μ-law encoded audio data or audio signals - GSM EFR encoded audio data or audio signals; AMR or AMR-WB encoded audio data or audio signals; G.729 encoded audio data or audio signals; - encoded audio data or audio signals according to one of the G.72x methods of the ITU-T standard for voice compression; - Fax data. The method of claim 35, wherein the identification feature in the second data stream comprises the useful signal of the first data stream. The method of claim 36, wherein the output of the first data stream is suppressed in the presence of the identification features in the second data stream. The method of any of claims 26 to 37, wherein the step of outputting the first and / or second data stream comprises the step of converting the first and / or second data stream into audio signals and / or video signals. Method according to one of Claims 26 to 38, in which the second data stream comprises at least one of the following useful signals: - video data; - image data; - data for a computer-assisted slide show; - audio data or audio signals. Method according to one of claims 26 to 39, wherein a transmission of the first and / or second data stream at least partially by A transmission of signals according to a mobile communication standard; - an internet connection between at least two nodes or - A wired communication connection takes place. Transmission device comprising: - a first device ( 151 ) for detecting a first useful signal; A second device ( 15 . 151 ) for detecting a second useful signal; An analysis device ( 12 . 12a . 110 ), which is designed to determine and provide at least one identification feature (L) for identifying the first or the respective subsection from a first subsection or subsections of the first useful signal; A coding device ( 11 . 11a . 110 ) carried out by the analysis device ( 12 . 12a ) to assign provided identification feature or a respective identification feature a first subsection or a respective subsection of the second useful signal; A first transmission stage ( 13 ) for transmitting the first useful signal; - A second transmission stage for transmitting the second useful signal. Transmission arrangement according to Claim 41, in which the first and second devices ( 131 . 15 . 151 ) for substantially simultaneously detecting the first and second payloads. Transmission arrangement according to one of Claims 41 to 42, in which the analysis device ( 12 . 12a ) for generating identification features, which has at least one of the following features of the first subsection of the first useful signal: Long term prediction coefficients; - coefficients derived from a linear predictive coding; A frequency information of a content of the first subsection; A HASH sequence; - a checksum of the first subsection. Transmission arrangement according to one of Claims 41 to 43, in which the coding device ( 11a ) a synchronization device ( 17 ) for an assignment of the provided identification feature to a subsection of the second useful signal, which has a predetermined temporal relationship to the first subsection of the first useful signal. Transmission arrangement according to one of Claims 41 to 44, in which the first transmission stage for transmission on a first transmission channel ( 23 ) and the second transmission stage for transmission on a second transmission channel ( 22 ) is executed. Transmission arrangement according to claim 45, in which the first and the second transmission stage form the same transmission stage and / or the first transmission channel and the second transmission channel are the same transmission channel. Transmission arrangement according to one of Claims 41 to 46, in which the first device ( 13 . 131 ) for capturing audio data and / or audio signals. Transmission arrangement according to one of Claims 41 to 47, in which the first device ( 13 . 131 ) is carried out for coding the first useful signal with at least one of the following coding types: a coding according to μ-law of a speech or audio signal; A coding according to GSM EFR of a voice or audio signal; A coding according to AMR of a voice or audio signal; A coding according to PCM of a voice or audio signal; A coding according to AMR or AMR-WB of a voice or audio signal; A coding according to G.729 of a voice or audio signal; A G.72x coding according to the ITU-T standard for voice compression. Transmission arrangement according to one of Claims 41 to 48, in which the second device ( 15 ) for detecting at least one of the following signals: - audio signals; - video signals; - data-based presentations; - Images; - text. Transmission arrangement according to one of Claims 41 to 49, in which the second device ( 15 ) or the coding device ( 11 . 11a ) for coding the first useful signal with at least one of the following coding types: H.26x coding of video signals; - MPEG or MPEG2 encoding of video signals. Receiving arrangement for the synchronization of first and second payload signals, wherein in the second payload identification features are stored, which are derived from sections of the first payload, the receiving arrangement comprising: - a device ( 31 . 32 ) for the controlled delivery of the first useful signal; An analysis device which is designed to provide an identification feature for identifying a first subsection from the received first useful signal; - a synchronizer ( 34 . 34a ) for synchronizing an output of the first and second useful signal on the basis of a comparison of the identification features stored in the second useful signal with those of the analysis device ( 36 . 36a ) provided identification features. A receive device according to claim 51, further comprising: - A second latch for storing the second useful signal. Receiving arrangement according to one of Claims 51 to 52, in which the first device ( 32 ) is executed to a buffering of the first useful signal. Receiving arrangement according to one of claims 51 to 53, wherein the first device is designed for a time-scaled change of the first useful signal. A receiving arrangement according to any one of claims 51 to 54, wherein the synchronizer comprises a comparison device configured to deliver a control signal to a synchronization device in response to a comparison of the identification features stored in the second payload with the identification features provided by the analyzer, the synchronization means with the first device coupled to control the output of the first useful signal. A receive device according to claim 55, wherein the comparison device is adapted to analyze a similarity or a correlation between the identification features stored in the second payload signal and the identification features provided. Receiving arrangement according to one of claims 51 to 56, further comprising: - a first decoding device ( 31 ), which is connected on the output side to the first intermediate memory and is designed to generate the first useful signal from a first signal received by the receiving device; A second decoding device ( 33 ), which is designed to generate the second useful signal from a second signal received by the receiving device. A receive arrangement according to claim 57, wherein the receiving arrangement is adapted to receive the first signal and the second signal in a common transmission channel. Receiving arrangement according to Claim 57, in which the second decoding device is designed to remove the identification features stored in the second useful signal. Receiving arrangement according to one of claims 51 to 59, wherein the first subsection of the first useful signal comprises at least one of the following signals: An analog audio signal; A GSM EFR encoded audio signal; An AMR encoded audio signal; A PCM encoded audio signal; An AMR or AMR-WB encoded audio signal; A G.729 coded audio signal An encoded audio signal according to one of the G.72x methods of the ITU-T standard for voice compression; - μ-law coded audio signal. Receiving arrangement according to one of Claims 51 to 60, in which the analysis device is designed to provide the identification feature which has one of the following features of the first subsection of the first useful signal: - Long term prediction coefficients; - coefficients derived from a linear predictive coding; A frequency information of a content of the first subsection; A HASH sequence; - a checksum of the first subsection. System comprising a transmitting arrangement and a receiving arrangement, - In which the transmission arrangement is designed to deliver a first useful signal and a second useful signal, wherein the second useful signal comprises identification features derived from the first useful signal; - Wherein the transmitting device is designed for transmitting the first and second useful signal to the receiving device; - Wherein the receiving arrangement is designed to determine identification features from the first useful signal; - In which the receiving arrangement is formed to a synchronous output derived from the first and second useful signal signals by comparing the determined identification features with the identification features contained in the second useful signal. A system according to claim 62, wherein the transmission of the first and second wanted signals is at least partially via a common transmission channel. A system according to any one of claims 62 to 63, wherein the transmission of one of the first and second payload signals is at least in part via a fixed network connection ( 23 ) or an internet connection ( 22 ) between at least two communication devices. A system according to any one of claims 62 to 64, wherein the receiving arrangement is adapted for time synchronous output of an audio signal and a video signal. Method for synchronizing a plurality of data streams, each comprising a plurality of successive subsections, wherein a first data stream (B; E) and a second data stream (N; P) are received, of which the second data stream in association with subsections of the second data stream to subsections contains identification identifiers (U; LPC, LSP; par_0 *, par_1 *) referring to the first data stream (B; E), and subparts of the data streams assigned to one another are identified via the identification features in order to synchronize the data streams. A method according to claim 66, characterized in that for the assignment of a respective subsection of the first data stream (B; E) to a respective subsection of the second data stream (N; P) several, to different sections of the first data stream referring identification features are taken into account. Method according to Claim 66 or 67, characterized in that, for a respective subsection of the first data stream (B; E), a set of several identification features (U; LPC, LSP; par_0 *, par_1 *) referring to this subsection are included in the second data stream (N ; P) and is taken into account in the determination of mutually associated subsections of the data streams. Method according to one of claims 66 to 68, characterized in that in each case at least one identification feature is determined for subsections of the first data stream (B; E) and with identification features (LPC, LSP; par_0 *, par_1 *) or identification feature sets that are contained in the second data stream is compared. A method according to claim 69, characterized in that the comparison is performed by applying at least one comparison criterion, preferably by determining at least one comparison measure from the comparison criterion or at least one distance measure (e (i, k)) from the distance criterion in an abstract feature space containing the identification features , Method according to Claim 70, characterized in that the identification feature (U) or identification feature record (LPC, LSP; par_0, par_1.) Determined by multiple application of the comparison criterion (DM) between, on the one hand, an identification point subsection of the first data stream (B; On the other hand, a plurality of identification features or identification feature sets (U, LPC, LSP, par_0 *, par_1 *) which are assigned to different subsections of the first data stream and which are transmitted to the second data stream (N; P) are assigned to the identification feature (U or identification feature set (LPC, LSP, par_0, par_1) of the point-of-arrival subsection and, on the other hand, to a particular identification feature set (s) contained in the second data flow (U; LPC, LSP; par_0 *, par_1 *) Comparative measures (e (i, k)) are determined, each of which is the Aufpunkt subsection and a certain Teilabschni tt of the second data stream are assigned and / or a subsection distance (k) between the Aufpunkt- section and the particular subsection of the second data stream. A method according to claim 70 or 71, characterized in that by multiple application of the comparison criterion (DM) between on the one hand in association with a Aufpunkt subsection of the second data stream (N; P) contained in this identification feature (U) or identification feature set (LPC , LSP; par_0 *, par_1 *) and, on the other hand, a plurality of identification features (U) or identification feature sets (LPC, LSP; par_0, par_1) determined for different subsections of the first data stream (B; Par_0 *, par_1 *) and, on the other hand, to an identification feature (U) or identification feature record (LPC, LSP, par_0.) Determined / determined for a particular subsection of the first data stream , par_1) (e (i, k)), each of which is the sub-point subsection and a specific n subsection of the first data stream are assigned and / or a subsection distance (k) between the Aufpunkt subsection and the particular subsection of the first data stream. A method according to claim 71 or 72, characterized in that on the basis of the determined comparison measures (e (i, k)) associated with each other subsections of the first and the second data stream are determined. Method according to one of claims 71 to 73, characterized in that a synchronization measure is determined on the basis of the determined comparison measures (e (i, k)), which at least one subsection or subsections of the first data flow (B; Partial segments of the second data stream (N; P) are assigned in pairs. Method according to one of Claims 71 to 74, characterized in that a plurality of sets of comparison measures (e (i, k)) are determined in each case for several different constituent point sections of the first or second data stream by repeated application of the comparison criterion (DM). Method according to claim 75, characterized in that comparison amounts (e (i, k)) of the sets of comparative measures assigned to a same subsection distance (k) are combined with each other to a respective combination comparison measure (E (k)). A method according to claim 76, characterized in that based on the determined combination comparison measures (E (k)) associated with one another subsections of the first and the second data stream are determined. Method according to claim 76 or 77, characterized in that on the basis of the determined combination comparison measures (E (k)) a synchronization measure is determined which comprises at least one subsection or subsections of the first data stream (B; E) of at least one subsection or subsections of the second data stream (N; P) in pairs. Method according to one of claims 76 to 78, characterized in that the respective combination comparison measure (E (k)) of a scaled sum or a scaled product or another mathematical combination of the same subsection distance associated comparison measures (e (i, k )), which increases at least monotonically with a respective incoming comparative measure or at least monotonically decreases. A method according to any one of claims 76 to 79, characterized in that the combination comparison measures (E (k)) are iteratively successively obtained from an initial set of combination comparison measures or a set of comparison measures (e (i, k)) Taking into account further subsections of the first data stream (B; E) and / or the second data stream (N; P) or by taking into account further sets of comparative measures (e (i, k)). Method according to one of Claims 71 to 80, characterized in that a respective set of comparative measures (e (i, k)) is supplemented by further comparison measures (e (i, k)) by taking into account further subsections of the first or second data stream. Method according to one of claims 66 to 81, characterized in that media data streams are synchronized. Method according to claim 82, characterized in that at least one first data stream (B; E) comprising an audio data stream or speech data stream (B; E) with at least one second data stream (N; P) comprising at least one further audio data stream or voice data stream and / or comprising at least one video data stream (B; E) is synchronized. The method of claim 83, wherein the first data stream (B; E) is a codec coded data stream, wherein parameters output from a codec based coder (LPC, LSP, P, G, aG) or data are included in the first data stream to enable reconstruction of the first data stream of underlying audio or speech information by decoding the first data stream, and at least partially (LPC, LSP) recorded in the second data stream (N; P) to synchronize the first with the second data stream to enable Method according to Claim 84, characterized in that the codec is a codec based on a prediction analysis, preferably a linear prediction analysis, on the underlying audio or speech information. A method according to claim 85, characterized in that the codec is a speech codec and is based on a language model. Method according to one of Claims 83 to 86, characterized in that the second data stream (N; P) is a codec-coded data stream, wherein parameters or data output by a coding device based on a codec are recorded in the second data stream in order to be reconstructed to enable the second data stream underlying audio or voice information or video information by decoding the second data stream. Method according to claim 84 and claim 87, characterized in that the first (B; E) and the second (N; P) data stream are directly or directly synchronized by performing the synchronization with respect to these data streams themselves. A method according to claim 84 and claim 87, characterized in that the first (B; E) and the second (N; P) data stream are synchronized indirectly or indirectly by obtaining successive data streams (F) obtained by decoding or recoding the first and second data streams , W) are synchronized. Method according to one of Claims 66 to 89, characterized in that the first data stream (B; E) and the second data stream (N; P) are determined by a first entity (B; 1 ) and to a second instance ( 3 ), and that the first and the second data stream are transmitted by the second instance ( 3 ) or an intermediate instance mediating the transmission between the first and the second instance, wherein in the case of the intermediate entity this combines the first and the second data stream into a combination data stream and transmits it to the second entity, preferably instead of the first and the second data stream. A method according to claim 90, characterized in that in a communication environment at least a first user terminal the or a first instance ( 1 ) and at least one second user terminal the or a second instance ( 2 ). A method according to claim 91, characterized in that at least one infrastructure of the communication environment forms the or an intermediate instance. Method according to claim 91 or 92, characterized in that in the case of a bidirectional or multidirectional communication, the first user terminal at the same time also has a second instance (with respect to at least one further first data stream and at least one further second data stream). 3 ) and the second user terminal in relation to the further first data stream and the further second data stream at the same time also a first instance ( 1 ). Method according to one of claims 90 to 93, characterized in that the first and the second data stream or the further first and the further second data stream via different transmission channels ( 23 . 22 ) and / or transferred between the entities on the basis of various transmission standards. A method according to claim 94, characterized in that one of the first and second data stream or one of the further first or second data stream via a circuit or circuit-switched Connection ( 23 . 24 ), preferably at least partially by radio according to a mobile telephony standard, and the other from the first and second data stream or from the further first or second data stream via a packet-switched connection ( 22 ), preferably according to a network or Internet transmission standard. Method according to Claim 95, characterized in that the first or the further first data stream (B; E) is transmitted via the circuit or circuit-switched connection (B; 23 . 24 ) and the second or the further second data stream (N; P) via the packet-switched connection ( 22 ) is transmitted. Method according to one of Claims 94 to 96, characterized in that one of the first and second data stream or one of the further first or second data stream with respect to the other is delayed by the first and second data stream or by the further first or second data stream during the transfer. Method according to one of Claims 90 to 97, characterized in that first the generation and transmission of the first or further first data stream (B; E) starts and later then the generation and transmission of the second or further second data stream (N; P) and then use the synchronization of the data streams. Method according to one of claims 66 to 98, carried out in connection with the implementation of the method according to at least one of claims 1 to 40 or as a partial method of the method according to at least one of claims 1 to 40. Method according to one of claims 66 to 99, comprising further method features resulting from at least one of claims 1 to 40. System for carrying out the method according to one of Claims 66 to 100, having at least one first system component (I) which is designed to generate the first data stream (B; E) and the second data stream (N; P) and to a second one System component ( 3 ), wherein the second system component or at least one intermediate component of the system conveying the transmission between the first and the second system components is designed to synchronize the first and the second data stream. A system according to claim 101, characterized in that the system is a communication system, or comprises a mobile telecommunications system, wherein at least a first user terminal or the first system component ( 1 ) and at least one second user terminal the or a second system component ( 3 ). System according to claim 102, characterized in that at least one infrastructure device of the communication system forms the or an intermediate component. System according to claim 102 or 103, characterized in that it is designed for a bidirectional or multidirectional communication, wherein the first user terminal in relation to at least one further first data stream and at least one further second data stream at the same time also a second system component ( 3 ) and the second user terminal in relation to the further first data stream and the further second data stream at the same time also a first system component ( 1 ). A system according to any one of claims 101 to 104, comprising further device features resulting from at least one of claims 41 to 65. A user terminal for a system according to any one of claims 101 to 105, wherein the user terminal ( 1 in carrying out the method according to any one of claims 66 to 100, generating the first and the second data stream (N; P) and transmitting the data streams. A user terminal for a system according to any one of claims 101 to 105, wherein the user terminal ( 1 ) is designed to receive and synchronize the first (B; E) and second (N; P) data stream or the further first and further second data stream when carrying out the method according to one of Claims 66 to 100. A user terminal according to claim 106, comprising a transmission arrangement according to any one of claims 41 to 50. A user terminal according to claim 107, comprising a receiving arrangement according to any one of claims 51 to 61. A system according to any of claims 101 to 105, comprising a plurality of user terminals according to claim 106 or 107 as a first system component ( 1 ) and / or as a second system component ( 3 ).

Images