WO2007065458A1

WO2007065458A1 - Method, circuit arrangement and terminal for transmitting voice and/or video data via ip-based networks

Info

Publication number: WO2007065458A1
Application number: PCT/EP2005/013118
Authority: WO
Inventors: Anton Kruk
Original assignee: Siemens Home And Office Communication Devices Gmbh & Co. Kg
Priority date: 2005-12-07
Filing date: 2005-12-07
Publication date: 2007-06-14

Abstract

The invention proposes a method, a circuit arrangement and a terminal for transmitting voice and/or video data via IP-based networks (IPN) which easily improve the quality of the real-time transmission of voice and/or video data via IP-based networks (IPN). The method involves a current data packet containing current compressed useful data (e.g. p(k+3)) being respectively transmitted by appending to the current compressed useful data (e.g. p(k+3)) the compressed useful data (e.g. p(k+2); p(k+1)) from a prescribed number of data packets which have been transmitted directly beforehand. The circuit arrangement and the terminal have appropriate means for carrying out the aforementioned method.

Description

description

Method, circuit arrangement and terminal for the transmission of voice and / or video data over IP-based networks

The invention relates to a method for the transmission of voice and / or video data via IP-based networks according to the preamble of claim 1. The invention further relates to a circuit arrangement and a terminal for performing the aforementioned method.

When real-time data (e.g. voice data and / or video data) is transmitted in so-called data packets over extensive IP-based networks, also known as the Internet, the voice and / and / or packet losses are impaired due to the sometimes large delays and packet losses. or video quality.

In order to counteract this loss of quality, it is known to use coding methods based on so-called robust codes. The disadvantage is that additional hardware and firmware are necessary.

Large jitter buffers can also be used for intermediate storage, which have the disadvantage that the time delay in the transmission of the data packets increases. This is contrary to the specifications for real-time transmission. In real-time transmission of user data, delays in the transmission of user data should be avoided as far as possible.

Data packets can also be sent repeatedly in order to ensure that if a data packet fails or is delayed, the useful data transported by such data packets ultimately still reaches the desired recipient. This is possible because the individual data packets belonging to an overall transmission between a transmitter and receiver are not necessarily all transmitted over the same route in the IP-based networks. Rather, the individual data packets are transmitted in a wide variety of ways, depending on which section load the individual data packets currently find. Depending on this, the individual data packets are then transmitted further over this or that section of the IP-based networks. It can happen that data packets are significantly delayed or noisy in the user data portion or are lost altogether. User data of such data packets can then contribute nothing to the reconstruction at the receiver of the causal information contained in the data, so that the quality of the entire transmission is ultimately reduced.

It therefore makes sense to send data packets repeatedly because this increases the chance, at least in principle, of bringing all the user data required for a correct reconstruction of the information contained in the transmitted user data to the recipient in good time. If they do not reach the recipient on one way or do not reach the recipient in sufficient quality, they reach the recipient on the other route in sufficient quality.

On the other hand, the repeated sending of data packets produces an increased volume of data on the IP-based networks, so that this also increases the likelihood that data packets will get stuck, be too delayed or noisy, or even be lost.

All in all, the ratio of net to gross data, which is important for the quality of the transmission, the ratio of the data, which on the one hand arrives at the recipient as user data and on the other hand is used to ensure that the user data can reach the recipient at all, is unsatisfactory. So that a data packet with user data can get from a sender to a receiver via IP-based networks, the data packet has, in addition to the user data part, a control information part, in which the information required by the system IP-based networks is summarized to be able to transmit the relevant data packet from the sender to the receiver. The control information component is very large in relation to the user data component, which means that control information is transmitted and less user data. This ratio also deteriorates to the disadvantage of the useful data in that only a particularly small proportion of useful data is transmitted in each case for real-time transmission, in particular in connection with voice and / or video data, so that this data, real in real time, is technically practical Real time, are brought as directly as possible to the recipient, who then, as said, restores or reconstructs them back in real time.

From document DE 101 33 518 A1 it is known, on the one hand, for the wireless transmission of voice data between at least two terminals, to combine the voice data in data packets in predefined data frames in a redundant manner and to transmit them at changing intervals at different intervals. The redundancy of the individual voice data is related to the selected frequency for the duration of a selected frequency, even if the frequency changes over time. If the frequency changes, then what has just been said applies only to this frequency.

On the other hand, it is known from document DE 101 33 518 A1 to divide the redundancy itself, to be precise over several different frequencies.

The aforementioned proposals are therefore based on the principle of introducing something explicitly, namely different frequencies, in order thereby to improve the transmission quality of voice data to be transmitted. The proposals focus on the transmission of voice data over air interfaces. It is an object of the present invention, based on a method of the type mentioned at the outset, to improve this method in such a way that the quality of the transmission of voice and / or video data over IP-based networks is improved in a simple manner. Furthermore, it is an object to specify a circuit arrangement and a terminal for carrying out the method according to the invention.

With regard to the method, this object is achieved according to the invention by such a method which has the feature specified in the characterizing part of claim 1.

Thereafter, each time a current data packet with current compressed user data is sent, the compressed current user data is appended with the compressed user data of a predetermined number of data packets sent immediately before.

In the method according to the invention, a total of somewhat longer data packets are transmitted than before, because not only current user data are transmitted with a data packet, but also user data from previous transmissions. But the scope of the total amount of data used for the transmission is limited, because for the majority of the useful data within a data packet, only the amount of control information that has to be used for a single data packet has to be used. Therefore, at least theoretically, there is no longer any control information transmission on the IP-based networks.

The additional data load of the IP-based networks remains limited to the additional share of user data is added to the current user data for each data packet. However, this additional share of useful data is low because it only takes short data sequences into account in order to meet the requirement for real-time transmission.

The method according to the invention exploits, inter alia, the peculiarity of IP-based networks in order to improve the quality of the transmission in real time between two terminals connected to it, which exchange voice and / or video data. This peculiarity is that successively transmitted data packets do not arrive at the same destination via the same route within the IP-based networks. The control mechanisms within IP-based networks treat data packets individually, at least from the first approach.

Due to this fact and due to the fact that useful data from previous transmissions are also contained within a data packet, the recipient can also reconstruct all of the desired useful information from the received useful data, in real time if one or the other Data packet delayed too much, its user data too noisy or the entire data packet has been lost.

•

A particular advantage of this method is that this method does not require any intervention in the IP-based networks. All measures for coding and decoding the corresponding data packets to be transmitted can be taken in the end devices. An intervention in the IP-based networks is not necessary. With this end-to-end approach, the method works regardless of the cause of the interference in the transmission, in particular when radio links such as Wireless Lan (WIAN) are interposed. The enlargement of the data packets to be transmitted according to the invention can be kept small by using highly compressing codes. A so-called jitter buffer for receiving the data packets is currently z. B. already used in Voice over IP transmissions and only needs to be slightly adjusted in size. Advantageous embodiments of the invention are the subject of dependent claims.

The specified number of user data that are attached from previous transmissions can then be determined statically or dynamically. The latter enables dynamic adaptation if a time-dependent deterioration in the transmission quality is detected on the basis of checked parameters. If the dynamic setting is made solely on the basis of reception parameters for the transmission direction, then no additional packets have to be transmitted for signaling.

For example, the so-called “Real-Time Transport Protocol ^w (RTP) or the“ Real-Time Control Protocol ”(RTCP) are used as the transmission protocol. The latter in particular offers possibilities for the dynamic setting of the predetermined number beforehand As explained above in connection with the method according to the invention, the invention is also achieved by a circuit arrangement according to claim 9 and by a terminal device according to claim 10. The circuit arrangement according to the invention and the terminal device according to the invention accordingly have means by which the circuit arrangement and the terminal are able to carry out the method according to the invention. An exemplary embodiment of the invention is explained in more detail below with reference to a drawing. 1 shows a basic illustration of a communication transmission link for voice and / or video data between a transmitter and a receiver via IP-based networks according to the prior art, and

Figure 2 is a schematic representation of an inventive

Method for the transmission of voice and / or video data via a communication transmission link according to FIG. 1. In FIG. 1, a transmitter S _{f is} a system of communication transmission links within IP-based networks IPN, which is represented in a stylized form, and which in principle is also a single one IP-based network can be, and a receiver E shown.

The transmitter sends voice and / or video data in real time to the receiver E via the IP-based network or networks IPN, which reconstructs the originally transmitted real-time information from the information received.

For this purpose, the transmitter S feeds the IP-based network (s) IPN with its currently running voice and / or video data (arrow) to be delivered to the receiver E in the form of user data current for respective small periods of time, which are packaged in data packets along with the actual user data additional control information for the notification of the destination and the nature of the user data contained.

The IP-based network or networks transmit the data packets transmitted by the transmitter S on the basis of those specifying the destination

Control information in the data packets to the recipient E. There is no fixed route for the individual data packets within the or the IP-based networks. The way it is determined for the individual data packets which further route section can be used in a favorable manner at the time of a data packet arriving at further route sections and is therefore assigned.

The receiver E accepts the incoming data packets (arrow) and reconstructs the current current user information in real time from the user data contained in the data packets and on the basis of the information about the type of user data also contained in the control information.

FIG. 2 shows the schematic sequence of real-time transmission of voice and / or video data. The transmission side is shown. On the receiver side, the receiver E filters out the individual user data units to the extent that it requires them for the reconstruction of the current user information and assembles them into the desired overall signal. It is clear that there is actually a time difference between the transmitted and reconstructed total signal, which is generally greater than the length of a received data packet. However, this time difference is so small that we can still speak of real-time transmission for the human eye and ear. On the transmitter side, the transmitter S packs the useful information to be transmitted into data packets, each with a relatively small proportion of useful data. This is necessary so that there are no great delays until a useful content is transmitted. If this time delay is too great, real-time transmission is no longer guaranteed.

Two columns can be seen in FIG. The left of the two columns shows how the transmitter S generates uncompressed data packets p (k), p (k + 1), p (k + 2), p (k + 3) etc. in the course of time t which each contain a successive part of the useful information to be transmitted. The respective useful information component is then compressed in each case, preferably using highly compressing codes. The right of the two columns shows data packets with compressed user data, identified in the figure by boxes that are shortened in length. After compression, the compressed user data are sent to the receiver E via the IP-based network or networks IPN. The transmission of compressed user data starts at the earliest after the comple ^¬ compression averaging the data to be transmitted to a part of each concerned. This is indicated in FIG. 2 by the fact that the data packets with the compressed useful data are located one time step lower than the respectively associated recompressed useful data.

For the transmission of compressed user data, as shown in FIG. 2, not only the current user data, e.g. p (k), p (k + 1), p (k + 2) etc., but also, if necessary, also the compressed useful data from a predetermined number of previous transmissions.

In the example according to FIG. 2, the first data packet transmitted with the compressed useful data p (k) has no further compressed useful data, since at this point in time it is not yet available in relation to a current transmission of voice and / or video data.

At the next point in time, the compressed useful data p (k + l) and p (k) are then combined and transmitted together in the same data packet. This is / and this also applies in a corresponding manner to the following cases, because the total duration of the data packet sent is still shorter than the current data packet generated in parallel with the associated uncompressed useful data due to the compression. In the present exemplary embodiment, the compressed useful data p (k + 2) and p (k + l) are then combined. The compressed useful data p (k + 3), p (k + 2) and p (k + l) are then combined. Then the compressed ones User data p (k + 4), p (k + 3) and p (k + 2) combined to form a data packet to be sent. In the next case, the compressed user data p (k + 5) and p (k + 4) are combined. Then the compressed user data p (k + 6) _/ p (k + 5), p (k + 4) and p (k + 3) are combined. Finally, FIG. 2 also shows how the compressed useful data p (k + 7) and p (k + 6) are combined in a data packet to be sent.

In the present exemplary embodiment, the dynamic selection of a predetermined number of compressed useful data to be combined is selected for the combination of compressed user data to be sent. It is also possible _/ each to consider a statically predefined number here.

The dynamic specification of the number of compressed useful data to be summarized can be based on feedback that comes from the receiver and on the basis of which the transmitter is provided with information about the quality with which the transmission of events takes place. If the quality is rather low, it is advantageous if a larger number of compressed useful data is transmitted from the transmitter side. If the quality is sufficient or good, a smaller number of compressed useful data can be selected.

Overall, the method presented is based, among other things, on the fact that the individual data packets with the compressed useful data arrive at the receiver E in different ways, so that useful data that has been lost, noisy or delayed too long is still of sufficient quality to the recipient E arrive.

Claims

1. Method for the transmission of voice and / or video data via IP-based networks, to which at least two terminals are wired or wirelessly coupled and via which voice and / or video data exchanged between coupled terminals as data packets with a user data portion in which the User data are included _/ and a control information component, in which information for the control of the data packet from a sending terminal to a receiving terminal is contained, is transmitted, characterized in that in each case for sending a current data packet with current compressed user data (e.g. p (k + 3)) the current compressed user data (eg p (k + 3)) the compressed user data (eg p (k + 2);

p (k + l)) are attached to a predetermined number of data packets sent immediately before.

2. The method according to claim 1, characterized in that it is statically set which predetermined number of data packets sent immediately before is used for attaching compressed useful data (e.g. p (k + 2); p (k + l)).

3. The method according to claim 1, characterized in that it is dynamically set which predetermined number of data packets sent immediately before is used for attaching compressed useful data (e.g. p (k + 2); p (k + l)).

4. The method according to claim 3, characterized in that the dynamic setting of the predetermined number of immediately transmitted data packets for attaching compressed useful data (e.g. p (k + 2);

p (k + l)) is controlled as a function of technical parameters which are determined during previous transmissions of data packets and continuously updated and transmitted to the sending terminal (S).

5. The method according to claim 3, characterized in that the setting is made solely on the basis of reception parameters for the transmission direction.

6. The method according to claim 4, characterized in that the transmission quality in connection with the previously sent data packets is used as a technical parameter.

7. The method according to any one of the preceding claims, characterized in that the number one and higher is used as the predetermined number of data packets sent immediately before after a start phase.

8. The method according to any one of the preceding claims, characterized in that the transmission of the data packets with a protocol known as "Real Time Transport Protocol" (RTP) and / or "Real Time Control Protocol" (RTCP) accomplished becomes.

9. Circuit arrangement with means for performing the method according to one of claims 1 to 7.

10. terminal, with means for wireless and / or wired coupling to IP-based networks (IPN), with means for transmitting voice and / or video data over the IP-based network (IPN) in the form of data packets and in real time, and with at least one single processor and program code means for performing the method according to one of claims 1 to 7.