Apparatus and Method for Transmitting a Sequence of Data Packets and Decoder and Apparatus for Decoding a Sequence of Data Packets

• US 20100284281A1
• Filed: 02/22/2008
• Published: 11/11/2010
• Est. Priority Date: 03/20/2007
• Status: Active Grant

• Current Assignee: Fraunhofer Gesellschaft Zur Foerderung Der Angewandten Forsching E.V.
• Original Assignee: Fraunhofer Gesellschaft Zur Foerderung Der Angewandten Forsching E.V.
• Inventors: Sperschneider, Ralph
• Granted Patent: US 8,385,366 B2
• Time in Patent Office: Days
• US Class Current: 370/242
• CPC Class Codes: H04L 1/0045 Arrangements at the receiver end
• CPC Class Codes: H04L 1/0084 Formats for payload data
• CPC Class Codes: H04L 2001/0097 Relays

Abstract

In the context of packet-oriented data transmission via a network, an apparatus for transmitting a sequence of data packets checks whether a packet is missing or faulty. In this case, a replacement packet is provided which is a valid packet with regard to a packet syntax which, however, has a predetermined contents characteristic. On the decoder side, a basic decoder recognizes the packet as a valid packet and decodes same, whereas an extension decoder can enable an error concealing measure on the basis of an indication in the replacement packet to the fact that this is a replacement packet, in order to provide a higher-quality audio reproduction.

Claims

1. An apparatus for transmitting a sequence of data packets representing an audio signal to a basic decoder or to an extension decoder, wherein the extension decoder comprises an extended functionality as compared to the basic decoder, comprising:

a receiver for receiving packets of the sequence, a packet syntax being specified for the packets;

a detector for detecting whether a packet of the sequence of packets is missing or faulty;

a provider for providing a replacement packet to replace the faulty packet or the missing packet, the replacement packet being a valid packet with regard to the packet syntax, audio contents of the replacement packet being predetermined audio contents reproducible by the basic decoder and comprising a predetermined contents characteristic and the replacement packet comprising an indication for the fact that it is a replacement packet, wherein the indication is implemented such that the indication is ignorable or interpretable by the basic decoder such that the replacement packet is decoded like a valid packet according to the predetermined contents characteristic, and that the indication is interpretable by the extension decoder to perform an error concealing measure which is generating contents for the replacement packet comprising a contents characteristic differing from the predetermined contents characteristic; and

an outputter for outputting the sequence of packets, the sequence of packets comprising at least one received packet and at least one replacement packet.

2. The apparatus of claim 1,

wherein a packet of the sequence receivable by the receiver for receiving comes from a transform-based audio encoder and comprises a short-time spectrum of a temporal sector of the audio signal, and

wherein the predetermined contents characteristic is that all the spectral values equal zero or together represent an energy smaller than two times the energy represented by a psycho-acoustic resting listening threshold.

3. The apparatus of claim 1, wherein the provider for providing comprises:

a memory for storing the replacement packet;

a memory fetcher for fetching the replacement packet from the memory whenever the detector for detecting detects a faulty packet or a missing packet.

4. The apparatus of claim 1, wherein the receiver for receiving is an input interface of a telephone base station and is connectable to a network implemented for packet-based data transmission.

5. The apparatus of claim 1, wherein the replacement packet comprises a payload region where the indication is located.

6. The apparatus of claim 1, wherein the packet syntax is implemented to define an extension payload field, and wherein the indication is formed by data in the extension payload field.

7. The apparatus of claim 6, wherein the data are generated in accordance with a letter code, the data comprising a meaning indicating at a data loss.

8. The apparatus of claim 7, wherein the data represent the expression “

FRAME_LOSS”

or “

data loss”

.

9. The apparatus of claims 7, wherein the letter code is the ASCII code.

10. The apparatus of claim 1, wherein the sequence of packets is generated by an audio encoder according to MPEG-1, layer 3, or MPEG-4 AAC, a bit reservoir function being disabled.

11. The apparatus of claim 1, which is implemented as a base station.

12. The apparatus of claim 1, wherein the packet syntax comprises a fill data field which is filled by a basic encoder when a minimum number of bytes provided for the frame is not needed for encoding a portion of the audio signal, and

wherein the indication is represented by a predetermined bit combination in the fill data field.

13. The apparatus of claim 1, wherein the detector for detecting is implemented to wait for a predetermined duration for a data packet comprising a sequence position indication and to signalize, when the predetermined duration has passed without detecting the data packet, a packet loss to the provider for providing.

14. The apparatus of claim 13, wherein the sequence position information sought is set by a sequence position indication of a preceding or subsequent valid packet.

15. The apparatus of claims 13, wherein the predetermined duration is settable and predeterminable by a QoS request.

16. The apparatus of claim 15, wherein the predetermined duration is smaller when the QoS request comprises a smaller delay and is greater when the QoS request allows a greater delay.

17. The apparatus of claim 1, wherein the outputter for outputting is implemented to output the sequence of packets as an unbroken sequence of packets and replacement packets representing a complete continuous sequence.

18. The apparatus of claim 1, wherein the receiver for receiving is implemented to receive packets but not, or at most partly, to decode the packets received in order to detect sequence position information in the packet.

19. A method for transmitting a sequence of data packets representing an audio signal to a basic decoder or to an extension decoder, wherein the extension decoder comprises an extended functionality as compared to the basic decoder comprising:

receiving packets of the sequence, a packet syntax being specified for the packets;

detecting whether a packet of the sequence of packets is missing or faulty;

providing a replacement packet to replace the faulty packet or missing packet, the replacement packet being a valid packet with regard to the packet syntax, audio contents of the replacement packet being predetermined audio contents reproducible by the basic decoder and comprising a predetermined contents characteristic and the replacement packet comprising an indication for the fact that it is a replacement packet, wherein the indication is implemented such that the indication is ignorable or interpretable by the basic decoder such that the replacement packet is decoded like a valid packet according to the predetermined contents characteristics, and that the indication is interpretable by the extension decoder to perform an error concealing measure which is generating contents for the replacement packet comprising a contents characteristic differing from the predetermined contents characteristic; and

outputting the sequence of packets, the sequence of packets comprising at least one received packet and at least one replacement packet.

20. An extension decoder for decoding a sequence of packets, wherein the extension decoder comprises an extended functionality as compared to the basic decoder, comprising:

a receiver for receiving the sequence of packets for which a packet syntax is specified, the sequence comprising at least one data packet and at least one replacement packet, the replacement packet being a valid packet with regard to the packet syntax, audio contents of the replacement packet being predetermined audio contents reproducible by the basic decoder and comprising a predetermined contents characteristic and the replacement packet comprising an indication for the fact that it is a replacement packet, wherein the indication is such that the replacement packet may be received by the basic decoder and that the replacement packet may be decoded like a valid packet according to the predetermined contents characteristics, and that the extension decoder is implemented to interpret the indication to perform an error concealing measure which is generating contents for the replacement packet comprising a contents characteristic differing from the predetermined contents characteristic;

a detector for detecting whether a packet comprises the indication and consequently is a replacement packet;

an error concealer for synthesizing synthesis audio contents for the replacement packet the characteristic of which differs from the predetermined contents characteristic, using the error concealer; and

an audio renderer for rendering the audio contents of a packet when the packet is not a replacement packet, and for rendering the synthesis audio contents when the packet is a replacement packet.

21. The extension decoder of claim 20, wherein the predetermined contents characteristic is a zero spectrum,

wherein the received packets of the sequence of packets have been generated by a transform-based audio encoder, and

wherein the audio renderer comprises a transform-based audio decoder (41, 42, 43).

22. The extension decoder of claim 21, wherein the error concealer is implemented to generate synthesis spectral values, and

wherein the audio renderer is implemented to convert the synthesis spectral values to a temporal representation.

23. The extension decoder of claim 20, wherein the error concealer is implemented to generate the synthesis audio contents by extrapolation of audio contents from a previous or subsequent intact or concealed audio packet, or

to generate the synthesis audio contents by interpolation of audio contents of a previous intact or concealed packet and a subsequent intact or concealed packet.

24. A method for decoding a sequence of packets in an extension decoder, wherein the extension decoder comprises an extended functionality as compared to the basic decoder, comprising:

receiving the sequence of packets for which a packet sequence is specified, the sequence comprising at least one data packet and at least one replacement packet, the replacement packet being a valid packet with regard to the packet syntax, the audio contents of the replacement packet being predetermined audio contents reproducible by the basic decoder and comprising a predetermined contents characteristic, and the replacement packet comprising an indication for the fact that it is a replacement packet, wherein the indication is such that the replacement packet may be received by the basic decoder and that the replacement packet can be decoded like a valid packet according to the predetermined contents characteristics, and that the extension decoder interprets the indication to perform an error concealing measure which is generating contents for the replacement packet comprising a contents characteristic differing from the predetermined contents characteristic;

detecting whether the packet comprises the indication and consequently is a replacement packet;

for the replacement packet, synthesizing synthesis audio contents the contents characteristic of which differs from the predetermined contents characteristic, by means of the error concealing measure; and

rendering the audio contents of a packet when the packet is not a replacement packet, and rendering the synthesis audio contents when the packet is a replacement packet.

25. A computer program comprising a program code for executing the method for transmitting a sequence of data packets representing an audio signal to a basic decoder or to an extension decoder, wherein the extension decoder comprises an extended functionality as compared to the basic decoder comprising:

receiving packets of the sequence, a packet syntax being specified for the packets;

detecting whether a packet of the sequence of packets is missing or faulty;

providing a replacement packet to replace the faulty packet or missing packet, the replacement packet being a valid packet with regard to the packet syntax, audio contents of the replacement packet being predetermined audio contents reproducible by the basic decoder and comprising a predetermined contents characteristic and the replacement packet comprising an indication for the fact that it is a replacement packet, wherein the indication is implemented such that the indication is ignorable or interpretable by the basic decoder such that the replacement packet is decoded like a valid packet according to the predetermined contents characteristics, and that the indication is interpretable by the extension decoder to perform an error concealing measure which is generating contents for the replacement packet comprising a contents characteristic differing from the predetermined contents characteristic; and

outputting the sequence of packets, the sequence of packets comprising at least one received packet and at least one replacement packet, when the method runs on a computer.

26. A computer program comprising a program code for executing the method for decoding a sequence of packets in an extension decoder, wherein the extension decoder comprises an extended functionality as compared to the basic decoder, comprising:

receiving the sequence of packets for which a packet sequence is specified, the sequence comprising at least one data packet and at least one replacement packet, the replacement packet being a valid packet with regard to the packet syntax, the audio contents of the replacement packet being predetermined audio contents reproducible by the basic decoder and comprising a predetermined contents characteristic, and the replacement packet comprising an indication for the fact that it is a replacement packet, wherein the indication is such that the replacement packet may be received by the basic decoder and that the replacement packet can be decoded like a valid packet according to the predetermined contents characteristics, and that the extension decoder interprets the indication to perform an error concealing measure which is generating contents for the replacement packet comprising a contents characteristic differing from the predetermined contents characteristic;

detecting whether the packet comprises the indication and consequently is a replacement packet;

for the replacement packet, synthesizing synthesis audio contents the contents characteristic of which differs from the predetermined contents characteristic, by means of the error concealing measure; and

rendering the audio contents of a packet when the packet is not a replacement packet, and rendering the synthesis audio contents when the packet is a replacement packet, when the method runs on a computer.