KR20130122552A - Method and apparatus for transmitting/receiving packet in a communication system - Google Patents

Abstract

The present invention relates to a packet configuration method in a communication system. The packet configuration method according to an embodiment of the present invention comprises a process for generating control data related to forward error correction (FEC), adding a header including a common header and an optional header to an application layer, processing the generated source payload, generating a data word block including information payload, and generating an FEC header for parity payload and FEC header data for source payload based on all or a part of the generated FEC control data; a process for generating FEC packets for source payload by combining the FEC header data for source payload with the source payload and generating FEC packets for parity payload by combining the common header data and the FEC header data for parity payload with the one or more generated parity payload; and a process for transmitting the generated FEC packets.

Description

METHOD AND APPARATUS FOR TRANSMITTING / RECEIVING PACKET IN A COMMUNICATION SYSTEM}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a communication system, and more particularly, to a method and apparatus for transmitting and receiving a packet in a communication system.

Data congestion is intensifying due to the diversification of various contents and the increase of high capacity contents such as high definition (HD) content and ultra high definition (UHD) content in a communication system. Due to such a situation, contents sent from a sender (for example, Host A) are not normally delivered to a receiver (for example, Host B), and a part of the contents is lost on a route.

In general, since data is transmitted in packet units, loss of content occurs in packet units. The packet is composed of one block (payload) of data to be transmitted, address information (for example, source address and destination address), and management information (for example, header). Therefore, when packet loss occurs in the network, the receiver cannot receive the lost packet, so that data and management information in the lost packet cannot be known. This causes user inconvenience in various forms such as deterioration of audio quality, deterioration of video quality of the video, broken screen, missing subtitles, and loss of files.

For this reason, application layer forward error correction (AL-FEC) is required as a method for recovering data loss generated in a network, and a method of constructing and transmitting FEC packets for this purpose is required.

The present invention provides an FEC packet for AL-FEC operation, which improves the reliability of a network by transmitting a packet packet generated by using at least one error correction code in a communication system supporting a packet-based communication scheme together with a data packet A method and apparatus for transmitting and receiving data are provided.

A transmission method according to an embodiment of the present invention is a method of transmitting a packet in a communication system, the method comprising: dividing a data stream into a data payload of a predetermined size; adding a common header to each divided data payload, A process of generating a source payload; Adding a first FEC Payload ID to the source payload and applying an FEC code to generate an FEC source packet for the source payload; Adding a second FEC payload ID to at least one parity payload and generating an FEC parity packet for the at least one parity payload by applying an FEC code; And transmitting the FEC source packet and the FEC parity packet.

A transmitting apparatus according to an embodiment of the present invention is a apparatus for transmitting a packet in a communication system, the apparatus comprising: dividing a data stream into a data payload of a predetermined size; adding a header to each divided data payload, (Source Payload); Adding a first FEC Payload ID to the source payload, generating an FEC source packet for the source payload by applying an FEC code, adding a second FEC Payload ID to the at least one parity payload, A control unit for generating an FEC parity packet for the at least one parity payload by applying the FEC parity packet; And a transmitter for transmitting the FEC source packet and the FEC parity packet.

A receiving method according to an embodiment of the present invention is a method for receiving a packet in a communication system, the method comprising: distinguishing whether a packet received from a transmitter is an FEC source packet or an FEC parity packet; Obtaining a source payload from the FEC source packet and obtaining a parity payload from the FEC parity packet, wherein the source payload divides the data stream into a data payload of a predetermined size, The FEC source packet includes a first FEC Payload ID added to the source payload and a FEC code applied thereto. The FEC parity packet is added to the parity payload with a second FEC Payload ID , And the FEC code is applied.

A receiving apparatus according to an embodiment of the present invention is a device for receiving a packet in a communication system, the apparatus comprising: a transmitter for separating an FEC source packet and an FEC parity packet from a packet received from a transmitter, And a control unit for obtaining a parity payload from the FEC parity packet, wherein the source payload divides the data stream into a data payload of a predetermined size, a header is added to each divided data payload, The FEC source packet is characterized in that a first FEC Payload ID is added to the source payload and an FEC code is applied thereto and a second FEC Payload ID is added to the parity payload and an FEC code is applied to the FEC parity packet .

According to an embodiment of the present invention, a higher quality service can be provided to the user. Further, the FEC packet configuration method according to the present invention makes it easy to distinguish FEC packets.

The present invention distinguishes each data stream from the control information different from the stream identification information or the FEC packet in the FEC packet, grasps the parity stream generated for FEC protection of each data stream, and performs FEC decoding Can be performed smoothly.

Figure 1 illustrates a network topology and a data flow;
2 is a block diagram of an MMT system according to an embodiment of the present invention;
3 is a diagram illustrating a structure of an MMT package according to an embodiment of the present invention;
4 is a diagram illustrating a configuration of configuration information included in an MMT package according to an embodiment of the present invention;
5 is a block diagram of an FEC packet format according to an embodiment of the present invention;
FIG. 6 is a block diagram of an FEC packet format when the Source Payload is an MMT Transport Packet according to an embodiment of the present invention; FIG.
FIG. 7 is a block diagram of an FEC packet format when Source Payload according to an embodiment of the present invention is an MMT Payload Format; FIG.
8 is a block diagram of a transmission apparatus according to an embodiment of the present invention;
9 is a block diagram of a receiving apparatus block according to an embodiment of the present invention;
FIG. 10 is an operation flow diagram for configuring an information block according to an embodiment of the present invention; FIG.
11 is an operation flow diagram of configuring an information word block according to another embodiment of the present invention;
FIG. 12 illustrates an example of a process of mapping an information symbol in an information word block when using an RS code; FIG.
13 is a diagram illustrating an example of a process of mapping an information symbol in an information word block when an LDPC code is used;
14 shows an RS frame configuration;
15 shows an LDPC frame configuration;
16 is a parity block mapping diagram for RS parity symbols;
17 is a parity block mapping diagram for LDPC parity symbols; And
18 is a structural view of an H matrix;

In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. The following terms are defined in consideration of the functions of the present invention, and may be changed according to the intentions or customs of the user, the operator, and the like. Therefore, the definition should be based on the contents throughout this specification.

First, terms used in the present invention are summarized as follows.

-FEC code: An error correction code for correcting an error or an erasure symbol.

-FEC Frame: A codeword generated by FEC encoding of data to be protected, which is composed of an information part and a parity part (repair part).

- Symbol: A symbol is a unit of data processed by the FEC code.

- Information symbol (s): Unprotected data or padding symbol (s) in the information part of the FEC frame, which is the information part of a FEC Frame).

- Codeword: An FEC frame generated by FEC encoding an information symbol (s).

- Parity Symbol (s): The parity symbol (s) of the FEC frame generated by FEC encoding from the information symbol (s).

- Packet: A transmission unit consisting of a header and a payload.

- Payload: The user data located in the packet to be transmitted from the sender (which is to be transmitted from the sender and which is placed inside a packet).

- Packet Header: Header for packets containing the payload.

Source Payload: A payload consisting of source symbols.

- Information Payload: A payload consisting of information symbols.

- Parity Payload: A payload consisting of parity symbols.

Source Block: A set of payloads consisting of one or more source payloads.

- Information block: A set of payloads consisting of one or more information payloads.

Parity Block: A set of payloads that consist of one or more parity payloads.

-FEC block: A set of payloads consisting of a set of codewords or an information word block and a parity block.

-FEC Delivery Block: A set of payloads consisting of source blocks and parity blocks.

-FEC Packet: Packet for transmitting FEC block.

- Source Packet: Packet for sending source block.

- Repair Packet: Packet to send Repair Block.

-FEC Packet Block: A set of packets for transmitting an FEC transport block

-MMT (MPEG Media Transport): An international standard being designed for efficient transmission of MPEG data.

- Source Flow: A Source Flow is a sequence of Source Payloads identified by the same Source Flow identifier to deliver one or more MMT Assets from one MMT Server to one or more MMT Clients.

- Parity Flow: A Parity Flow is a sequence of Parity Payloads, which are generated by FEC.

- FEC Flow: A FEC Flow consists of a Source Flow and its associated one or more Parity Flows.

- FEC Payload ID: Information that identifies the Information (sub-) payload (s) carried by an FEC Source Packet or the Parity Payload (s) carried by an FEC Parity Packet.

- FEC Source Payload ID: FEC Payload ID for Source Packet.

- FEC Payload ID: FEC Payload ID for Parity Packet.

- Access Unit (AU): An access unit is the smallest data entity that can be attributed. For non-timed data, whose timing information is not associated AU is not defined

Media Fragment Unit (MFU) is a generic container, independent of any specific media codec, containing coded media data that is independently consumable by a media decoder. It is equal to or smaller than an Access Unit (AU) and contains information that can be used by delivery layers.

- MMT Processing Unit: MMT Processing Unit (MPU) is a generic container, independent of any specific media codec, containing one or more AUs and additional delivery and consumption related information. For non-timed data, the MPU contains a portion of the data without an AU boundaries identified. It defines coded media data units that can be completely and independently processed in MMT. In this context processing means encapsulation into MMT Package or packetization for delivery.

- MMT Asset: An MMT Asset is a logical data entity that is composed of one or more MPUs. An MMT Asset is the largest data unit for which the same composition information and transport characteristics are applied. An MMT Asset contains only one kind of data type including packed or multiplexed data (eg a portion of audio ES, a portion of video ES, an MPEG-U Widget Package, a portion of MPEG-2 TS, a portion of MP4 file, and a portion of MMT Package).

- MMT Composition Information: MMT Composition Information (MMT-CI) is description about spatial and temporal composition of MMT Assets.

- MMT Media Characteristics for Transmission: MMT Media Characteristics for Transmission (MMT-MCT) MMT-MCT is represented by the parameters agnostic to specific delivery environment.

- MMT Package: An MMT Package is a logically structured collection of data, which is composed of one or more MMT Assets, MMT-CI and MMT-MCT. MMT Package may also have assigned descriptive information such as an identifier.

- MMT Payload Format: MMT Payload Format (MMT-PF) is a format of payload for MMT Package or MMT signaling message to be carried by MMT Protocol or Internet application layer protocols (eg RTP).

- MMT Transport Packet: An MMT Transport Packet (MMTP) is an application layer protocol for delivering MMT-PF over IP network.

Hereinafter, parity and repair are used in the same sense and in combination.

1A and 1B are diagrams illustrating a network topology and a data flow.

Referring to FIG. 1A, a network topology includes Host A 102 acting as a transmitter and Host B 108 acting as a receiver, where Host A 102 and Host B 108 are one or more routers 104 and 106. Is connected through. Host A 102 and Host B 108 are connected to routers 104 and 106 via Ethernet 118 and 122, which routers 104 and 106 are fiber, satellite communication or other means 120 possible. ) Can be connected to each other. Data flow between host A 102 and host B 108 is through link layer 116, internet layer 114, transport layer 112, and application layer 110.

Referring to FIG. 1B, the application layer 130 generates data 130 to be transmitted through the AL-FEC. The data 130 may be RTP packet data obtained by splitting data compressed by an audio / video (AV) codec stage using RTP (Real Time Protocol) or MMT packet data according to MMT. have. The data 130 is converted by the transport layer 112 into a UDP packet 132 into which, for example, a User Datagram Protocol (UDP) header is inserted. The Internet layer 114 attaches an IP header to the UDP packet 132 to generate an IP packet 134, and the link layer 116 adds a frame header 136 to the IP packet 134 and a frame footer if necessary. A frame 116 to be transmitted is attached with a footer).

2 is a block diagram of a MPEG Media Transport (MMT) system according to an embodiment of the present invention.

2 is a diagram showing the configuration of the MMT system, the right is a diagram showing the detailed structure of the delivery function (Delivery Function).

The media coding layer 205 compresses audio and / or video data and transmits it to an encapsulation function layer 210 (E. Layer).

The encapsulation function layer 210 packages compressed audio / video data in a form similar to a file format and delivers the compressed audio / video data to a delivery function layer 220.

The transport function layer 220 (or “D. Layer”) formats the output of the encapsulation function layer 210 in an MMT payload, and then adds an MMT transport packet header to the transport protocol layer 230 in the form of an MMT transport packet. To pass on. Alternatively, the transport function layer 220 delivers the output of the encapsulation function layer 210 to the transport protocol layer 230 in the form of an RTP packet using an existing RTP protocol. Thereafter, the transport protocol layer 230 converts the transport protocol into any one of a user datagram protocol (UDP) and a transmission control protocol (TCP) and then transmits it to the IP layer 240. Finally, IP layer 240 converts the output of transport protocol layer 230 into IP packets and transmits them using IP protocol.

The FEC packet of the present invention can be in the form of at least one of an MMT payload format, an MMT transmission packet, and an RTP packet.

The control function layer 200 (C. Layer) manages a presentation session and a delivery session.

3 is a diagram illustrating the structure of an MMT package. As shown in FIG. 3, the MMT package 310 is transmitted and received with the client 350 through the D. Layers 330-1 and 330-2 of the network, and the MMT assets 303 are provided. -1 to 303-3, composition information 301, and transport characteristics 305-1 and 305-2.

In addition, the MMT package 310 has functionality and operations for utilizing configuration information. The configuration information includes a list of MMT assets 303-1 to 303-3, configuration information 301, and transmission characteristics 305-1 and 305-2.

Description information describes the MMT package 310 and the MMT assets 303-1 through 303-3. The configuration information 301 assists in the consumption of the MMT assets 303-1 through 303-3. Transmission characteristics 305-1 and 305-2 provide hints for the delivery of MMT assets 303-1 through 303-3.

The MMT package 310 describes transmission characteristics for each MMT asset. The transport characteristics 305-1 and 305-2 include error resiliency information, and simple transport characteristic information for one MMT asset may or may not be lost. In addition, the transmission characteristics 305-1 and 305-2 may include a quality of service (QoS) of each MMT asset (loss tolerant, delay tolerant).

4 illustrates a configuration of configuration information included in the MMT package and subordinate information thereof.

The configuration information includes additional information together with identification information 312 of the package, asset list information 314 which is a component of the package, composition information 316, transport characteristics 318, and contents, as shown in FIG. It provides structural information such as how these components are contained in the package and where they are contained.

5 is a FEC packet format according to an embodiment of the present invention.

The FEC source packet 510 as the FEC packet for the source payload 520 includes a Common Header 512, an Optional Header 514, an FEC Payload ID 516 for the Source Payload, and a Data 518. The FEC source packet 510 is generated by adding an FEC Payload ID 516 for the source payload to the Source Payload 520 composed of the Common Header 512, the Optional Header 514 and the Data 518.

The FEC Parity Packet 538, which is an FEC packet for the Parity Payload 540, includes a Common Header 532, an FEC Payload ID 536 for the Parity Payload, and a Parity Payload 538. The FEC Parity Packet 538 is generated by adding a Common Header 532 to the Parity Payload 538 generated by the FEC encoding and an FEC Payload ID 536 for the Parity Payload.

5, the FEC Payload ID 516 is placed in front of the Data 518 or the Parity Payload 538, but is not limited thereto. For example, at the end of Data 518 and Parity Payload 538. [

It is preferable that the common header is located at the same position in the FEC packet. Also, the Common Header includes a Type information field so that the FEC Packet received at the FEC Packet Receiving Terminal can easily distinguish whether it is an FEC Source Packet or an FEC Parity Packet. In addition, the Common Header may include an information field indicating whether to apply the FEC. In this case, the information field indicating whether the FEC is applied is stored as a separate packet for transmitting control information different from the FEC packet, and then transmitted. Data of FEC Source Packet can have various data types. Various data may be audio data, video data, file data, timed data, non-timed data, MPU, MFU, and the like. The various data can be classified by the value of the Type field of the Common Header. In particular, in the case of the FEC Parity Packet, the value of the Type field of the Common Header 532 is set as information indicating that the parity is included (in the case of multiple parities, the parity 1 and the parity 2 are separately added in the case of the Two Stages Coding Structure , It is possible to distinguish between a parity packet and a source packet.

The common header information refers to a header having an information field commonly applied to a parity payload among header information for a data payload. The common header information includes information for distinguishing whether the received packet is a packet for a data payload or a packet for a parity payload. In addition, information indicating whether FEC of the received FEC packet is applied is transmitted in the FEC packet or as a separate control information packet different from the FEC packet. The information indicating whether the Common Header is FEC-applied is FEC-encoded after generating the Source Payload, that is, FEC is applied to the Common Header before FEC-encoding, or FEC is not applied to the Common Header when the Source Payload is generated After FEC coding, FEC payload ID for the source payload is added to the Source Payload to generate FEC packet for the source payload, and FEC application information of the Common Header may be FEC applied. In the latter case, the receiver recognizes that the packet is FEC-applied from the FEC packet for the received source payload, changes the FEC application information of the Common Header to not applied again, and then generates FEC decryption by generating the source payload. Accordingly, by placing the common header at a fixed position such as the head or tail of the FEC packet, it is possible to distinguish whether the received FEC packet is a packet for a data payload or a packet for a parity payload.

Table 1 below shows FEC type information indicating whether the MMT packet header is applied with FEC and whether the MMT packet header is an FEC source packet or an FEC repair (parity) packet.

<Table 1> - value of FEC_type

Value Description 0 MMT packet without AL-FEC protection One MMT packet with AL-FEC protection (FEC source packet) 2 MMT packet for repair symbol (s) (FEC repair packet) 3 Reserved for future use

FEC type = 0 indicates that the FEC is not applied to the MMT packet.

FEC type = 1 indicates an FEC-applied MMT packet.

FEC type = 2 indicates an MMT packet for a parity payload generated after FEC encoding.

For the data to be transmitted by applying the FEC, the transmitter generates MMT packets (i.e., source payloads) whose FEC type is set to "1 ", generates information payloads, and then performs FEC encoding to generate parity payload . The parity payload generated by the FEC encoding is generated by adding an MMT header with the FEC type set to "2", an MMT payload header and an FEC payload ID for the parity payload to generate an FEC packet for the parity payload, Along with an FEC packet for loading. The receiving end separates the FEC type information of the received MMT packets from the FEC type information of the FEC type information and whether the FEC source packet or the FEC repair packet is applied. Then, the FEC decoding is performed using the FEC source packet and the FEC repair packets. And restores the packet.

Alternatively, the transmitting end generates MMT packets (i.e., source payloads) whose FEC type is set to " 0 " for data to be transmitted by applying FEC, generates information payloads, . The FEC type of the MMT packet header is changed to "1 " when the FEC source packet is generated for the source payload by adding the FEC payload ID to the MMT packet after the FEC encoding is performed The FEC packet is set to FEC type = 0, and the FEC packet for the source payload after the FEC coding is set to FEC type = 1. The parity payload generated by the FEC encoding is generated by adding an MMT header with the FEC type set to "2", an MMT payload header and an FEC payload ID for the parity payload to generate an FEC packet for the parity payload, Lt; / RTI &gt; for the &lt; RTI ID = 0.0 &gt; The receiving end distinguishes whether the FEC type and the FEC source packet or the FEC repair packet are applied based on the FEC type information of the MMT packet header of the received MMT packets. The FEC source packet (FEC type = 1) and the FEC repair packet ), The FEC source packet removes the FEC payload ID, changes to FEC type = 0 to switch to MMT packets (source payloads), performs FEC decoding using the parity payload of the FEC repair packet, Recovers lost packets.

Optional Header 514 of the FEC Source Packet 510 is information applied only to the source packet, and information indicating the fragmentation state information of the MPU / MFU, the header length information, and the asset related to the data of the packet (in the present invention) Although the information may be an Asset ID, it is preferable to improve transmission efficiency by transmitting compressed Asset ID information mapped with the Asset ID, in which case, the Asset ID and the compressed Asset ID information mapped thereto are transmitted out-of-band. At least one of the above).

In the case of the FEC header, the FEC Payload ID for the source packet and the FEC Payload ID for the parity packet may be the same or different according to the FEC-related control information arrangement method according to the FEC control information according to the information block generation method. The FEC Payload ID information includes FEC Flow information, FEC coding structure information, Source Packet number information, Information Payload number information, Parity Packet number information, Packet Sequence Number (or Source / Parity Packet Sequence Number or Information / Parity Payload ID Information indicating the order of information payload and parity payload in the payload information), and Block Boundary Info (or Source Block Number).

In the MMT system shown in FIG. 2, the FEC packet is an output of the MMT D.1 layer when FEC is applied, and is input into an application protocol such as the MMT D.2 Layer or RTP. However, if FEC is not applied, FEC Packet is Source Payload because it does not need FEC Payload ID and it is output of MMT D.1 Layer. Although not shown, the FEC Packet according to the present invention is a Source Payload itself that includes FEC Payload ID when FEC is applied and FEC Payload ID when FEC is not applied.

The MMT Server sends one or more MMT Assets to the MMT Client. Each Asset consists of one or more MPUs and each MPU is packed into one or more MMT Payload Format (MMT-PF) in D.1 Layer. D.2 Header is added and converted into MMT Transport Packet (MMT_TP) and transmitted to the lower layer. Considering the transmission of multiple MMT Assets, the D.2 Header of the MMT Transport Packet for each Asset stores information for identifying each Asset (Asset_ID, for example) Indicates which asset data is being transmitted. A plurality of MMT Assets transmitted are constituted by MMT Asset individually, and one Parity Flow is generated or protected by FEC, or two or more Assets constitute one Source Flow, and one Parity Flow is generated by FEC Protected or some Assets are protected by FEC and some Assets may not be protected by FEC. Also, it can be protected by creating two or more parity flows in one source flow like Two Stage FEC coding structure or Layer-Aware FEC coding structure.

When protected by FEC, one Source Flow consisting of one or more MMT Assets transmits Parity Payloads in Parity Flow generated by FEC after making MMT TP like MMT Asset. Each D.2 Header (For example, Asset ID) for identifying the parity payloads are stored and transmitted. When two or more parity flows such as Two Stage FEC or LA (Layer-Aware) -FEC coding structure are generated, each Parity Flow can be classified into Parity Flow ID. In this case, FEC Flow IDs corresponding to the number of FEC flows transmitted as FEC Out-of-Band signals are defined, and mapping information for Source Flow and Parity Flow corresponding to each FEC Flow ID is provided.

For example, when a Video Asset, an Audio Asset, a Widget Asset, and a File Asset are transmitted, Video, Audio, and Widget Assets are composed of one Source Flow and protected by Two Stage FEC coding structure. The FEC Out-of-Band Signal transmits the following information when it is composed of One Stage FEC coding structure.

- Video Asset: Asset_ID = 1

- Audio Asset: Asset ID = 2

- Widget Asset: Asset ID = 3

- File Asset: Asset ID = 4

- Number of FEC Flow = 2

■ FEC Flow ID = 1

FEC coding structure: Two Stage FEC coding structure

Source Flow: Asset ID 1, 2, 3

Parity Flow 1: Asset ID 101

Parity Flow2: Asset ID 102

■ FEC Flow ID = 2

FEC coding structure: One Stage FEC coding structure

Source Flow: Asset ID 4

Parity Flow: Asset ID 103

And

- In MMP TP Header that transmits Video Asset, Asset ID = 1

- For MMP TP Header that transmits Audio Asset, Asset ID = 2

- The MMP TP Header that transmits Widget Asset has Asset ID = 3

- The MMP TP Header that transmits Parity Flow1 has Asset ID = 101

- In MMP TP Header that transmits Parity Flow2, Asset ID = 102

- Asset ID = 4 for MMP TP Header that transmits File Asset

- Asset ID = 103 is set in the MMP TP Header that transmits Parity Flow for File Asset, and the value of the corresponding Asset ID field is set to 1, 2, and 3 from the Asset ID = 103 in the FEC Out-of-Band signal and MMT TP Header, 3, 101, and 102 constitute one FEC flow, and Asset ID values 4 and 103 constitute another FEC flow, thereby performing FEC decoding smoothly.

6 is a FEC packet format when the Source Payload according to an embodiment of the present invention is an MMT Transport Packet.

The FEC source packet 610 includes a D2 header (Common Header 612 + Optional Header 614), an MMT Payload Format 616, and an FEC Payload ID 618. On the other hand, the FEC Parity Packet 630 includes a Common Header 632, an FEC Payload ID 636, and a Parity Payload 638. In the Common Header, an ID field for identifying the Assets is constructed. In case of the Parity Packet, an ID value for identifying the Parity Flow is set. The Common Header may include a Global Sequence Number field, a Delivery Time Stamp, and the like. 6, the FEC Parity Pakcet may come with an optional header after the Common Header (i.e., the D2 header of the FEC Source Packet and the D2 Header of the FEC Parity Packet may be the same). Since the MMT TP Header performs the function of the protocol, it is possible to drop the FEC parity packet when the network entity drops the packet according to the congestion condition of the network, so that it can have the same header structure as the FEC Source Packet Do.

FIG. 7 is a FEC packet format when Source Payload according to an embodiment of the present invention is an MMT Payload Format.

The FEC source packet 710 includes a D2 header 712, a D1 header (Common header 714), an optional header 716, a D1 payload 718, an FEC payload ID 720, 730 includes a D2 Header 732, a Common Header 736, an FEC Payload ID 738, and a Parity Payload 740. In this case, the Asset ID field of the D2 header is prepared, the Asset ID field for the FEC Source Packet is set to an ID value for identifying each Asset, the Asset ID field for the FEC Parity Packet is set to an ID value Is set. The same information as in FIG. 5 is stored in the Common Header.

8 is a block diagram of a transmission apparatus according to an embodiment of the present invention.

The FEC control information generation unit 601 determines whether FEC is applied or not, and generates FEC-related control information when FEC is applied.

The source payload generating unit 603 receives the data stream for transmission of the MMT Asset from the upper Encapsulation Layer, divides the data stream into a data payload having a predetermined size, and adds a Common Header and an Optional Header to generate a Source Payload. Based on the FEC control information of the FEC control information generation unit 601, the source payload generation unit 603 generates a source block composed of a predetermined number of source payloads from the sequence of Source Payloads having the same FEC flow to the information block generation unit 605 ).

The information block generation unit 605 generates an information block from the input Source Payload generation unit 603 based on the FEC control information, and outputs the information block to the FEC coding unit 607.

The FEC coding unit 607 generates a predetermined parity data based on the input information block and the FEC control information, and inputs the generated parity data to the source payload generating unit 603.

The source payload generating unit 603 generates a parity payload based on the inputted parity data based on the FEC control information.

The FEC packet generation unit 609 generates a FEC packet by adding a Common Header and an FEC Header to the Parity Payload and adds an FEC Header to the generated Source Payload to generate an FEC Source Packet, 611). The transmitting unit 611 transmits the FEC packet to the lower layer.

When the FEC is not applied, the source payload generating unit 603 and the FCE packet generating unit 609 generate a source payload based on the data stream and the FEC control information, and transmit it to the transmitting unit 611 as an FEC packet .

In the MMT system of FIG. 2, the FEC packet is transmitted through an application protocol such as MMT D.2 Layer or RTP.

A control unit not shown in the figure divides the data stream into data payloads of a predetermined size, adds a header to each divided data payload to generate a source payload, and adds a first FEC Adding a Payload ID, applying an FEC code to generate an FEC source packet for the source payload, adding a second FEC Payload ID to at least one or more parity payloads, applying an FEC code to the at least one parity payload And generates an FEC parity packet for loading.

The transmitting unit transmits the FEC source packet and the FEC parity packet.

9 is a block diagram of a receiving apparatus according to an embodiment of the present invention.

The FEC packet receiving unit 701 receives the packet stream and discriminates whether the FEC is applied from the common header of the corresponding packet and whether the corresponding packet is a source packet or a parity packet. When the FEC is applied, the FEC packet receiving unit 701 receives the FEC header of the source packet and the FEC header And transmits the FEC-related control information to the FEC controller. If there are multiple FEC flows, the FEC controller divides the control information for each flow and controls the FEC decoding for each flow.

The source payload reconstructing unit 703 transmits the data of the packet (i.e., source payload) to which the FEC is not applied from the received FEC packets to the data stream unit. If FEC is applied, it distinguishes source payload which is not received by source payload received in packets having the same FEC flow and outputs corresponding data when all source payloads are received based on FEC control information. Otherwise, it outputs the received FEC Parity Packet And outputs the parity payloads to the FEC Block reconstructing unit 705 together with the received source payloads.

The FEC Block reconstructing unit 705 reconstructs the FEC Block composed of the information block and the parity block based on the FEC control information from the received Source Payload and Parity Payloads (performs Erasure processing on the lost payloads at this time) (707).

The FEC decoding unit 707 performs decoding based on the FEC control information from the input FEC control unit 709, restores the lost information payloads, and outputs the recovered information payloads to the FEC Block reconstructing unit 705.

The FEC Block reconstructing unit 705 restores the source payload based on the FEC control information from the restored information payloads (using the information payload reconstructed from the source payloads if necessary), and sends the restored source payload to the source payload reconstructing unit 703 Output.

The source payload reconfiguration unit 703 transmits data of the restored source payload and the received source payload to the upper layer.

9, the FEC controller 709 distinguishes between an in-band signal and an out-band signal in generating FEC-related control information, and sets an FEC packet And the out-of-band signal is transmitted through the C. layer in the case of the MMT system as shown in FIG. 2 or transmitted to the receiving end through the Session Description Protocol (SDP).

A control unit, not shown in the figure, distinguishes between the FEC source packet and the FEC parity packet from the packet received from the transmitter, obtains the source payload from the FEC source packet, and obtains the parity payload from the FEC parity packet. In this case, the source payload is a data stream divided into a data payload of a predetermined size, a header is added to each divided data payload, and the FEC source packet is added with a first FEC Payload ID to the source payload , The FEC code is applied, and the FEC parity packet is a parity payload in which a second FEC Payload ID is added and an FEC code is applied.

10 is a diagram illustrating an embodiment of an information word block configuration in the FEC block generation unit 520 according to an embodiment of the present invention.

10, when eight source payloads having a variable packet size, that is, SPL # 0 to SPL # 7 are input, the payload size of each payload is made equal to a payload having a maximum length, for example, S_max After the padding data is added, an information word block composed of 8 information word payloads, i.e., IPL # 0 to IPL # 7, is generated. In the embodiment of FIG. 10, although the maximum length S_max of the source payload and the length of the information payload are set to be the same, the length of the information payload may have a value smaller than S_max depending on the system complexity and memory requirements.

11 is a diagram illustrating another embodiment of an information word block configuration in the FEC packet generation unit 609 according to an embodiment of the present invention.

11, when 8 source payloads having a variable packet size, that is, SPL # 0 to SPL # 7 are inputted, the payload sizes are arranged in a row, and then the maximum length of the information payload, , S_max is divided into units, thereby generating an information word block composed of five information word payloads, i.e., IPL # 0 to IPL # 4. Note that padding data is included in the last information payload. According to the embodiment of FIG. 11, since the boundaries of the source block and the information payload do not match, information necessary for extracting the source payload from the information word block such as the length of each source payload is included in the information word block Or transmitted to the receiving end in a separate manner. In the embodiment of FIG. 11, although the maximum length S_max of the source payload and the length of the information payload are set to be the same, the length of the information payload may have a value smaller than S_max depending on the system complexity and memory requirements have.

In FIG. 8, the FEC encoding unit 607 calculates a parity symbol using an FEC encoding algorithm determined by inputting the information word block, generates a parity payload composed of the parity symbols, and outputs the parity payload as a parity block .

If K is less than or equal to 200 from the FEC encoder information word block, an information word symbol for RS coding may be generated as shown in FIG. 12 by mapping the source block to the information word block. Alternatively, as shown in FIG. 13, May be generated.

14 and 15, the parity symbols are generated by performing RS and LDPC encoding on each information word symbol. 15, although shortening and puncturing are not shown, shortening and puncturing are performed similarly to the case of FIG. 20 for various K and P by using an LDPC code having a predetermined length, Symbols can be generated. At this time, only the shortening may be performed, and only the puncturing may be performed.

From the generated parity symbols, an RS parity block and an LDPC parity block are generated as shown in FIGS.

The following shows RS code and LDPC code specifications. The primitive polynomial of the RS (N, K) code over the finite field GF (2 ^ 8) (RS (N, K) code over the finite field GF (2 ^ 8 + x ^ 4 + x ^ 3 + x ^ 2 + 1.

The symbols in GF (2 ^ 8) can be represented as (a ^ 7, a ^ 6, a ^ 5, a ^ 4, a ^ 3, a ^ 2, a, 1).

Where a = 00000010 (binary number).

Each RS code word rsc includes a finite field whose information represented by rsc = (e0, e1, ..., e199, p200, ..., p239) expressed by a vector is 200 bytes and the parity is 40 bytes (240,40) code over the finite field GF (2 ^ 8) on the GF (2 ^ 8).

The LDPC (K + P, k) code (LDPC (K + P, K) code over the finite field GF (2)) on the finite field GF (2) consists of K information bits and P parity bits Is a QC-LDPC structure. Where K = L x 400, P = L x 80, and L = 1, 2, 4, 8, or 16.

Particularly, the parity part of the LDPC has an approximate triangular matrix form as shown in FIG. 16 shows the structure of the H matrix.

In Fig. 18, K = 400 and P = L x 80 (L = 1, 2, 4, 8 or 16).

Although the embodiments of the RS code and the LDPC code have been described above, it is to be noted that not only the codes limited to the detailed description described above can be used, but all the codes commonly referred to as FEC codes such as Raptor or Raptor Q, .

A method of generating an FEC packet according to an embodiment of the present invention will now be described.

And sets the payload type of the Common header of each packet to match the corresponding payload. That is, the payload type of the packet for the source payload indicates the source payload (audio, video, etc.) and the payload type of the packet for the repair payload is the repair payload. Sequence numbers for source packets are continuously given, and sequence numbers for repair packets are continuously given. The start sequence number is used to identify the boundaries of the repair blocks in the corresponding FEC block. For example, Start from the start sequence number. That is, the boundaries of the repair packets (or the repair blocks) can be known by setting the sequence numbers for the source packets and the repair packets in the FEC block to be correlated. And stores the start sequence number of the FEC block as FEC block boundary information in the header of each packet. In addition, when FEC is selectively applied, FEC flag (Flag) information is stored. In addition, when the number of source packets or the number of repair packets of the FEC block is variably operated, information on the number of packets (or the number of source packets) and the number of source packets (or the number of repair packets) of the FEC block are stored.

According to an embodiment of the present invention, the FEC configuration related information or other encoding configuration related information is signaled and transmitted to the receiver so that the transmitter can selectively transmit FEC-applied content. According to an embodiment of the present invention, the FEC can be selectively applied according to the network conditions and the quality of service (QoS) of the content. According to an embodiment of the present invention, all or a part of the FEC control information including the FEC configuration related information or other encoding configuration related information may be periodically repeatedly transmitted, or the FEC control information may be transmitted by the in-band signaling method of the present invention By transmitting all or a part of the related information, it is possible to inform the new receiver about the FEC configuration information even in the situation where the service is already in progress, so that the newly appeared receiver also performs the FEC decoding to recover the lost data, Can be provided.

When there are a plurality of data streams (Source Flow) output from the application layer, the transmission system stores information capable of distinguishing the plurality of streams in an FEC packet (for example, FEC Flow ID information) It is preferable that the same information is also stored in an FEC packet for a stream (Parity Flow) composed of the generated parity payload so that the receiver can know which stream of parity payloads are related to each of the plurality of streams.

Or if there are a plurality of data streams (Source Flow) output from the application layer, information that can distinguish between each data stream and each parity payload stream is stored in the FEC packet (e.g., Source Flow ID, Parity Flow ID) and transmits the mapping information (for example, FEC Flow ID 1 = Source Flow ID 1 + Parity Flow ID 1) with the parity stream generated for FEC protection of each data stream to a separate control information .

Accordingly, the receiving apparatus distinguishes each data stream from the stream identification information in the FEC packet or from the control information separate from the FEC packet, grasps the parity stream generated for FEC protection of each data stream, and performs FEC decoding Can be performed smoothly.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments, but is capable of various modifications within the scope of the invention. Therefore, the scope of the present invention should not be limited by the described embodiments, but should be determined by the scope of the appended claims and equivalents thereof.

Claims (20)

In a method for transmitting a packet in a communication system,
Dividing the data stream into data payloads of a predetermined size, and adding a common header to each divided data payload to generate a source payload;
Adding a first FEC Payload ID to the source payload and applying an FEC code to generate an FEC source packet for the source payload;
Adding a second FEC payload ID to at least one parity payload and generating an FEC parity packet for the at least one parity payload by applying an FEC code; And
And transmitting the FEC source packet and the FEC parity packet. The method of claim 1,
Wherein the common header includes an information field commonly applied to the source payload and the parity payload. 3. The method of claim 2,
Wherein the common header indicates at least one of FEC application, FEC source packet, and FEC parity packet. The method of claim 1,
Wherein the first FEC Payload ID identifies the FEC source payload sent by the FEC source packet. The method of claim 1,
And the second FEC Payload ID identifies the FEC parity payload transmitted by the FEC parity packet. An apparatus for transmitting a packet in a communication system,
Dividing the data stream into data payloads of a predetermined size, and adding a header to each divided data payload to generate a source payload;
Adding a first FEC Payload ID to the source payload, generating an FEC source packet for the source payload by applying an FEC code, adding a second FEC Payload ID to the at least one parity payload, A control unit for generating an FEC parity packet for the at least one parity payload by applying the FEC parity packet; And
And a transmitter for transmitting the FEC source packet and the FEC parity packet. The method according to claim 6,
Wherein the common header includes an information field commonly applied to the source payload and the parity payload. The method of claim 7, wherein
Wherein the common header indicates at least one of FEC application, FEC source packet, and FEC parity packet. The method according to claim 6,
Wherein the first FEC Payload ID identifies the FEC source payload transmitted by the FEC source packet. The method according to claim 6,
Wherein the second FEC Payload ID identifies the FEC parity payload transmitted by the FEC parity packet. A method for receiving a packet in a communication system,
A step of discriminating whether a packet received from a transmitter is an FEC source packet or an FEC parity packet;
Obtaining a source payload from the FEC source packet and obtaining a parity payload from the FEC parity packet,
The source payload divides a data stream into a data payload of a predetermined size, a header is added to each divided data payload,
The FEC source packet has a first FEC Payload ID appended to the source payload and has an FEC code applied thereto,
Wherein the FEC parity packet is a parity payload to which a second FEC Payload ID is added and an FEC code is applied thereto. 12. The method of claim 11,
Wherein the common header includes an information field commonly applied to the source payload and the parity payload. The method of claim 12,
Wherein the common header indicates at least one of whether FEC is applied, whether it is an FEC source packet, or whether it is an FEC parity packet. 12. The method of claim 11,
Wherein the first FEC Payload ID identifies the FEC source payload sent by the FEC source packet. 12. The method of claim 11,
Wherein the second FEC Payload ID identifies the FEC parity payload transmitted by the FEC parity packet. An apparatus for receiving a packet in a communication system,
A control unit for distinguishing between a FEC source packet and an FEC parity packet from a packet received from a transmitter, obtaining a source payload from the FEC source packet, and obtaining a parity payload from the FEC parity packet,
The source payload divides a data stream into a data payload of a predetermined size, a header is added to each divided data payload,
The FEC source packet has a first FEC Payload ID appended to the source payload and has an FEC code applied thereto,
Wherein the FEC parity packet is a parity payload to which a second FEC Payload ID is added and an FEC code is applied thereto. 17. The method of claim 16,
Wherein the common header includes an information field commonly applied to the source payload and the parity payload. 18. The method of claim 17,
Wherein the common header indicates at least one of FEC application, FEC source packet, and FEC parity packet. 17. The method of claim 16,
Wherein the first FEC Payload ID identifies the FEC source payload sent by the FEC source packet. 17. The method of claim 16,
Wherein the second FEC Payload ID identifies the FEC parity payload transmitted by the FEC parity packet.

Images