JP2015136059A - Communication device, communication data generation method, and communication data processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a configuration which allows for discrimination whether or not the storing coded data of a communication packet is a data based on a coded data capable of random accessing.SOLUTION: A transmitter generates a packet storing a NAL unit, i.e., the component of a coded data or a NAL unit fragment obtained by further dividing the NAL unit, and transmits the packet while setting access point information indicating whether or not a data becoming a random access point is included, as additional information, in the NAL unit, i.e., the original data of the packet storage data. A receiver discriminates whether or not the packet storage data is generated based on a coded data capable of random accessing, with reference to the packet additional information.

Description

  The present disclosure relates to a communication device, a communication data generation method, and a communication data processing method. More specifically, for example, the present invention relates to a communication apparatus, a communication data generation method, and a communication data processing method that execute transmission or reception of data via broadcast waves or a network.

  OTT (Over The Top) is known as a data distribution method that enables distribution of contents such as image data and audio data regardless of the service form of each communication carrier. The delivery content by OTT is called OTT content, and the delivery service of image (video) data using OTT is called OTT video and OTT-V (Over The Top Video).

  For example, DASH (Dynamic Adaptive Streaming HTTP) is used as a basic technology for data streaming distribution according to OTT-V. DASH is a standard for adaptive streaming technology that uses the HyperText Transfer Protocol (HTTP) protocol.

  In adaptive (adaptive) streaming, the content distribution server enables reproduction of distribution content by various clients. Therefore, a segmented file of moving image content of a plurality of bit rates, and a manifest file describing these attribute information and URL Create and hold

  The client acquires the manifest file from the server, selects the optimum bit rate content according to the size of the display unit of the own device and the available communication band, receives the selected content, and plays it. It is possible to dynamically change the bit rate according to fluctuations in the network bandwidth, and it is possible for the client side to switch and receive the optimal content according to the situation at any time, reducing the occurrence of video interruptions Playback is realized. Note that adaptive (adaptive) streaming is described in, for example, Japanese Patent Application Laid-Open No. 2011-87103.

The MPEG-DASH standard is a standard that defines the specifications for streaming distribution of moving picture and audio data encoded according to MPEG (Moving Picture Expert Group) according to the DASH.
The MPEG-DASH standard includes the following two standards.
(A) A standard relating to a manifest file (MPD: Media Presentation Description) for describing metadata that is management information of moving images and audio files.
(B) Standard relating to a file format (segment format) for moving image content transmission.
When MPEG data is streamed according to DASH, processing according to the MPEG-DAH standard is performed.

However, DASH, which is a standard for adaptive (adaptive) streaming technology, is based on point-to-point HTTP streaming technology, which is a one-to-one communication process.
For this reason, when applying to streaming distribution of contents (programs) that many clients may view at the same time, such as sports broadcasts, it is necessary to support CDN (Content Delivery Network).

  However, there is a cost limitation in constructing point-to-point HTTP streaming using CDN, and it is difficult to realize scalability comparable to broadcast distribution. As described above, DASH uses a streaming protocol based on HTTP, and has a problem that it is not suitable for content distribution that many clients view simultaneously, such as broadcast distribution.

JP 2011-87103 A

  The present disclosure provides a communication device, a communication data generation method, and communication data that simultaneously provide content to a large number of clients in content distribution using a streaming protocol based on HTTP, and realize real-time reproduction with little delay in each client. An object is to provide a processing method.

The first aspect of the present disclosure is:
A data processing unit for generating a packet storing encoded data;
A communication unit that transmits the packet generated by the data processing unit;
The data processing unit
Generate a packet that stores a NAL unit that constitutes a GOP (Group of Pictures), which is a processing unit of encoded data, or a NAL unit fragment obtained by further dividing the NAL unit,
As the additional information for the packet, the communication is performed by generating a packet in which access point information indicating whether the NAL unit stored in the packet or the NAL unit that is the division source of the NAL unit fragment includes data to be a random access point is set. In the communication device that transmits via the unit.

  Furthermore, in an embodiment of the communication apparatus according to the present disclosure, the access point information is data generated based on a NAL unit in which packet storage data includes encoded data specified by SAP (Streaming Access Point). It is information indicating whether or not.

  Furthermore, in an embodiment of the communication apparatus according to the present disclosure, the data processing unit includes a sub GOP media segment storing media data including one or more NAL units that are GOP configuration data and metadata corresponding to the media data. Are generated as packet storage data.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the data processing unit generates a NAL unit fragment obtained by further dividing a NAL unit that constitutes a GOP, and stores a plurality of packets storing the NAL unit fragment, and the NAL A packet storing metadata corresponding to the unit is generated.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the data processing unit divides the encoded data to be transmitted into data equal to or less than a maximum data transfer unit (MTU) defined on the communication path. A packet storing the generated divided data is generated.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the data processing unit generates an HTTP packet and records the additional information in an HTTP header in the generated packet.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the data processing unit records the additional information in an extension header in a generated packet.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the data processing unit records the additional information in an LCT header in a generated packet.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the communication unit broadcasts or multicasts a packet generated by the data processing unit via a broadcast wave.

Furthermore, the second aspect of the present disclosure is:
A communication unit that receives the encoded data storage packet transmitted by the transmission device; and
A data processing unit that inputs a packet received by the communication unit and executes processing;
Each of the encoded data storage packets received by the communication unit is:
A packet that stores a NAL unit that constitutes a GOP (Group of Pictures) that is a processing unit of encoded data, or a NAL unit fragment that is a further division of the NAL unit. As additional information, the NAL unit or NAL stored in the packet The communication apparatus is a packet in which access point information indicating whether or not a NAL unit that is a unit fragment division source includes data to be a random access point is set.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the data processing unit refers to the access point information that is additional information recorded in the received packet, and the data stored in the received packet is converted to SAP (Streaming). A process for determining whether or not the data is generated based on the NAL unit including the encoded data specified by (Access Point) is executed.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the data processing unit refers to the access point information that is additional information recorded in the received packet, and selectively acquires and decodes data including the access point. Execute playback process.

  Furthermore, in one embodiment of the communication device according to the present disclosure, the communication unit receives an HTTP packet in which the additional information is recorded in an HTTP header, and the data processing unit acquires the additional information from the HTTP header of the received packet. To do.

  Furthermore, in an embodiment of the communication device according to the present disclosure, the communication unit receives a packet in which the additional information is recorded in an extension header, and the data processing unit acquires the additional information from the extension header of the received packet. .

  Furthermore, in an embodiment of the communication device according to the present disclosure, the communication unit receives a packet in which the additional information is recorded in an LCT header, and the data processing unit acquires the additional information from the LCT header of the received packet. .

  Furthermore, in an embodiment of the communication device according to the present disclosure, the communication unit receives the packet via a broadcast wave.

Furthermore, the third aspect of the present disclosure is:
A communication data generation method executed in a data transmission device,
The data processor
Generate a packet that stores a NAL unit that constitutes a GOP (Group of Pictures), which is a processing unit of encoded data, or a NAL unit fragment obtained by further dividing the NAL unit,
As a supplementary information for the packet, a communication unit is generated by generating a packet in which access point information indicating whether or not the NAL unit stored in the packet or the NAL unit that is the division source of the NAL unit fragment includes data serving as a random access point is set. In the communication data generation method to be transmitted via

Furthermore, the fourth aspect of the present disclosure is:
A communication data processing method executed in a data receiving device,
A communication unit receiving an encoded data storage packet transmitted by a transmission device;
A data processing unit executes a data processing step of inputting a packet received by the communication unit and executing a process,
Each of the encoded data storage packets received by the communication unit is:
A packet that stores a NAL unit that constitutes a GOP (Group of Pictures) that is a processing unit of encoded data, or a NAL unit fragment that is a further division of the NAL unit. As additional information, the NAL unit or NAL stored in the packet A packet in which access point information indicating whether or not the NAL unit that is the source of the unit fragment includes data to be a random access point is set,
The data processing step is a communication data processing method which is a step of executing a process by inputting a packet in which access point information received by the communication unit is set.

  Other objects, features, and advantages of the present disclosure will become apparent from a more detailed description based on embodiments of the present disclosure described below and the accompanying drawings. In this specification, the system is a logical set configuration of a plurality of devices, and is not limited to one in which the devices of each configuration are in the same casing.

According to the configuration of an embodiment of the present disclosure, a configuration is realized in which it is possible to determine whether or not the stored encoded data of a communication packet is data based on encoded data that can be accessed randomly.
Specifically, the transmission device generates a packet storing a NAL unit that is a component of encoded data or a NAL unit fragment obtained by further dividing the NAL unit, and is the original data of the packet storage data as additional information Access point information indicating whether or not the NAL unit includes data to be a random access point is set and transmitted. The receiving apparatus refers to the packet additional information to determine whether or not the packet storage data is data generated based on randomly accessible encoded data.
With this configuration, it is possible to realize a configuration in which it is possible to determine whether or not the stored encoded data of the communication packet is data based on randomly accessible encoded data for each packet.
Note that the effects described in the present specification are merely examples and are not limited, and may have additional effects.

It is a figure explaining an example of 1 composition of a communications system which performs processing of this indication. It is a figure explaining the transmission data of a transmitter. It is a figure explaining the structural example of a DASH segment. It is a figure explaining a fragment production | generation sequence. It is a figure explaining the structural example which set the media data (mdat) in a fragment to the data which subdivided 1 GOP instead of the data of 1 GOP unit. It is a figure explaining the structural example of the HTTP packet which set the HTTP header to the sub GOP media segment. It is a figure explaining the recording information of the HTTP header of a media segment HTTP packet. It is a figure explaining the recording information of the HTTP header of a media segment HTTP packet. It is a figure explaining the recording information of the HTTP header of an initialization segment HTTP packet. It is a figure explaining the structural example of an IP packet. It is a figure explaining the protocol stack of a transmitter and a receiver. It is a figure explaining the protocol stack of a receiver. It is a figure explaining the recording information of the HTTP header of a media segment HTTP packet. It is a figure explaining the recording information of the HTTP header of a media segment HTTP packet. It is a figure explaining the structure of a MAC frame. It is a figure explaining a NAL unit fragment corresponding | compatible HTTP packet. It is a figure explaining the additional information recorded on the HTTP header of a metadata storing type NAL unit fragment corresponding | compatible HTTP packet. It is a figure explaining the additional information recorded on the HTTP header of a media data storage type NAL unit fragment corresponding | compatible HTTP packet. It is a figure which shows the flowchart explaining the process sequence which a transmitter performs. It is a figure which shows the flowchart explaining the process sequence which a receiver performs. It is a figure which shows the flowchart explaining the process sequence which a receiver performs. It is a figure explaining the example of the additional information recorded on an extension header. It is a figure explaining the Example which recorded additional information on the extension header. It is a figure explaining the structural example of the LCT header set according to a FLUTE protocol. It is a figure explaining the data recording structure with respect to the header extension part of a LCT header. It is a figure explaining the data recorded on the header extension part of a LCT header. It is a figure explaining the structure of HEVC encoding data. It is a figure explaining the hardware structural example of a communication apparatus.

Hereinafter, details of the communication device, the communication data generation method, and the communication data processing method of the present disclosure will be described with reference to the drawings. The description will be made according to the following items.
1. 1. Configuration example of communication system 2. Problems with simultaneous content distribution 3. Transmission data segmentation processing configuration 4. Example in which additional information is recorded in HTTP header Packet configuration 6. Configuration and processing of transmission device and reception device 7. Regarding a configuration to which access point information that makes it possible to identify whether or not a NAL unit stored in a packet includes an SAP (stream access point) is added 8. Communication processing configuration using a NAL unit fragment storage packet obtained by further dividing a NAL unit 9. Processing sequence of transmitting device and receiving device 10. Embodiment in which additional information is recorded in extension header 11. Example in which additional information is recorded in LCT header 12. Application example to HEVC encoded data 14. Hardware configuration example of each device Summary of composition of this disclosure

[1. Communication system configuration example]
First, a configuration example of a communication system that executes the processing of the present disclosure will be described with reference to FIG.
As illustrated in FIG. 1, the communication system 10 includes a transmission device 20 that is a communication device that transmits content such as image data and audio data, and a reception device 30 that is a communication device that receives the transmission content of the transmission device 20. .

Specifically, the transmission device 20 is a device that provides content, such as a broadcasting station 21 and a content server 22.
On the other hand, the receiving device 30 is a client device of a general user, and specifically includes, for example, a television 31, a PC 32, a portable terminal 33, and the like.

  Data communication between the transmission device 20 and the reception device 30 is performed as communication using at least one of bidirectional communication via a network such as the Internet, one-way communication using a broadcast wave, or the like, or both.

Content transmission from the transmission device 20 to the reception device 30 is executed in accordance with the MPEG-DASH standard, which is a standard of adaptive (adaptive) streaming technology.
As described above, the MPEG-DASH standard includes the following two standards.
(A) a standard concerning a manifest file (MPD: Media Presentation Description) for describing metadata that is management information of a moving image or an audio file;
(B) Standards related to file format (segment format) for video content transmission;
Content distribution from the transmission device 20 to the reception device 30 is executed according to the MPEG-DASH standard.

The transmission device 20 encodes the content data, and generates a data file including the encoded data and the metadata of the encoded data. The encoding process is performed in accordance with, for example, an MP4 file format defined in MPEG. When the transmission apparatus 20 generates an MP4 format data file, the encoded data file is called “mdat”, and the metadata is called “moov” or “moof”.
Details of these encoded data will be described later.

  The content provided by the transmission device 20 to the reception device 30 is various data such as music data, video data such as movies, television programs, videos, photos, documents, pictures and charts, games and software.

Transmission data of the transmission device 20 will be described with reference to FIG.
As shown in FIG. 2, the transmission apparatus 20 that performs data transmission in accordance with the MPEG-DASH standard stores an initialization segment 50 and a media segment 60 in a packet, respectively. Send to. Each of these segments is called a DASH segment.

Each of the media segments 60 is a segment in which MPEG-coded content data is divided and stored.
The initialization segment 50 is a segment that stores initial setting information required when the content stored in the media segment 60 is reproduced on the receiving device 30 side, such as codec setting information.
Note that the transmission device 20 divides and stores content such as one movie or program into a number of media segments 60 and sequentially transmits them.

  The transmitting device 20 stores the initialization segment 50 and the media segment 60 shown in FIG. 2 in an HTTP packet, and further generates and transmits an IP packet storing the HTTP packet. Details of the configuration of the transmission packet will be described later.

  The receiving apparatus 30 shown in FIG. 2 first receives one initialization segment 50 and executes a setting process such as a codec in accordance with setting information stored in the initialization segment. Subsequently, the media segments 60 are sequentially received and decoded according to the playback order to perform playback processing.

[2. Problems in simultaneous simultaneous content distribution]
As described above, DASH, which is a standard for adaptive streaming technology, is based on point-to-point HTTP streaming, and is not suitable for the simultaneous distribution of content that many clients view simultaneously. There is.
However, it is considered that content can be simultaneously provided to a large number of clients (reception devices) without delay by using multicast and broadcast (MC / BC) together.

  Transport protocols applicable to multicast and broadcast (MC / BC) type streaming include, for example, RTP (Real-time Transport Protocol) and FLUTE (File Delivery over Uni-directional Transport).

  A configuration example of a DASH segment that can be used when content stream distribution is performed in accordance with the HTTP streaming-based DASH standard using the FLUTE protocol will be described with reference to FIG.

As described above with reference to FIG. 2, the DASH segment is
(A) Initialization segment (Initialization Segment),
(B) Media segment,
There are two types.

(A) The initialization segment (initialization segment) is a segment that stores initialization data such as setting information necessary for executing content reproduction, such as setting of a decoder in the receiving device 30.
(B) A media segment is a segment that stores encoded content to be reproduced.

As shown in FIG. 3, (a) the initialization segment includes the following pieces of information.
(A1) Header information (dash) including segment file type information, etc.
(A2) metadata (moov) including initialization information such as codec (encoding mode) information of media data (mdat) that is encoded content transmitted by the media segment;

On the other hand, the (b) media segment includes the following pieces of information as shown in FIG.
(B1) Header information (msdh) including segment file type information, etc.
(B2) Access information (sidx) indicating boundary information of a plurality of sub-segments (Sub-Segment) stored in the media segment, random access points of media data (mdat) that is encoded content stored in the media segment, etc. ),
(B3) A plurality of sub-segments (Sub-Segment) 70,

In addition, the plurality of sub-segments (Sub-Segment) 70 includes one or a plurality of fragments (Fragments) 80.
The fragment 80 includes the following data.
Media data (mdat) that is encoded content to be played back,
Metadata (moof) corresponding to media data (mdat).

Note that (b) a random access point recorded in the access information (sidx) of the media segment is called SAP (Stream Access Point) in DASH.
The SAP indicates the start byte position of the start picture of the image sequence that can reset all the states necessary for decoding the stream, for example. Specifically, it is information indicating the position of an I picture of MPEG data, for example.

  In the metadata (moof) corresponding to the media data (mdat) stored in the fragment 80, for example, a presentation time or the like as the playback time information of the fragment storage media data (mdat) is recorded.

Currently, the media data (mdat) stored in one fragment 80 is normally set in a processing unit (chunk) of a content stream to be controlled by DASH. The processing unit (chunk) is, for example, GOP (Group of Pictures) which is a processing unit of MPEG (Moving Picture Expert Group) encoding.
In many cases, the GOP is operated by setting data with a reproduction time of about 0.5 to 2 seconds.

However, if the media data (mdat) stored in one fragment 80 is data of one GOP unit, a delay in data distribution and reproduction processing may occur, and real-time reproduction may be hindered.
For example, when a live relay image is distributed, the transmission device 20 performs encoding processing of live image data input from a camera and sequentially generates media segments 60. When the media data (mdat) stored in each fragment 80 of the media segment 60 is encoded data in GOP units, the transmission device 30 generates media data (mdat) that is encoded data in GOP units, and then Metadata (moof) describing attribute information such as presentation time of the generated media data (mdat) is generated.

Therefore, if each data is generated in this sequence, the GOP metadata (moof) can be generated only after the data range of the encoded data of each GOP unit is determined.
Therefore, the generation of the metadata (moof) is performed after the data range of the media data (mdat) is determined, and the generation process of each metadata (moof) is performed in one GOP unit time (0.5 to 2). It is necessary to wait for a time equivalent to (second).

  A fragment generation sequence when media data (mdat) stored in the fragment 80 constituting the media segment 60 is encoded data in GOP units will be described with reference to FIG.

The example shown in FIG. 4 is a sequence example when the media data (mdat) stored in the media segment 60 is stored in accordance with the MP4 file format (encoded format) defined in MPEG.
The data portion of the MP4 file format is divided into samples as basic storage units. Further, each sample is composed of one or more NAL units. The NAL unit is fragmented data of MPEG encoded data, for example, in units of slices.
As shown in FIG. 4A, encoded data corresponding to a plurality of GOPs is composed of a plurality of NAL units.

In FIG.
The upper row (a) NAL units that constitute the GOP. The lower row (b) shows the fragment generation processing sequence.
(B) The time axis is shown at the bottom of the fragment generation processing sequence. Time elapses from left to right, and the transmission apparatus 20 executes each process according to this time axis to generate a fragment.
The transmission device generates a media segment storing the generated fragment, generates an HTTP packet storing the media segment, generates an IP packet storing the HTTP packet, and then transmits the IP packet. .

The fragment generation processing sequence shown in FIG.
The transmission apparatus 20 generates a fragment storing encoded data according to the MP4 file format in the following sequence.
Time t0 to t1: Sample 1 (sample1) storing a plurality of NAL units constituting the GOP is generated.
Time t2 to t3: Sample 2 (sample2) storing a plurality of NAL units constituting the GOP is generated.
Time t4 to t5: Sample 3 (sample3) storing a plurality of NAL units constituting the GOP is generated.
Up to this point, the generation of the sample storing all the NAL units constituting one GOP is completed.
These samples 1 to 3 (samples 1 to 3) are set as media data (mdat) in the fragment of the media segment.

Time t6 to t7: Metadata (moof) that is attribute information of GOP encoded data stored in samples 1 to 3 (samples 1 to 3) is generated.
Time t8 to t9: A fragment is generated by combining media data (mdat) composed of samples 1 to 3 (samples 1 to 3) and metadata (moof) of samples 1 to 3 (samples 1 to 3).

Thereafter, the transmission device 20 generates a media segment including the fragment generated according to the above-described processing, generates a packet including the media segment as a payload, and transmits the packet to the reception device 30.
When generating a fragment, the transmission device 20 checks the reproduction time of media data (mdat) stored in the fragment in GOP units, and generates attribute information in GOP data units such as a presentation time according to the reproduction time. Therefore, it is necessary to record in the metadata (moof).

  Currently, image data to be distributed is mainly high-definition image data. However, in the future, it is expected that image quality will be further improved and data distribution with a large amount of data such as 4K images will increase. There is a possibility that the bit rate of the stream becomes large in order to cope with such large volume data distribution.

  The amount of data for each GOP unit increases as the amount of data increases due to higher image quality. Therefore, as described with reference to FIG. 4, when one GOP unit fragment is generated and transmitted sequentially, the processing interval on the transmission side becomes long. In addition, the amount of data per unit of transmission data increases. Therefore, when a sufficient communication band cannot be secured, the possibility of transmission delay in network transmission increases.

  Further, the packet reception interval on the receiving device side becomes longer, the amount of data received per packet increases, and the amount of data buffer required on the receiving device side also increases. Also, if retransmission processing is performed when a packet reception error occurs, the amount of delay increases rapidly, and there is a high possibility that real-time reproduction will break.

[3. Transmission data segmentation processing configuration]
In order to solve the above problem, a configuration example in which transmission data from the transmission device 20 to the reception device 30 is subdivided to reduce one transmission packet, that is, the amount of transmission data per unit will be described below.

  In the fragment generation sequence with reference to FIG. 4, the media data (mdat) to be set in the fragment is set as 1 GOP unit data, and the metadata (moof) corresponding to the 1 GOP unit media data (mdat) is generated. . That is, each metadata (moof) is metadata corresponding to media data (mdat) in units of 1 GOP.

  In the sequence shown in FIG. 4, the generation timing of metadata (moof) is determined according to the data amount of one GOP. Accordingly, when the amount of data included in one GOP increases, the generation timing of metadata (moof) is also delayed, and as a result, a delay occurs in fragment generation processing, media segment generation processing, and transmission packet generation processing. As a result, the amount of data per packet increases and the possibility of delivery delay increases.

  As a configuration for preventing such a situation, a configuration example in which media data (mdat) in a fragment is set to data obtained by subdividing 1 GOP instead of 1 GOP unit data will be described with reference to FIG.

  In FIG. 5, media data (mdat) stored in a fragment is not one GOP data but data obtained by subdividing one GOP data, specifically, one NAL unit or a plurality of pieces of data constituting one GOP. An example of a NAL unit is shown.

FIG. 5 (a) shows the NAL units constituting the GOP, as in FIG. 4 (a).
FIGS. 5B1 to 5B3 show a plurality of media segments in which NAL units that are constituent data of one GOP are distributed and stored as media data (mdat).
As shown in FIGS. 5 (b1) to (b3), a media segment storing media data (mdat) composed of one or more NAL units, which are subdivided data of GOP data, is hereinafter referred to as a sub GOP media segment. .

FIG. 5 shows an example in which one GOP data is stored in three sub GOP media segments, but the number of sub GOP media segments for storing one GOP data is set to an arbitrary number of 2 or more. Is possible.
In addition, the sub GOP media segments in FIGS. 5B1 to 5B3 are all examples in which media data (mdat) is set in a plurality of NAL units, but the media data (mdat) is only one NAL unit. It is good also as a setting.

  The transmission apparatus 20 generates sub GOP media segments shown in FIGS. 5B1 to 5B3, sets each of the generated sub GOP media segments as a payload of an individual HTTP packet, and transmits a network or broadcast wave. Send through.

The metadata (moof) stored in the sub GOP media segment in FIGS. 5B1 to 5B3 is metadata including attribute information corresponding to the media data (mdat) stored in each sub GOP media segment.
With this setting, the data amount of one packet transmitted via the network or broadcast wave is reduced, and the time required for one packet generation process on the transmission device side is shortened. In addition, the possibility of packet delay is reduced, and the retransmission processing delay in the event of packet loss or the like is also reduced. As a result, real-time reproduction without error in the receiving device 30 is realized.

A configuration example of an HTTP packet in which an HTTP header is set in a sub GOP media segment will be described with reference to FIG.
FIG. 6 shows two HTTP packet configuration examples.
The difference between the packets shown in FIGS. 6A and 6B is whether or not the packet has [sidx] for storing random access information and the like.
As described above with reference to FIG. 3, sidx is boundary information of a plurality of sub-segments (Sub-Segment) stored in the media segment, and media data (encoded content stored in the media segment ( mdat) random access point or the like. In DASH, the access information is called SAP (Stream Access Point). The SAP indicates the start byte position of the start picture of the image sequence that can reset all the states necessary for decoding the stream, for example. Specifically, for example, it corresponds to an I picture position of MPEG data.

The HTTP packet illustrated in FIG. 6B is a packet that does not include sidx. The HTTP packet shown in FIG. 6B does not include picture data serving as an access point in the media data (mdat) stored in this packet.
When the packet storage media data (mdat) does not include data serving as an access point, data indicating the access point is also unnecessary. Therefore, sidx is not set in the HTTP packet shown in FIG.
On the other hand, sidx is set in a packet having a sub GOP media segment storing media data (mdat) including data serving as an access point, as shown in FIG.

As described with reference to FIGS. 5 to 6, in this embodiment, one GOP configuration data is divided into a plurality of packets and transmitted.
The receiving device 30 sequentially receives the plurality of packets, and acquires GOP configuration data divided and stored in each packet. When performing decoding processing in units of GOPs, it is necessary to collect all GOP configuration data stored in a plurality of packets, arrange the GOP configuration data (NAL units) in the correct order, and reconfigure the GOP data.
Hereinafter, a setting example of additional information for smoothly executing processing on the receiving device side, such as information applied to the GOP reconfiguration processing, will be described.

[4. Example in which additional information is recorded in HTTP header]
First, an embodiment will be described in which additional information for smoothly executing processing on the receiving device side, such as information applied to GOP reconfiguration processing, is recorded in an HTTP header of an HTTP packet.

The recording information of the HTTP header will be described with reference to FIGS.
FIG. 7 shows a configuration example of an HTTP packet in which one GOP data is divided and stored in three sub GOP media segments, as described above with reference to FIG.

In the HTTP header of these three HTTP packets, as shown in FIG.
(1) Segment identifier (Content-Location)
(2) GOP internal position identification information (X- (Start / Middle / End) ofGOP)
These two pieces of identification information are recorded.

  (1) The segment identifier is data including the content position information of the segment stored in the HTTP packet, the type of the segment, and the identification information of the GOP to which the media data (mdat) stored in the packet belongs. Specifically, the position information of GOP data (access information such as URL) may be recorded.

  The receiving device 30 that receives the packet can determine that the HTTP packet in which the same segment identifier (Content-Location) is recorded is an HTTP packet in which media data (mdat) belonging to the same GOP is stored.

(2) GOP in-position identification information is data indicating in which position in one GOP the media data (mdat) stored in the HTTP packet is located.
The packet with GOP internal identification information = X-StartofGOP is a packet in which the NAL unit in the head area of the GOP data is stored as media data (mdat).
A packet with GOP in-line identification information = X-MiddleofGOP is a packet in which a NAL unit in an intermediate area of GOP data is stored as media data (mdat).
The packet with GOP in-line identification information = X-EndofGOP is a packet in which the NAL unit in the end area of the GOP data is stored as media data (mdat).

When one GOP data is divided into four or more sub GOP media segments, a plurality of packets in which GOP in-line identification information = X-MiddleofGOP is set are generated. The arrangement in the GOP data of the media data (mdat) stored in these packets can be determined by packet header information other than the HTTP header.
For example, it can be determined by referring to the packet sequence number recorded in the LCT header of the LCT packet storing the HTTP packet. A specific transmission packet configuration and an LCT header configuration will be described later.

  Therefore, for example, in a configuration in which an HTTP packet is stored in an LCT packet and transmitted, the intra-GOP position identification information set in the HTTP header may be set to identify only the packet storing the data in which the GOP position is the head area. That is, it is also possible to record only the GOP in-position identification information = X-StartofGOP, and the subsequent GOP data is arranged with reference to the sequence number of the LCT header.

FIG. 8 shows a data configuration example of the HTTP header of the HTTP packet storing the media segment.
As shown in FIG. 8, for example, the following HTTP header information is recorded in the HTTP header.
"...
HTTP / 1.1 206 Partial Content
Date: Fri, 04 Oct 2013 11:14:20 GMT
Content-type: application / mp4
Content-Location: http: // a. com / x. mp4
X-StartOfGOP
... "

In the above HTTP header information,
The segment identifier is
“Content-Location: http://a.com/x.mp4”
It is.
This segment identifier includes identification information of the GOP to which the media data (mdat) stored in the HTTP packet belongs. Specifically, it is position information (access information) of the GOP data.
It can be determined that an HTTP packet in which the same segment identifier (Content-Location) is recorded is an HTTP packet storing media data (mdat) belonging to the same GOP.

In the HTTP header information,
The GOP internal identification information is
"X-StartOfGOP"
It is. This GOP in-position identification information is data indicating where in one GOP the media data (mdat) stored in the HTTP packet is located.
The media data (mdat) stored in the HTTP packet is one GOP,
In the case of the top area, “X-StartOfGOP”,
In the case of an intermediate area, “X-MiddleOfGOP”,
In the case of the end region, “X-EndOfGOP”,
Any one of these data is recorded in the HTTP header as GOP in-line identification information.

Next, the recording information of the HTTP header of the HTTP packet storing the initialization segment will be described with reference to FIG.
Segment identification information indicating that the HTTP packet is a packet storing the initialization segment is recorded in the HTTP header of the HTTP packet storing the initialization segment.

FIG. 9 shows a data configuration example of the HTTP header of the HTTP packet storing the initialization segment.
As shown in FIG. 9, for example, the following HTTP header information is recorded in the HTTP header.
"...
HTTP / 1.1 206 Partial Content
Date: Fri, 04 Oct 2013 11:14:20 GMT
Content-type: application / mp4
Content-Location: http: // a. com / x-init. mp4 "
X-InitializationSegment
... "

In the above HTTP header information,
The segment identifier is
“Content-Location: http://a.com/x-init.mp4”
It is.
As the segment identifier, access information such as the URL of the initialization segment may be recorded.

In the above HTTP header information,
Segment identification information
"X-InitializationSegment"
It is.
This segment identification information is information indicating that the segment stored in the HTTP packet is an initialization segment.
This segment identification information is recorded in the HTTP header storing the initialization segment.

[5. Packet configuration]
Next, a configuration example of a packet transmitted from the transmission device to the reception device will be described.
FIG. 10 shows a configuration example of the following two IP packets.
(1) Initialization segment storage IP packet (2) Media segment storage IP packet

(1) The initialization segment storage IP packet has the following configuration.
IP header UDP header LCT header HTTP header Initialization segment configuration data [dash]
Initialization segment configuration data [moov]

  The IP header, UDP header, LCT header, and HTTP header are header information set in accordance with the communication protocols of the IP protocol, UDP protocol, FLUTE protocol, and HTTP protocol, respectively.

On the other hand, (2) the media segment storage IP packet has the following configuration.
IP header UDP header LCT header HTTP header Media segment configuration data [msdh]
Media segment configuration data [sidx]
Media segment configuration data [moof]
Media segment configuration data [mdat]

The IP header, UDP header, LCT header, and HTTP header are header information set in accordance with the communication protocols of the IP protocol, UDP protocol, FLUTE protocol, and HTTP protocol, respectively.
As described above, the media segment configuration data [sidx] is set for a packet having media data (mdat) applicable to random access, but need not be set for other packets.

The transmission device 20 generates the IP packet shown in FIG. 10 and transmits it to the reception device 30.
The receiving device 30 receives each packet shown in FIG. 10 received from the receiving device 20, analyzes each header information, extracts a segment, and executes initial setting and content reproduction according to the segment storage data.

[6. Configuration and processing of transmission device and reception device]
Next, the configuration and processing of the transmission device and the reception device will be described with reference to FIG.
First, the configuration and protocol stack of the transmission device 20 and the reception device 30 will be described with reference to FIG.

The transmission apparatus 20 that generates and transmits the IP packet described with reference to FIG. 10 includes a data processing unit 21 and a communication unit 22 as illustrated in FIG.
The data processing unit 21 executes transmission packet generation processing.
Specifically, for example, as described above, media data (mdat) including only a part of configuration data of GOP (Group of Pictures), which is a processing unit of encoded data, and metadata (moof) corresponding to media data are included. The stored sub GOP media segment is generated as packet storage data. Further, packet additional information such as an in-GOP position identifier indicating the position in the GOP of the media data which is the storage data of the sub GOP media segment is generated, and a packet recorded in the HTTP header is generated.

The communication unit 22 transmits the packet generated by the data processing unit 21.
The communication unit 22 broadcasts or multicasts a packet generated by the data processing unit 21 via a broadcast wave or a network such as the Internet.

The data processing unit 21 and the communication unit 22 execute a packet generation process according to a protocol stack including the following layers, and transmit the generated packet. Hereinafter, the layer configuration of the protocol stack of the transmission apparatus 20 will be described. The protocol stack of the transmission apparatus 20 has the following layer configuration from the upper layer to the lower layer.
(1) DASH Server: Application layer that executes processing such as segment generation according to the DASH standard (2) Video / Audio / Subtitle etc: Application layer that generates and acquires content to be transmitted (3) Fragmented MP4: MP4 file Application layer that generates encoded data according to the format and generates a segment (4) HTTP: Layer that generates an HTTP packet having an HTTP header according to the HTTP protocol (5) FLUTE / ALC (LCT): An LCT header according to the FLUTE protocol (6) UDP: Layer that generates a UDP packet having a UDP header according to the UDP protocol (7) IP Layer that generates an IP packet having an IP header in accordance with the IP protocol (8) PHY: IP packets or physical layer a communication unit or the like for generating and transmitting the MAC frame storing an IP packet,

In addition, the receiving device 30 that receives the IP packet described with reference to FIG. 10 includes a data processing unit 31 and a communication unit 32 as illustrated in FIG.
The communication unit 32 receives a packet transmitted from the transmission device 20, and the data processing unit 31 inputs the packet received by the communication unit 31 and performs data processing.

The data processing unit 31 and the communication unit 32 receive and analyze packets according to a protocol stack including the following layers. The protocol stack of the receiving device 30 has the following layer configuration from the upper layer to the lower layer.
(1) DASH Client: Application layer that executes processing such as segment analysis according to the DASH standard (2) Video / Audio / Subtitle etc: Application layer that executes reception content acquisition, playback processing, etc. (3) Fragmented MP4: (4) HTTP: Layer for analyzing HTTP packet having HTTP header according to HTTP protocol (5) FLUTE / ALC (LCT): LCT according to FLUTE protocol Layer for analyzing FLUTE packet having header (6) UDP: Layer for analyzing UDP packet having UDP header according to UDP protocol (7) IP: Layer (8) PHY for analyzing the IP packet having the IP header according to the P protocol: IP packets or physical layer a communication unit or the like for receiving a MAC frame storing an IP packet,

  Note that, as described above, when the IP device broadcasts or multicasts an IP packet, the transmission device 20 uses either a transmission via a network, a transmission via a broadcast wave, or both communication paths. To perform parallel transmission processing.

The receiving device 30 performs packet reception processing via a broadcast wave and any communication path of a network such as the Internet, or both communication paths.
Generation and analysis of transmission / reception packets via a network such as the Internet can be performed by replacing the FLUTE / ALC (LCT) layer and the UDP layer shown in FIG. 11 with TCP layers.

  An example of a protocol stack of the receiving apparatus 30 that performs processing by appropriately switching between an IP packet received via a broadcast wave and an IP packet received via a network such as the Internet will be described with reference to FIG.

The protocol stack of the receiving device 30 shown in FIG. 12 shows the layer configuration of the protocol stack corresponding to the following two communication systems.
(1) Broadcast system (2) Network communication system

(1) The broadcasting system has the layer configuration described with reference to FIG.
(2) The network communication system has a configuration in which the FLUTE / ALC (LCT) layer of the broadcast system layer and the UDP layer are replaced with a TCP layer.
The TCP layer performs analysis of a TCP packet having a TCP header.
The signaling layer is a layer for performing switching control of each communication system.

  The receiving device 30 selectively uses a packet received via a broadcast wave and a packet received via a network such as the Internet by appropriately switching and using each layer of the broadcast system and the network communication system. Thus, the packet storage content can be acquired and the playback process can be executed.

For example, when a delay occurs in packet reception from the network, it is possible to switch to the broadcast system, receive a packet corresponding to the same content via a broadcast wave, and continue content reproduction.
The segment identifier and the intra-GOP location identifier described above are recorded in the delivery packet via both the network communication system and the broadcast system, and the GOP data can be reconstructed by referring to these identification information. Thus, error-free decoding processing and content reproduction are realized.

[7. Concerning a configuration in which access point information that can identify whether or not a NAL unit stored in a packet includes an SAP (stream access point) is added]
Next, as a configuration example for realizing the processing efficiency improvement of the receiving apparatus, identification information that makes it possible to identify whether the NAL unit that is the division source of the NAL unit fragment stored in the packet includes SAP (access point data) or not is provided. The added configuration will be described.

As described above with reference to FIG. 3 and others, SAP (Stream Access Point) is storage location information of data serving as a random access point. In DASH, a random access point is called SAP (Stream Access Point). The SAP indicates the start byte position of the start picture of the image sequence that can reset all the states necessary for decoding the stream, for example. Specifically, it is information indicating the I picture position of MPEG data, for example.
As described above with reference to FIG. 3B, the SAP is stored in [sidx] that is meta information of the media segment.

On the receiving device side, it is possible to acquire encoded data at the data position indicated by SAP, perform decoding processing from the acquired data, and perform reproduction.
Therefore, when special reproduction processing such as reproduction from the middle of content is performed, SAP becomes important indispensable data. For example, an image of a random access point can be reproduced by acquiring and decoding encoded data specified by SAP without arranging all of the GOP data.

  In addition, the receiving apparatus can also preferentially process random access point data specified by the SAP, for example, when reception of all encoded data packets in units of GOP is not in time due to distribution delay or the like. By selecting random access point data and decoding and reproducing it, the image display on the display unit can be continued. As described above, the SAP is important data when determining the processing priority.

  A random access point is surely present in one GOP unit data. However, as described with reference to FIGS. 5 to 10, the storage data of one packet is not data of 1 GOP unit, but a packet in which 1 GOP is subdivided and one or a plurality of NAL units are stored is a packet. The NAL unit stored in may or may not include a random access point.

In the configuration described with reference to FIGS. 5 to 10, for example, as illustrated in FIG. 7, a packet including metadata “sidx” and a packet not including the metadata are set in the sub GOP media segment HTTP packet.
If the NAL unit stored in the sub GOP media segment HTTP packet includes data corresponding to a random access point, [sidx] in which SAP is recorded is set.
However, [sidx] is metadata including not only SAP, which is random access point information of media data, but also boundary information of other data, etc., and it becomes an access point to packet storage data only by the presence / absence of [sidx]. It cannot be determined whether or not data is included.

Therefore, in the encoded data distribution configuration using the above-described sub GOP media segment HTTP packet, it is difficult for the receiving apparatus side to determine whether or not the access point data is included in the packet in units of received packets.
Hereinafter, a configuration will be described in which such a problem is solved, and the reception apparatus can determine whether or not the access point data is included in the packet for each received packet.

With reference to FIG. 13 and subsequent figures, a description will be given of a configuration in which it is possible to determine whether or not access point data is included in each packet in a packet unit in the above-described encoded data distribution configuration using the sub GOP media segment HTTP packet. .
FIG. 13 shows an example of the configuration of an HTTP packet in which one GOP data is divided and stored in three sub GOP media segments, similar to FIGS. 5 and 7 described above.

The following additional information described above with reference to FIG. 7 is recorded in the HTTP headers of these three HTTP packets.
(1) Segment identifier (Content-Location)
(2) GOP internal position identification information (X- (Start / Middle / End) ofGOP)
Furthermore, in the example shown in FIG. 13, the following additional information is additionally recorded in the HTTP header.
(3) Access point information (X-SAP)
These three pieces of information are recorded.

  (1) The segment identifier is data including the content position information of the segment stored in the HTTP packet, the type of the segment, and the identification information of the GOP to which the media data (mdat) stored in the packet belongs. Specifically, the position information of GOP data (access information such as URL) may be recorded.

  The receiving device 30 that receives the packet can determine that the HTTP packet in which the same segment identifier (Content-Location) is recorded is an HTTP packet in which media data (mdat) belonging to the same GOP is stored.

(2) GOP in-position identification information is data indicating in which position in one GOP the media data (mdat) stored in the HTTP packet is located.
The packet with GOP internal identification information = X-StartofGOP is a packet in which the NAL unit in the head area of the GOP data is stored as media data (mdat).
A packet with GOP in-line identification information = X-MiddleofGOP is a packet in which a NAL unit in an intermediate area of GOP data is stored as media data (mdat).
The packet with GOP in-line identification information = X-EndofGOP is a packet in which the NAL unit in the end area of the GOP data is stored as media data (mdat).

When one GOP data is divided into four or more sub GOP media segments, a plurality of packets in which GOP in-line identification information = X-MiddleofGOP is set are generated. The arrangement in the GOP data of the media data (mdat) stored in these packets can be determined by packet header information other than the HTTP header.
For example, it can be determined by referring to the packet sequence number recorded in the LCT header of the LCT packet storing the HTTP packet. A specific transmission packet configuration and an LCT header configuration will be described later.

  Therefore, for example, in a configuration in which an HTTP packet is stored in an LCT packet and transmitted, the intra-GOP position identification information set in the HTTP header may be set to identify only the packet storing the data in which the GOP position is the head area. That is, it is also possible to record only the GOP in-position identification information = X-StartofGOP, and the subsequent GOP data is arranged with reference to the sequence number of the LCT header.

(3) The access point information (X-SAP) is information indicating whether or not encoded data included in the media data (mdat) stored in the HTTP packet includes data serving as a random access point.
By referring to the access point information (X-SAP), the receiving apparatus can immediately determine whether or not the NAL unit in the media data stored in the HTTP packet includes access point data.
Therefore, for example, when decoding and reproduction are performed with priority given only to access point data, the processing target packet can be selected with reference to the access point information (X-SAP) recorded in the HTTP header. That is, the access point information (X-SAP) recorded in the HTTP header can be processed by selecting only a packet indicating that the access point is included in the packet storage data (NAL unit).

In the example shown in FIG. 13, the access point information (X-SAP) is set in all HTTP packets in FIGS. 13 (b1) to (b3), and the access point information (X-SAP) is set in each HTTP packet. The encoded data included in the stored media data (mdat) is configured to indicate whether data serving as a random access point is included.
In addition to such a configuration, for example, only when encoded data included in media data (mdat) stored in an HTTP packet includes data serving as a random access point, access point information is included in the HTTP header of the HTTP packet. (X-SAP) may be recorded. That is, when the encoded data included in the media data (mdat) stored in the HTTP packet does not include data serving as a random access point, the access point information (X-SAP) is not recorded in the HTTP header. For example, when data serving as a random access point is not included in the HTTP packet shown in FIGS. 13B2 and 13B3, the access point information (X-SAP) is not recorded in the HTTP header.
In the case of this setting, the receiving apparatus randomly selects the encoded data included in the media data (mdat) stored in the HTTP packet according to whether or not the access point information (X-SAP) is recorded in the HTTP header. It is determined whether data serving as an access point is included.

  This access point information (X-SAP) is information that makes it possible to identify the presence / absence of randomly accessible data in units of packets. On the receiving device side, it is possible to select packets storing randomly accessible data with reference to the access point information (X-SAP) of each packet. For example, the receiving apparatus can perform a data reproduction process from a random access point quickly by preferentially processing a packet storing randomly accessible data.

FIG. 14 shows a data configuration example of an HTTP header of an HTTP packet storing a media segment.
As shown in FIG. 14, for example, the following HTTP header information is recorded in the HTTP header.
"...
HTTP / 1.1 206 Partial Content
Date: Fri, 04 Oct 2013 11:14:20 GMT
Content-type: application / mp4
Content-Location: http: // a. com / x. mp4
X-StartOfGOP
X-SAP
... "

In the above HTTP header information,
Segment identifier “Content-Location: http://a.com/x.mp4”
GOP internal identification information “X-StartOfGOP”,
These are the same information as described above with reference to FIG.
That is, the segment identifier is position information (access information) of GOP data that functions as identification information of the GOP to which the media data (mdat) stored in the HTTP packet belongs.
The GOP in-position identification information is data indicating where in one GOP the media data (mdat) stored in the HTTP packet is located.
In the case of the top area, “X-StartOfGOP”,
In the case of an intermediate area, “X-MiddleOfGOP”,
In the case of the end region, “X-EndOfGOP”,
Any one of these data is recorded in the HTTP header as GOP in-line identification information.

In the HTTP header information shown in FIG.
The access point information “X-SAP” is information indicating whether or not the encoded data (NAL) included in the media data (mdat) stored in the HTTP packet includes data serving as a random access point.

  By referring to the access point information (X-SAP), the receiving apparatus can immediately determine whether or not the access point is included in the NAL unit in the packet storage media data for each received packet.

  The receiving apparatus can select only the packets including the access points, decode the data stored in these packets with priority, and reproduce them. For example, when packet reception of all encoded data in units of GOP is not in time due to distribution delay, etc., it is possible to perform processing such as continuing image display on the display unit by preferentially decoding and reproducing random access point data. Become.

[8. Communication processing configuration using NAL unit fragment storage packet obtained by further dividing NAL unit]
As described above, the data amount of one transfer packet on the network can be reduced by performing data transmission / reception using one NAL unit which is GOP divided data or a packet including a plurality of NAL units. .

  However, for image data with a large amount of data, such as 4K content or 8K content, which is high-quality image data, for example, the data amount of the NAL unit generated by the encoding process of these large-capacity image data increases. That is, it is assumed that the data amount of one NAL unit itself becomes very large.

In data transfer processing in the IP layer, for example, when performing data transfer via Ethernet (registered trademark), an IP packet is a MAC frame equal to or less than the maximum data transfer unit (MTU: Maximum Transfer Unit) defined by Ethernet (registered trademark). Must be generated and transferred.
That is, when a packet is transferred through a network in which the maximum data transfer unit (MTU) is specified, an IP packet having a data amount greater than or equal to the MTU is subjected to fragment processing for dividing it into a data amount that is specified by the MTU or less. A process of generating and transferring a plurality of MAC frames storing each divided data is essential.

For example, the maximum data transfer amount (MTU) of each MAC frame defined by Ethernet (registered trademark) is about 1500 bytes.
The MAC frame has a configuration shown in FIG. 15, for example, and has a configuration in which a MAC header is set at the head of the IP packet described above with reference to FIG. For example, in the MAC frame shown in FIG. 15, it is necessary to set the data amount from the IP header to mdat, which is the payload of the MAC frame, to MTU = 1500 bytes or less.

When packet transmission is performed from the transmission device 20 to the reception device 30, fragmentation corresponding to the MTU is performed in communication between devices on the transmission device 20 or the reception device 30 side or in a relay device between the transmission device 20 and the reception device 30. That is, there is a possibility that the packet division process and the divided data reconstruction process are repeatedly executed.
When such a situation occurs, a delay occurs in the time from content input in the transmission device 20 to content reproduction in the reception device 30, and the possibility of content reproduction delay in the reception device 30 increases.

Below, the structure which prevents such a situation is demonstrated.
In the embodiment described below, the transmission device 20 sets the data size of the packet to a predetermined size or less in the HTTP layer in the packet generation processing stage. Specifically, the data size is smaller than the maximum data transfer unit (MTU: Maximum Transfer Unit) assumed in the communication path. Specifically, for example, a packet storing a NAL unit fragment obtained by further dividing one NAL unit is generated and transmitted.

As described above, for example, a general maximum data transfer amount (MTU) of a frame unit of a MAC frame defined by Ethernet (registered trademark) is about 1500 bytes.
The data processing unit of the transmission device 20 generates a packet in consideration of the MTU at the generation stage of the HTTP packet. For example, the packet generation is performed so that the frame size of the MAC frame storing the HTTP packet is 1500 bytes or less.

  The packet processing according to the MTU is performed in the IP layer of the transmission device, the relay device, or the reception device by executing packet generation processing in consideration of the maximum data transfer amount (MTU) in the communication path in the data processing unit of the transmission device 20 Fragmentation, which is a division process, is not necessary, and data transfer can be executed smoothly.

  As described above, the data processing unit of the transmission device 20 executes a process of setting the packet storage data to a small fragment equal to or smaller than the MTU defined in the communication path at the generation stage of the HTTP packet. By this processing, for example, overhead due to fragmentation processing in the IP layer of the communication path is reduced.

  In the example described above with reference to FIG. 5 and others, one or a plurality of NAL units are set as media data (mdat), and further combined with metadata (moof) that is attribute data thereof, one fragment is generated, An HTTP packet storing a sub GOP media segment having this fragment was generated.

In the embodiment described below, an HTTP packet in which a NAL unit fragment (NALf) obtained by dividing one NAL unit is stored as media data (mdat) is generated.
The attribute data (moof) corresponding to the NAL unit is stored and distributed in another independent HTTP packet different from the media data (mdat) storage packet.

With reference to FIG. 16, the structural example of the HTTP packet in a present Example is demonstrated.
FIG. 16A shows the sub GOP media segment described above with reference to FIGS. That is, it is a segment in which one or more NAL units, which are data obtained by dividing GOP, are stored as media data (mdat).
However, the sub GOP media segment shown in FIG. 16A is an example in which only one NAL unit is stored as media data (mdat). The number of NAL units is not limited to 1 and may be a plurality of NAL units.

  In the present embodiment, the sub GOP media segment as shown in FIG. 16A is further divided to generate a plurality of HTTP packets. The division process is executed in consideration of the maximum data transfer amount (MTU) in the communication path. For example, when a MAC frame storing a divided HTTP packet is generated, the MAC frame is set to be equal to or less than a maximum data transfer unit (MTU: Maximum Transfer Unit) allowed for the MAC frame.

In the division process, a NAL unit division process for dividing one or a plurality of NAL units is performed. Hereinafter, the data after the division of the NAL unit will be described as a NAL unit fragment (NALf).
In addition, an HTTP packet generated by dividing the configuration data of the sub GOP media segment is referred to as a NAL unit fragment compatible HTTP packet.

In the example illustrated in FIG. 16, the configuration data of the sub GOP media segment illustrated in FIG. 16A is divided into four HTTP packets (b1) to (b4).
Each of the four NAL unit fragment-corresponding HTTP packets shown in (b1) to (b4) has a data amount other than the MAC header of 1500 bytes or less when the HTTP packet has the MAC frame configuration shown in FIG. Set the amount of data as follows.

  The HTTP packet corresponding to the NAL unit fragment shown in (b1) of FIG. 16 is the metadata area of the sub GOP media segment shown in (a), that is, the msdh (type), sidx, moof, and HTTP packets storing these metadata. It is. That is, it is a metadata storage type NAL unit fragment compatible HTTP packet.

  Also, the HTTP packets corresponding to the NAL unit fragments shown in FIGS. 16B2 to 16B4 are generated by dividing the NAL unit in the media data (mdat) area of the sub GOP media segment shown in FIG. ) Is stored in the HTTP packet. That is, it is a media data storage type NAL unit fragment compatible HTTP packet.

  Note that only one NAL unit is stored in the sub GOP media segment shown in FIG. 16 (a), but one NAL unit is stored in the sub GOP media segment as shown in FIG. 16 (a). It may be a unit, or it may be two or more NAL units. The configuration data of the NAL unit fragment is not limited to one NAL unit, and may be data extending over a plurality of NAL units. For example, a configuration in which one NAL unit fragment is set by combining the latter half of the preceding NAL unit and the beginning of the subsequent NAL unit is also possible.

  The transmission device 20 divides one sub GOP media segment in this way to generate a plurality of HTTP packets, further generates IP packets from these HTTP packets, and sequentially transmits them.

However, when one original NAL unit and its attribute information are divided and distributed in this way, it is necessary to reconstruct the NAL unit and its attribute information in order to perform decoding and reproduction processing on the receiving device 30 side. There is.
Information necessary for this is recorded in the HTTP header as additional information.

The following additional information is recorded in the HTTP header of the metadata storing type NAL unit fragment correspondence HTTP packet shown in FIG.
* NAL unit fragment header [X-NALUFragmentSubSegmentHeader]
* Movie fragment sequence number [X-MovieFragmentSequenceNumber]
* Access point information [X-SAP]

The NAL unit fragment header contains the HTTP packet
Whether the metadata storage type NAL unit fragment compatible HTTP packet,
Whether it is a media data storage type NAL unit fragment compatible HTTP packet,
This is identification information for identifying.

  The movie fragment sequence number is a sequence number recorded in the metadata (moof) of the sub GOP media segment before the division, and is sequence information of the NAL unit stored in the sub GOP media segment. As this sequence number, the sequence number recorded in the metadata (moof) of the sub GOP media segment before division is copied and recorded as it is.

The access point information indicates whether data serving as a random access point is included in the NAL unit stored in the sub GOP segment (sub GOP media segment in FIG. 16A) from which the HTTP packet is generated. Information.
By referring to this access point information (X-SAP), the receiving apparatus determines whether the HTTP packet is data generated based on the configuration data of the sub GOP segment including the NAL unit serving as a random access point. Can be immediately determined.

  It should be noted that the access point information (X-SAP) recorded in the HTTP header of the metadata storing type NAL unit fragment compatible HTTP packet of FIG. 16 (b1) includes the access point data in the sub GOP segment of the packet generation source. , By referring to [sidx] that is metadata, it is possible to acquire the position of the encoded data of the SAP, that is, the random access point.

On the other hand, the following additional information is recorded in the HTTP header of the media data storage type NAL unit fragment compatible HTTP packet shown in FIGS. 16 (b2) to (b4).
* Movie fragment sequence number [X-MovieFragmentSequenceNumber]
* Position identification information in the NAL unit [X-NALUFragmentIndicator]
* Access point information [X-SAP]

  As described above, the movie fragment sequence number is a sequence number recorded in the metadata (moof) of the sub GOP media segment before the division, and is sequence information of the NAL unit stored in the sub GOP media segment. As this sequence number, the sequence number recorded in the metadata (moof) of the sub GOP media segment before division is copied and recorded as it is.

The intra-NAL unit position identification information is information indicating the position of the NAL unit fragment (NALf) stored in the packet corresponding to the position of the NAL unit before division.
A packet with NAL unit position identification information = X-StartNALUFragmentSubSegment is a packet in which the NAL unit fragment (NALf) of the head area of the NAL unit before division is stored as media data (mdat).
The packet with NAL unit position identification information = X-Middle NALU Fragment SubSegment is a packet in which the NAL unit fragment (NALf) in the intermediate area of the NAL unit before division is stored as media data (mdat).
A packet with NAL unit position identification information = X-EndNALUFragmentSubSegment is a packet in which the NAL unit fragment (NALf) of the end area of the NAL unit before division is stored as media data (mdat).

When a NAL unit stored in one sub GOP media segment is divided into four or more media data storage type NAL unit fragment compatible HTTP packets, a plurality of NAL unit location identification information = X-Middle NALUFragmentSubSegment is set Will be generated. The arrangement of NAL unit fragments (NALf) of media data (mdat) stored in these packets can be determined by packet header information other than the HTTP header.
For example, it can be determined by referring to the packet sequence number recorded in the LCT header of the LCT packet storing the HTTP packet. A specific transmission packet configuration and an LCT header configuration will be described later.

  Therefore, for example, in a configuration in which an HTTP packet is stored in an LCT packet and transmitted, the intra-NAL unit position identification information set in the HTTP header may be set so as to identify only the packet storing the data in the head area. That is, only NAL unit position identification information = X-StartNALUFragmentSubSegment may be recorded, and subsequent data may be arranged with reference to the sequence number of the LCT header.

The access point information indicates whether data serving as a random access point is included in the NAL unit stored in the sub GOP segment (sub GOP media segment in FIG. 16A) from which the HTTP packet is generated. Information.
By referring to this access point information (X-SAP), the receiving apparatus determines whether the HTTP packet is data generated based on the configuration data of the sub GOP segment including the NAL unit serving as a random access point. Can be immediately determined.

FIG. 17 shows a configuration of an HTTP packet corresponding to a metadata storage type NAL unit fragment and a data configuration example of an HTTP header.
As shown in FIG. 17, for example, the following HTTP header information is recorded in the HTTP header.
"...
HTTP / 1.1 206 Partial Content
Date: Fri, 04 Oct 2013 11:14:20 GMT
Content-type: application / mp4
X-NALUFragmentSubSegmentHeader
X-MovieFragmentSequenceNumber: 234567
X-SAP
... "

In the above HTTP header information,
The NAL unit fragment header is
"X-NALUFragmentSubSegmentHeader"
It is.
The NAL unit fragment header is recorded as information indicating that the HTTP packet is a metadata storage type NAL unit fragment compatible HTTP packet.

In the above HTTP header information,
The movie fragment sequence number is
"X-MovieFragmentSequenceNumber"
It is.
As described above, this movie fragment sequence number is a sequence number recorded in the metadata (moof) of the sub GOP media segment before division, and is the sequence information of the NAL unit stored in the sub GOP media segment. As this sequence number, the sequence number recorded in the metadata (moof) of the sub GOP media segment before division is copied and recorded as it is.

In the above HTTP header information,
Access point information
"X-SAP"
It is.
This access point information is information indicating whether or not data serving as a random access point is included in the NAL unit stored in the sub GOP segment from which the HTTP packet is generated.

Next, a configuration of a media data storage type NAL unit fragment compatible HTTP packet for storing media data and a data configuration example of an HTTP header will be described with reference to FIG.
As shown in FIG. 18, for example, the following HTTP header information is recorded in the HTTP header.
"...
HTTP / 1.1 206 Partial Content
Date: Fri, 04 Oct 2013 11:14:20 GMT
Content-type: application / mp4
X-StartOfGOP
X-MovieFragmentSequenceNumber: 234567
X-StartNALUFragmentSubSegment
X-SAP
... "

In the above HTTP header information,
The GOP internal identification information is
"X-StartOfGOP"
It is. This GOP in-position identification information is data indicating where in one GOP the media data (mdat) stored in the HTTP packet is located.
In the case of the top area, “X-StartOfGOP”,
In the case of an intermediate area, “X-MiddleOfGOP”,
In the case of the end region, “X-EndOfGOP”,
Any one of these data is recorded in the HTTP header as GOP in-line identification information.

In the above HTTP header information,
The movie fragment sequence number is
"X-MovieFragmentSequenceNumber"
It is.
As described above, this movie fragment sequence number is a sequence number recorded in the metadata (moof) of the sub GOP media segment before division, and is the sequence information of the NAL unit stored in the sub GOP media segment. As this sequence number, the sequence number recorded in the metadata (moof) of the sub GOP media segment before division is copied and recorded as it is.

In the above HTTP header information,
The position identification information in the NAL unit is
"X-StartNALUFragmentSubSegment"
It is.
This intra-NAL unit position identification information is information indicating to which position of the NAL unit before the division the NAL unit fragment (NALf) stored in the packet.

A packet with NAL unit position identification information = X-StartNALUFragmentSubSegment is a packet in which the NAL unit fragment (NALf) of the head area of the NAL unit before division is stored as media data (mdat).
The packet with NAL unit position identification information = X-Middle NALU Fragment SubSegment is a packet in which the NAL unit fragment (NALf) in the intermediate area of the NAL unit before division is stored as media data (mdat).
A packet with NAL unit position identification information = X-EndNALUFragmentSubSegment is a packet in which the NAL unit fragment (NALf) of the end area of the NAL unit before division is stored as media data (mdat).

In the above HTTP header information,
Access point information
"X-SAP"
It is.
This access point information is information indicating whether or not data serving as a random access point is included in the NAL unit stored in the sub GOP segment from which the HTTP packet is generated.

[9. Processing sequence of transmitting device and receiving device]
Next, a processing sequence executed by the transmission device and the reception device will be described with reference to the flowcharts shown in FIGS.

First, the processing sequence executed by the transmission apparatus 20 will be described with reference to the flowchart shown in FIG.
The flow shown in FIG. 19 is the metadata storage type NAL unit fragment corresponding HTTP packet generated based on the sub GOP media segment shown in FIG. 16A described with reference to FIG. 16, or the media data storage type NAL unit. It is a flow explaining the generation | occurrence | production of a fragment corresponding | compatible HTTP packet, and the sequence of a transmission process.
This processing is executed in the data processing unit of the transmission device 20. The data processing unit includes a CPU having a program execution function, and executes processing according to the flow shown in FIG. 19 according to, for example, a program stored in the storage unit.
Hereinafter, the processing of each step will be described sequentially.

(Step S101)
First, the data processing unit of the transmission device executes encoding processing of content to be transmitted. For example, an encoding process according to the MP4 file format is executed.

(Step S102)
Next, the transmission apparatus generates media data (mdat) corresponding to the sub GOP media segment. This process is a process for generating media data (mdat) to be stored in the sub GOP media segment described above with reference to FIG. Media data (mdat) including a part of GOP data which is MP4 encoded data, that is, one or more NAL units constituting the GOP, is generated.

(Step S103)
Next, the transmission apparatus generates metadata (moof) that is attribute information corresponding to the media data (mdat) generated in step S102.

(Step S104)
Next, the transmission apparatus generates msdh and sidx, which are metadata corresponding to the sub GOP media segment. This metadata is stored in the sub GOP media segment described above with reference to FIG.

(Step S105)
Next, the transmission apparatus divides the metadata (mdat) for storing the sub GOP media segment to generate a NAL unit fragment (Nalf). For example, according to the general MTU of the MAC frame, the division process is performed such that the payload when the MAC frame is used is about 1500 bytes or less.

(Step S106)
Next, the transmission apparatus generates a metadata storage type NAL unit fragment corresponding HTTP packet and a packet header of the media data storage type NAL unit fragment corresponding HTTP packet storing the NAL unit fragment.
This is the HTTP header information generation process described above with reference to FIGS. 16, 17, and 18.

The following information is recorded in the HTTP header of the metadata storing type NAL unit fragment compatible HTTP packet.
First, a NAL unit fragment header indicating that this HTTP packet is a metadata storing type NAL unit fragment compatible HTTP packet is recorded.
Furthermore, a movie fragment sequence number corresponding to the arrangement information of the NAL unit stored in the sub GOP media segment is recorded. As this sequence number, the sequence number recorded in the metadata (moof) of the sub GOP media segment before division is copied and recorded as it is.
Further, access point information “X-SAP” is recorded.
This access point information is information indicating whether or not data serving as a random access point is included in the NAL unit stored in the sub GOP segment from which the HTTP packet is generated.

On the other hand, the following information is recorded in the packet header of the media data storage type NAL unit fragment compatible HTTP packet.
GOP in-position identification information: data indicating where in one GOP the media data (mdat) stored in the HTTP packet is located,
In the case of the top area, “X-StartOfGOP”,
In the case of an intermediate area, “X-MiddleOfGOP”,
In the case of the end region, “X-EndOfGOP”,
Record any of these data.

  Movie fragment sequence number: As described above, this is the sequence number recorded in the metadata (moof) of the sub GOP media segment before division, and is the sequence information of the NAL unit stored in the sub GOP media segment. As this sequence number, the sequence number recorded in the metadata (moof) of the sub GOP media segment before division is copied and recorded as it is.

  In-NAL unit position identification information: Information indicating which position of the NAL unit fragment (NALf) stored in the packet corresponds to the position of the NAL unit before division.

  Access point information “X-SAP”: The access point information is information indicating whether data serving as a random access point is included in the NAL unit stored in the sub GOP segment from which the HTTP packet is generated. .

  In step S106, the transmission apparatus generates a packet header of the metadata storage type NAL unit fragment correspondence HTTP packet and the media data storage type NAL unit fragment correspondence HTTP packet storing the NAL unit fragment as described above.

(Step S107)
Next, the transmission apparatus generates the following HTTP packet in which the HTTP header generated in step S106 is set in step S107.
Metadata storing type NAL unit fragment corresponding HTTP packet Media data storing type NAL unit fragment corresponding HTTP packet Each of these HTTP packets is generated.

(Steps S108 to S109)
Next, the transmission device sets an LCT header, a UDP header, and an IP header for the generated HTTP packet, and generates and transmits an IP packet. The transmission processing is executed via either a communication network such as the Internet or broadcast waves, or both communication paths.

  Note that the flow shown in FIG. 19 is a flow for explaining the generation of a packet generated based on the media segment and the transmission processing sequence. When generating a packet based on the initialization segment, after generating the configuration data of the initialization segment, a process of recording a segment identifier indicating the initialization segment in the HTTP header is performed.

Next, a processing sequence from packet reception to content reproduction executed in the receiving apparatus will be described with reference to flowcharts shown in FIGS.
This process is executed in the data processing unit of the receiving device 30. The data processing unit includes a CPU having a program execution function, and executes processing according to the flow shown in FIGS. 20 to 21 according to a program stored in the storage unit, for example.
Hereinafter, the processing of each step will be described sequentially.

(Step S201)
First, in step S201, the receiving apparatus inputs reproduction content designation information by the user. For example, a content list such as a program guide received in advance from the transmission device is displayed on the display unit, and the playback content is determined based on user input information for this display information.

(Step S202)
Next, the receiving device receives the packet including the initialization segment corresponding to the selected content selected as the reproduction target, and acquires the initialization segment.
As described above with reference to FIG. 9, segment identification information is recorded in the HTTP header of the HTTP packet storing the initialization segment, and the initialization segment is stored by referring to this identification information. It can be confirmed that this is an HTTP packet.

(Step S203)
The receiving device executes initialization processing of the receiving device according to the received storage data of the initialization segment. Specifically, the codec setting parameters and the like stored in the initialization segment are acquired, and codec settings and the like are performed according to the acquisition parameters.

(Step S204)
Next, the receiving apparatus receives the following packet, which is a divided packet of the sub GOP media segment corresponding to the selected content selected in step S201.
An IP packet storing an HTTP packet corresponding to a metadata storage type NAL unit fragment;
IP packet storing media data storage type NAL unit fragment compatible HTTP packet,
Each of these packets is received.

(Step S205)
Next, the receiving apparatus acquires the following attribute information of the data stored in the packet from the HTTP header of the received packet.
* NAL unit fragment header * GOP position identification information * Movie fragment sequence number * NAL unit position identification information * Access point information

As described above, the NAL unit fragment header is information indicating that the HTTP packet is a metadata storage type NAL unit fragment compatible HTTP packet.
The GOP in-position identification information is data indicating where in one GOP the media data (mdat) stored in the HTTP packet is located.
The movie fragment sequence number is a sequence number recorded in the metadata (moof) of the sub GOP media segment before the division, and is sequence information of the NAL unit stored in the sub GOP media segment.
The intra-NAL unit position identification information is information indicating the position of the NAL unit fragment (NALf) stored in the packet corresponding to the position of the NAL unit before division.
The access point information “X-SAP” is information indicating whether or not data serving as a random access point is included in the NAL unit stored in the sub-GOP segment from which the HTTP packet is generated.

(Step S206)
Next, in step S206, the receiving apparatus determines whether or not to execute special reproduction in which access point data is preferentially decoded and reproduced.
For example, it is determined whether or not to execute special playback processing such as playback processing from the middle of content such as random access playback processing. This determination process is determined based on user input information or the like.
When the special reproduction process is executed, the process proceeds to step S207. When the normal reproduction process is executed, the process proceeds to step S208.

(Step S207)
If it is determined in step S206 that special reproduction processing such as random access reproduction processing is to be executed, the process proceeds to step S207. In step S207, the access point information [X-SAP] recorded in the HTTP header of the packet selects a packet indicating data with an access point. That is, a packet indicating that the NAL unit stored in the sub GOP segment from which the HTTP packet is generated includes data serving as a random access point is selected as a decoding target, the decoding process is executed, and the reproduction process is performed. Do.
In this case, decoding and reproduction can be performed without arranging the entire GOP.
It progresses to step S211 after the process of step S207.

(Step S208)
On the other hand, if it is determined in step S206 that the special reproduction process such as the random access reproduction process is not performed and the normal reproduction process is performed, the process proceeds to step S208. In step S208, the receiving apparatus rearranges the NAL unit fragments stored in the received plurality of media data storage type NAL unit fragment corresponding HTTP packets according to the identification information acquired from the HTTP header, and reconstructs GOP unit data.

(Steps S209 to S210)
Next, the receiving apparatus performs a decoding process on the reconstructed GOP data, and performs a reproduction process of the decoded data.

(Step S211)
Next, the receiving apparatus determines whether or not the data reproduction process has been completed. If the data reproduction process has not been completed, the receiving apparatus returns to step S204, and repeatedly executes the processes from step S204.
If it is determined in step S211 that the reproduction process has ended, the process ends.

For example, when only an image at a specific reproduction position is reproduced, such as random access reproduction, the metadata storage type NAL unit fragment compatible HTTP packet is referred to sidx data that is random access point information and is calculated from the sidx data. Only one or more NAL unit fragments (NALf) can be processed. That is, for example, only encoded image data corresponding to an I picture is selected, and a process of decoding and reproducing this is executed.
In this case, it is possible to select and reproduce only the data of the random access point without performing the process of rearranging the entire GOP data.

[10. Example in which additional information is recorded in extension header]
In the above-described embodiment, additional information for smoothly executing processing on the receiving apparatus side, such as information applied to NAL unit fragment (NALf) arrangement, GOP reconfiguration processing, and access point information, is set in the HTTP packet. Examples have been described.
However, the recording destination of the additional information is not limited to the HTTP header, and various recording destinations can be used.
Hereinafter, an embodiment in which additional information is recorded in the extension header will be described.

With reference to FIGS. 22 and 23, an embodiment in which additional information is recorded in the extension header will be described.
The example shown in FIGS. 22 and 23 is an example in which a new extension header is inserted into the packet area following the HTTP header, and additional information is recorded in the extension header.

In FIG.
A list of additional information to be recorded in the extension header is shown, and FIG. 23 shows an example of three types of HTTP packets having an extension header.
In FIG.
(1) Initialization segment HTTP packet,
(2) Media segment-based metadata storage type NAL unit fragment compatible HTTP packet,
(3) Media segment-based media data storage type NAL unit fragment compatible HTTP packet,
Examples of these three types of HTTP packets are shown.
An extension header is set after the HTTP header of each packet.

Additional information according to the list shown in FIG. 22 is recorded in this extension header. An example of additional information to be recorded will be described with reference to FIG. As shown in FIG. 22, the additional information includes the following information.
(A) Segment identification information (Initialization Segment Identifier),
(B) In-GOP position identification information (Sub GOP Indicator),
(C) Movie Fragment Sequence Number (Movie Fragment Sequence Number),
(D) NAL unit fragment header (NALU Fragment Sub Segment Header),
(E) NAL unit internal location identification information (NALU Fragment Identifier)
(F) Access point information (SAPIndicator)

For example, (a) segment identification information is 8-bit data, 1 is set for an initialization segment, and 0 is set for other segments.
In the case of a media segment, if the content location information (ContentLocation) recorded in the HTTP header is set as the location information of GOP unit data, each GOP data can be distinguished by referring to this content location information.

(B) The position identification information in the GOP is set as the following settings, for example, as 8-bit data.
GOP head position data (StartofGOP) = 1
GOP intermediate position data (MiddleofGOP) = 2
Data at end position of GOP (EndofGOP) = 3

  (C) The movie fragment sequence number is 32-bit data, and the sequence number recorded in the metadata (moof) of the sub GOP media segment before division is recorded. This is arrangement information of NAL units stored in the sub GOP media segment.

  (D) The NAL unit fragment header is recorded as 8-bit data, with metadata storage type NAL unit fragment corresponding packet = 1 and other packets = 0.

(E) The NAL unit position identification information records, as 8-bit data, information indicating which position of the NAL unit fragment before the division is the NAL unit fragment (NALf) stored in the packet.
Start position of the NAL unit before split (StarNALunitFragmentSubSegment) = 1,
Middle position of NAL unit before division (MiddleNALunitFragmentSubSegment) = 2,
End position of NAL unit before division (EndNALUnitFragmentSubSegment) = 3,
Each bit value is recorded with this setting.

  (F) The access point information (SAPIndicator) is the same data as [X-SAP] described as the recording information of the HTTP header, and is 8-bit data. If the packet is a packet generated from the NAL unit storage segment including the access point, [1] is recorded, and if it is another packet, [0] is recorded.

In FIG.
(1) Initialization segment HTTP packet,
(2) Media segment-based metadata storage type NAL unit fragment compatible HTTP packet,
(3) Media segment-based media data storage type NAL unit fragment compatible HTTP packet,
The example which set each additional information of (a)-(f) shown in FIG. 22 to the extension header of these three types of HTTP packets is shown.

  The transmission device sets such an extension header and transmits it to the reception device. On the other hand, the receiving device refers to the segment identification information in the extension header and determines whether the segment stored in the packet is an HTTP packet storing an initialization segment or an HTTP packet corresponding to a metadata storage type NAL unit fragment It is determined whether the packet is a media data storage type NAL unit fragment compatible HTTP packet.

  Further, when the receiving apparatus identifies that the packet is a media data storage type NAL unit fragment-compatible HTTP packet, the receiving apparatus refers to each NAL unit by referring to the GOP position identification information in the extension header or the NAL unit position identification information. It becomes possible to rearrange fragments (NALf) and further reconstruct GOP.

  Furthermore, by referring to the access point information (SAPIndicator), it can be determined whether or not the packet is a packet generated from the NAL unit storage segment including the access point.

[11. Example in which additional information is recorded in LCT header]
As described above, transport protocols applicable to multicast and broadcast (MC / BC) type streaming include, for example, RTP (Real-time Transport Protocol) and FLUTE (File Delivery over Uni-directional Transport).

In the packet set according to the FLUTE protocol, an LCT header which is header information according to the FLUTE protocol is set.
That is, the LCT header in the IP packet described above with reference to FIG.
Hereinafter, an embodiment will be described in which additional information for smoothly executing processing on the receiving apparatus side, such as information applied to NAL unit fragment (NALf) rearrangement and GOP reconfiguration processing, is recorded in the LCT header. .

A configuration example of the LCT header set in accordance with the FLUTE protocol is shown in FIG.
Examples of main data fields set in the LCT header include the following fields.
CCI (Congestion Control Information): This field records the length of each field, congestion control information, and the like.
TSI (Transport Session Identifier): This field records session information of packet transfer.
TOI (Transport Object Identifier): A field for recording a packet transfer sequence and the like.
Header Extension (Header Extensions): A field in which various extension data can be recorded.

As described above with reference to FIG. 7, the data transmission order can be confirmed with reference to the above TOI as a substitute for the position information in GOP.
As described above, for example, only the X-StartofGOP indicating the head position is recorded in the GOP position information that is additional information recorded in the HTTP header or the like, and the arrangement order of the other GOP data is the TOI in the LCT header. The packet sequence may be acquired by referring to the NAL units in the GOP in the correct order according to the packet sequence.

  Similarly, the sequence information of NAL unit fragments (NALf) can be rearranged by referring to the above TOI and confirming the data transmission order. For example, in-NAL unit position identification information, which is additional information to be recorded in an HTTP header or the like, is recorded in the HTTP header only for the packet storing the NAL unit fragment at the head position. Subsequent NAL unit fragment sequences can be arranged by analyzing the packet sequence from the TOI of the LCT header.

  As shown in FIG. 24, a header extension (Header Extensions) capable of recording various data is set in the LCT header. Additional information for smoothly executing processing on the receiving device side, such as rearrangement of NAL unit fragments and information applied to GOP reconfiguration processing, can be recorded in this header extension section.

A data recording configuration for the header extension portion of the LCT header will be described with reference to FIG.
There are two types of formats for recording data in the header extension part of the LCT header.
FIG. 25 (1) shows a format in which the length of recording information can be freely set.
FIG. 25 (2) shows a format in which the length of recorded information is fixed.

In HET (Header Extension Type), extension header identification information (numerical value) indicating the type of extension header is recorded. The format of FIG. 25 (1) can use values up to 127, and the format of FIG. 25 (2) is specified to be set to 128 or more.
In HET (Header Extension Length), the length of the extension header is recorded.
The HEC (Neader Extension Content) is a field for recording extension header content, and can record arbitrary extension information.

  Using the format in which the length of the recording information shown in FIG. 25 (2) is fixed, the processing on the receiving apparatus side such as information to be rearranged in NAL unit fragments and information applied to the GOP reconstruction processing is smoothly executed. FIG. 26 shows a configuration example of recording data when additional information for recording is recorded.

As shown in FIG.
(A) The extension header identification information (HET) records, for example, [200] as the identification value of new information as 8-bit data.
In addition, the extended information recording unit (HEC)
(B) segment identification information,
(C) GOP position identification information,
(D) Movie fragment sequence number,
(E) NAL unit fragment header,
(F) NAL unit position identification information,
(G) Access point information Each of these pieces of identification information is recorded.

For example, (b) segment identification information is 8-bit data, 1 is set for the initialization segment, and 0 is set for the other segments.
In the case of a media segment, if the content location information (ContentLocation) recorded in the HTTP header is set as the location information of GOP unit data, each GOP data can be distinguished by referring to this content location information.

(C) The position identification information in GOP is set as the following setting, for example, as 8-bit data.
GOP head position data (StartofGOP) = 1
GOP intermediate position data (MiddleofGOP) = 2
Data at end position of GOP (EndofGOP) = 3

  (D) The movie fragment sequence number is 32-bit data, and the sequence number recorded in the metadata (moof) of the sub GOP media segment before division is recorded. This is arrangement information of NAL units stored in the sub GOP media segment.

  (E) The NAL unit fragment header is recorded as 8-bit data, metadata storing type NAL unit fragment corresponding packet = 1, and other packets = 0.

(F) In the NAL unit position identification information, information indicating which position of the NAL unit before the division the NAL unit fragment (NALf) stored in the packet is recorded as 8-bit data.
Start position of the NAL unit before split (StarNALunitFragmentSubSegment) = 1,
Middle position of NAL unit before division (MiddleNALunitFragmentSubSegment) = 2,
End position of NAL unit before division (EndNALUnitFragmentSubSegment) = 3,
Each bit value is recorded with this setting.

  (G) The access point information (SAPIndicator) is the same data as [X-SAP] described as the recording information of the HTTP header, and is 8-bit data. If the packet is a packet generated from the NAL unit storage segment including the access point, [1] is recorded, and if it is another packet, [0] is recorded.

  The transmission apparatus sets the LCT header including such extension header information and transmits it to the reception apparatus. On the other hand, the receiving device refers to the segment identification information of the extension header information in the LCT header, determines whether it is an initialization segment storage packet, a metadata storage type NAL unit fragment correspondence packet, or a media data storage type NAL. It is determined whether the packet is a unit fragment compatible packet. Further, when the receiving apparatus identifies that the packet is a media data storage type NAL unit fragment compatible packet, the receiving apparatus refers to the NAL unit position identification information or the GOP position identification information recorded in the extended header information of the LCT header. As a result, the NAL unit fragment (NALf), which is the media data (mdat) stored in each packet, is rearranged so as to be correctly arranged, and the GOP can be reconfigured.

  Furthermore, by referring to the access point information (SAPIndicator), it can be determined whether or not the packet is a packet generated from the NAL unit storage segment including the access point.

[12. Application example for HEVC encoded data]
In the above-described embodiment, the processing example for the MP4 encoded data according to the MPEG encoding method has been described. However, the process of the present disclosure can be applied to other encoded data. For example, the present invention can be applied to encoded data according to HEVC (High Efficiency Video Coding).

The configuration of HEVC encoded data will be described with reference to FIG.
As shown in FIG. 27, one HEVC stream generated by the HEVC encoding process has a plurality of sequences (CVS: Coded Video Sequence) and an End of sequence NAL unit (EoB).

The CVS, which is a component of the HEVC stream, includes a plurality of GOPs and an end of sequence NAL unit (EoS).
One GOP has an IRAP (Intra Random Access Point) access unit serving as a random access point as head data, and has a plurality of LP (Leading Picture) access units and a TP (Trailing Picture) access unit.

The LP access unit and the TP access unit are access units that can be decoded with reference to some reference picture.
However, when decoding is started from the IRAP access unit, normal decoding and reproduction can also be performed for TPAU.

  The access unit (AU) is composed of a plurality of NAL units (Network Abstraction Layer Units), and the access unit (AU) always includes one or more slice segment NAL units.

As shown in FIG. 27, a GOP as an encoding processing unit is also set in encoded data generated by HEVC encoding processing, and this GOP is further divided into a plurality of NAL units.
One or a plurality of NAL units, which are GOP configuration data set in HEVC encoded data, are stored in the sub GOP media segment described above with reference to FIG.

  Furthermore, as described above with reference to FIG. 16 and others, it is possible to generate a NAL unit fragment (NALf) obtained by further dividing a NAL unit and store it in an HTTP packet.

As described above, a NAL unit fragment (NALf) obtained by further dividing a NAL unit that is divided data of GOP data can also be generated for HEVC encoded data. Accordingly, the processing described with reference to FIG. 16 and the like is possible, and the processing of the present disclosure can be applied.
It should be noted that the present disclosure is not limited to MP4 and HEVC, as long as the encoding configuration has an encoding processing unit equivalent to GOP and a unit (NAL) in which GOP data is divided is set. is there.

[13. About hardware configuration example of each device]
Finally, with reference to FIG. 28, a hardware configuration example of each device that executes the above-described processing will be described.
FIG. 28 illustrates a hardware configuration example of a communication device applicable as the transmission device 20 and the reception device 30.

  A CPU (Central Processing Unit) 201 functions as a data processing unit that executes various processes according to a program stored in a ROM (Read Only Memory) 202 or a storage unit 208. For example, processing according to the sequence described in the above-described embodiment is executed. A RAM (Random Access Memory) 203 stores programs executed by the CPU 201, data, and the like. These CPU 201, ROM 202, and RAM 203 are connected to each other by a bus 204.

  The CPU 201 is connected to an input / output interface 205 via a bus 204. The input / output interface 205 is connected to an input unit 206 including various switches, a keyboard, a mouse, and a microphone, and an output unit 207 including a display and a speaker. Yes. The CPU 201 executes various processes in response to a command input from the input unit 206, and outputs a processing result to the output unit 207, for example.

  A storage unit 208 connected to the input / output interface 205 includes, for example, a hard disk and stores programs executed by the CPU 201 and various data. The communication unit 209 functions as a data communication transmission / reception unit and a broadcast wave transmission / reception unit via a network such as the Internet or a local area network, and communicates with an external device.

  A drive 210 connected to the input / output interface 205 drives a removable medium 211 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory such as a memory card, and executes data recording or reading.

  Data encoding or decoding can be executed as processing of the CPU 201 as a data processing unit, but a configuration including a codec as dedicated hardware for executing encoding processing or decoding processing may be adopted.

[14. Summary of composition of the present disclosure]
As described above, the embodiments of the present disclosure have been described in detail with reference to specific embodiments. However, it is obvious that those skilled in the art can make modifications and substitutions of the embodiments without departing from the gist of the present disclosure. In other words, the present invention has been disclosed in the form of exemplification, and should not be interpreted in a limited manner. In order to determine the gist of the present disclosure, the claims should be taken into consideration.

The technology disclosed in this specification can take the following configurations.
(1) a data processing unit for generating a packet storing encoded data;
A communication unit that transmits the packet generated by the data processing unit;
The data processing unit
Generate a packet that stores a NAL unit that constitutes a GOP (Group of Pictures), which is a processing unit of encoded data, or a NAL unit fragment obtained by further dividing the NAL unit,
As the additional information for the packet, the communication is performed by generating a packet in which access point information indicating whether the NAL unit stored in the packet or the NAL unit that is the division source of the NAL unit fragment includes data to be a random access point is set. A communication device for transmitting via a unit.

  (2) The access point information is information indicating whether the packet storage data is data generated based on a NAL unit including encoded data specified by SAP (Streaming Access Point). The communication device according to 1).

  (3) The data processing unit generates, as packet storage data, media data including one or more NAL units that are GOP configuration data and a sub-GOP media segment storing metadata corresponding to the media data. ) Or (2).

  (4) The data processing unit generates a NAL unit fragment obtained by further dividing the NAL unit constituting the GOP, and stores a plurality of packets storing the NAL unit fragment and metadata corresponding to the NAL unit. The communication device according to (1) or (2).

  (5) The data processing unit stores the divided data generated by dividing the encoded data to be transmitted into data equal to or less than the maximum data transfer unit (MTU) specified on the communication path. The communication device according to any one of (1) to (4).

  (6) The communication device according to any one of (1) to (5), wherein the data processing unit generates an HTTP packet and records the additional information in an HTTP header in the generated packet.

  (7) The communication device according to any one of (1) to (5), wherein the data processing unit records the additional information in an extension header in a generated packet.

  (8) The communication device according to any one of (1) to (5), wherein the data processing unit records the additional information in an LCT header in a generated packet.

  (9) The communication device according to any one of (1) to (8), wherein the communication unit performs broadcast distribution or multicast distribution of the packet generated by the data processing unit via a broadcast wave.

(10) a communication unit that receives the encoded data storage packet transmitted by the transmission device;
A data processing unit that inputs a packet received by the communication unit and executes processing;
Each of the encoded data storage packets received by the communication unit is:
A packet that stores a NAL unit that constitutes a GOP (Group of Pictures) that is a processing unit of encoded data, or a NAL unit fragment that is a further division of the NAL unit. As additional information, the NAL unit or NAL stored in the packet A communication apparatus which is a packet in which access point information indicating whether or not a NAL unit which is a unit fragment division source includes data to be a random access point is set.

(11) The data processing unit
With reference to the access point information, which is additional information recorded in the received packet, the data stored in the received packet is generated based on a NAL unit including encoded data specified by SAP (Streaming Access Point). The communication device according to (10), wherein a process for determining whether or not the received data is determined.

  (12) The data processing unit refers to the access point information that is additional information recorded in the received packet, selects and acquires data including the access point, and executes decoding and reproduction processing (10) Or the communication apparatus as described in (11).

  (13) The communication unit receives an HTTP packet in which the additional information is recorded in an HTTP header, and the data processing unit acquires the additional information from the HTTP header of the received packet. A communication device according to claim 1.

(14) The communication unit receives a packet in which the additional information is recorded in an extension header,
The communication device according to any one of (10) to (12), wherein the data processing unit acquires the additional information from an extension header of a received packet.

  (15) The communication unit receives a packet in which the additional information is recorded in an LCT header, and the data processing unit acquires the additional information from the LCT header of the received packet. The communication apparatus as described in.

  (16) The communication device according to any one of (10) to (15), wherein the communication unit receives the packet via a broadcast wave.

(17) A communication data generation method executed in a data transmission device,
The data processor
Generate a packet that stores a NAL unit that constitutes a GOP (Group of Pictures), which is a processing unit of encoded data, or a NAL unit fragment obtained by further dividing the NAL unit,
As a supplementary information for the packet, a communication unit is generated by generating a packet in which access point information indicating whether or not the NAL unit stored in the packet or the NAL unit that is the division source of the NAL unit fragment includes data serving as a random access point is set. A method for generating communication data to be transmitted via the Internet.

(18) A communication data processing method executed in the data receiving device,
A communication unit receiving an encoded data storage packet transmitted by a transmission device;
A data processing unit executes a data processing step of inputting a packet received by the communication unit and executing a process,
Each of the encoded data storage packets received by the communication unit is:
A packet that stores a NAL unit that constitutes a GOP (Group of Pictures) that is a processing unit of encoded data, or a NAL unit fragment that is a further division of the NAL unit. As additional information, the NAL unit or NAL stored in the packet A packet in which access point information indicating whether or not the NAL unit that is the source of the unit fragment includes data to be a random access point is set,
The data processing step is a communication data processing method which is a step of executing processing by inputting a packet in which access point information received by the communication unit is set.

  The series of processing described in the specification can be executed by hardware, software, or a combined configuration of both. When executing processing by software, the program recording the processing sequence is installed in a memory in a computer incorporated in dedicated hardware and executed, or the program is executed on a general-purpose computer capable of executing various processing. It can be installed and run. For example, the program can be recorded in advance on a recording medium. In addition to being installed on a computer from a recording medium, the program can be received via a network such as a LAN (Local Area Network) or the Internet and can be installed on a recording medium such as a built-in hard disk.

  Note that the various processes described in the specification are not only executed in time series according to the description, but may be executed in parallel or individually according to the processing capability of the apparatus that executes the processes or as necessary. Further, in this specification, the system is a logical set configuration of a plurality of devices, and the devices of each configuration are not limited to being in the same casing.

As described above, according to the configuration of an embodiment of the present disclosure, it is possible to determine in units of packets whether the stored encoded data of a communication packet is data based on encoded data that can be accessed randomly. This configuration is realized.
Specifically, the transmission device generates a packet storing a NAL unit that is a component of encoded data or a NAL unit fragment obtained by further dividing the NAL unit, and is the original data of the packet storage data as additional information Access point information indicating whether or not the NAL unit includes data to be a random access point is set and transmitted. The receiving apparatus refers to the packet additional information to determine whether or not the packet storage data is data generated based on randomly accessible encoded data.
With this configuration, it is possible to realize a configuration in which it is possible to determine whether or not the stored encoded data of the communication packet is data based on randomly accessible encoded data for each packet.

DESCRIPTION OF SYMBOLS 10 Communication system 20 Transmission apparatus 21 Data processing part 22 Communication part 30 Reception apparatus 31 Data processing part 32 Communication part 50 Initialization segment 60 Media segment 201 CPU
202 ROM
203 RAM
204 Bus 205 Input / Output Interface 206 Input Unit 207 Output Unit 208 Storage Unit 209 Communication Unit 210 Drive 211 Removable Media

Claims (18)

A data processing unit for generating a packet storing encoded data;
A communication unit that transmits the packet generated by the data processing unit;
The data processing unit
Generate a packet that stores a NAL unit that constitutes a GOP (Group of Pictures), which is a processing unit of encoded data, or a NAL unit fragment obtained by further dividing the NAL unit,
As the additional information for the packet, the communication is performed by generating a packet in which access point information indicating whether the NAL unit stored in the packet or the NAL unit that is the division source of the NAL unit fragment includes data to be a random access point is set. A communication device for transmitting via a unit.   The access point information is information indicating whether or not the packet storage data is data generated based on a NAL unit including encoded data specified by SAP (Streaming Access Point). Communication equipment. The data processing unit
The communication apparatus according to claim 1, wherein a sub GOP media segment storing media data including one or more NAL units as GOP configuration data and metadata corresponding to the media data is generated as packet storage data. The data processing unit
The NAL unit fragment which further divided | segmented the NAL unit which comprises GOP is produced | generated, The packet which stored the several packet which stored the NAL unit fragment, and the metadata corresponding to the said NAL unit is produced | generated. Communication device. The data processing unit
2. The packet according to claim 1, wherein the encoded data to be transmitted is divided into data equal to or less than a maximum data transfer unit (MTU) defined on the communication path, and a packet storing the divided data is generated. Communication device. The data processing unit
The communication apparatus according to claim 1, wherein an HTTP packet is generated, and the additional information is recorded in an HTTP header in the generated packet. The data processing unit
The communication apparatus according to claim 1, wherein the additional information is recorded in an extension header in the generated packet. The data processing unit
The communication apparatus according to claim 1, wherein the additional information is recorded in an LCT header in a generated packet. The communication unit is
The communication apparatus according to claim 1, wherein the packet generated by the data processing unit is broadcast or multicast via a broadcast wave. A communication unit that receives the encoded data storage packet transmitted by the transmission device; and
A data processing unit that inputs a packet received by the communication unit and executes processing;
Each of the encoded data storage packets received by the communication unit is:
A packet that stores a NAL unit that constitutes a GOP (Group of Pictures) that is a processing unit of encoded data, or a NAL unit fragment that is a further division of the NAL unit. As additional information, the NAL unit or NAL stored in the packet A communication apparatus which is a packet in which access point information indicating whether or not a NAL unit which is a unit fragment division source includes data to be a random access point is set. The data processing unit
With reference to the access point information, which is additional information recorded in the received packet, the data stored in the received packet is generated based on a NAL unit including encoded data specified by SAP (Streaming Access Point). The communication apparatus according to claim 10, wherein a process for determining whether or not the received data is data is executed. The data processing unit
The communication apparatus according to claim 10, wherein the access point information, which is additional information recorded in a received packet, is referred to, data including an access point is selectively acquired, and decoding and reproduction processing are executed. The communication unit is
Receiving an HTTP packet in which the additional information is recorded in an HTTP header;
The data processing unit
The communication apparatus according to claim 10, wherein the additional information is acquired from an HTTP header of a received packet. The communication unit is
Receiving a packet in which the additional information is recorded in an extension header;
The data processing unit
The communication apparatus according to claim 10, wherein the additional information is acquired from an extension header of a received packet. The communication unit is
Receiving a packet in which the additional information is recorded in the LCT header;
The data processing unit
The communication apparatus according to claim 10, wherein the additional information is acquired from an LCT header of a received packet. The communication unit is
The communication device according to claim 10, wherein the packet is received via a broadcast wave. A communication data generation method executed in a data transmission device,
The data processor
Generate a packet that stores a NAL unit that constitutes a GOP (Group of Pictures), which is a processing unit of encoded data, or a NAL unit fragment obtained by further dividing the NAL unit,
As a supplementary information for the packet, a communication unit is generated by generating a packet in which access point information indicating whether or not the NAL unit stored in the packet or the NAL unit that is the division source of the NAL unit fragment includes data serving as a random access point is set. A method for generating communication data to be transmitted via the Internet. A communication data processing method executed in a data receiving device,
A communication unit receiving an encoded data storage packet transmitted by a transmission device;
A data processing unit executes a data processing step of inputting a packet received by the communication unit and executing a process,
Each of the encoded data storage packets received by the communication unit is:
A packet that stores a NAL unit that constitutes a GOP (Group of Pictures) that is a processing unit of encoded data, or a NAL unit fragment that is a further division of the NAL unit. As additional information, the NAL unit or NAL stored in the packet A packet in which access point information indicating whether or not the NAL unit that is the source of the unit fragment includes data to be a random access point is set,
The data processing step is a communication data processing method which is a step of executing processing by inputting a packet in which access point information received by the communication unit is set.

Images