JP2013017031A - Image distribution device, reproduction device, and image communication system - Google Patents

Abstract

The present invention provides a video communication system that reduces packet loss when receiving and reproducing content distributed by an information communication server.
A packetization processing unit for generating a video data packet, an error correction processing unit for generating an FEC packet for each set of a predetermined number of video data packets, and a header for transmitting the video data packet and the FEC packet. The error correction processing unit 115 can generate FEC packets corresponding to a plurality of different schemes, and the header processing unit 112 includes all of a plurality of different FEC restoration schemes for each group of video data packets. Each time an FEC packet is generated, transmission of the video data packet and the generated FEC packet is started, and only the FEC packet related to the FEC restoration method specified by the playback device is transmitted among the generated FEC packets. [Selection figure] Fig. 2

Description

  The present invention relates to a video distribution device, a playback device, and a video communication system, and more particularly to a video distribution corresponding to a difference in network environment on the receiving side.

  In a video communication system composed of an information communication server that generates video in real time and a terminal that receives and displays the video and feeds back input from the user to the information communication server, a protocol used for communication is IP. Using a communication protocol such as HTTP (HyperText Transfer Protocol) based on TCP (Transmission Control Protocol) that establishes a one-to-one connection configured on (Internet Protocol), or RTP (Real-time Transport Protocol) Can be considered.

  Here, RTP communication does not retransmit even if the transmission data does not reach the other party, so if a packet loss occurs that causes a loss of part of the transfer data on the network, the video and audio will be disturbed. It is necessary to deal with. For video communications that require real-time response, lost packets must be restored in real time, so the receiver can correct errors in real time by sending data packets together with redundant packets for data restoration. In general, a packet is restored using an FEC (Forward Error Correction) method.

  The packet loss often occurs because the amount of data flowing on the network increases and the network load exceeds the allowable range of the bandwidth of the network communication path, that is, a so-called bandwidth shortage. In order to solve this problem, each end host connected to the network has a corresponding FEC proxy and quality measuring means, and measures the error correction repair state, and performs a session unit between end hosts that perform bidirectional real-time communication. A technology has been proposed that determines the quality state of the network, grasps the congestion state of each area of the network, eliminates the congestion state due to the over-capacity of the network that is the cause of packet loss, and reduces the packet loss in the entire network ( For example, see Patent Document 1).

  Also, by increasing the ratio of redundant packets, the ability to restore lost packets by FEC can be enhanced. However, the amount of transmission data increases, the network load increases, and the ratio of lost packets may increase. . In order to avoid such a situation, a method is disclosed in which a packet is received by a plurality of FEC restoration methods in which redundancy performance is changed, and an optimum FEC restoration method is applied (for example, refer to Patent Document 2).

JP 2006-045461 A JP 2010-161550 A

  As described above, in FEC, the degree of error correction repair is determined by the amount of redundant packets for restoration (hereinafter referred to as FEC packets) that are transmitted together with data. In other words, the lost packet restoration rate increases as more FEC packets are sent, but there is a problem in that the amount of transmission data increases and the network is overloaded and bandwidth shortage is likely to occur.

  In the case of a non-business network device, the local network can adjust the congestion state by adjusting the route by the server by informing the distribution server of the error correction rate from the receiver side as in Patent Document 1. Up to the entrance, i.e. to the router. In general, route adjustment is not possible within the local network, and the allowable bandwidth is often smaller than that of the external network. When RTP data finally arrives at the local operation terminal, packet loss occurs. Will be. Therefore, it is necessary for the operating terminal to develop a method for adjusting streaming quality maintenance using FEC according to its own data reception capability and the allowable bandwidth of the local network.

  The technique disclosed in Patent Document 2 is assumed to be in the form of television broadcasting over a network, and is premised on a plurality of clients receiving video data distributed from a server. Therefore, a plurality of FEC schemes, that is, FEC packets based on FEC schemes having different redundancy, have already arrived at the router that is the entrance of the local network, and each terminal in the local network selects an FEC packet acquired from the router. Change the FEC method.

  Such a form can be applied to a form of moving picture distribution in which a client and a server exchange data on a one-to-one basis. In order to deliver a video to a single client, FEC based on a plurality of FEC methods is used. Delivering packets is inefficient. Therefore, as a method for selecting and receiving an optimal FEC method in a client connected in a local area network, a method suitable for a moving image distribution form in which the client and the server exchange data on a one-to-one basis is required. .

  In view of the above-described problems, an object of the present invention is to provide a video communication system in which packet loss is reduced when content distributed by an information communication server is received and reproduced.

  In order to achieve the above object, one aspect of the present invention provides a video distribution device that distributes video information in response to a request from a playback device that plays back a video via a network, the video information relating to the request A video data packet generator for generating a video data packet for transmitting the video information via a network and storing the video data packet in a storage medium, and a predetermined number based on the generated video data packet An error correction packet generator for generating an error correction packet for correcting an error in packet transmission and storing it in a storage medium for each of the video data packets, and the generated video data packet and error correction packet A packet transmitter for transmitting to the playback device via a network, and the error correction packet The generator can generate an error correction packet corresponding to each of a plurality of different error correction methods, and the packet transmission unit can generate the plurality of different errors for each set of a predetermined number of the video data packets. Each time the error correction packet is generated for all the correction methods, transmission of the video data packet in which the error correction packet is generated and the generated error correction packet is started.

  According to another aspect of the present invention, there is provided a playback device that receives and plays back video information via a network, and includes a plurality of video data packets generated based on video information to be played back and errors in packet transmission. A packet receiver for receiving an error correction packet for restoring the lost video data packet; and the selected error correction based on the received error correction packet for the video data packet lost due to an error in packet transmission. A packet restoring unit for restoring by a method, an error rate calculating unit for calculating an error rate that is an error occurrence rate in packet transmission based on the plurality of received video data packets, and the calculated error rate is predetermined. If the specified threshold value is exceeded, the error correction method Characterized in that it comprises a mode change notifying unit for notifying the video distribution device further.

  According to still another aspect of the present invention, there is provided a video communication system including the above-described video distribution device and the above-described playback device.

  According to the present invention, when a plurality of users use devices on a home network, it is possible to provide a function that can be easily resumed.

It is a figure which shows the operation | use form of the video communication system which concerns on embodiment of this invention. It is a figure which shows the function structure of the information communication server which concerns on embodiment of this invention. It is a figure which shows the function structure of the operating terminal which concerns on embodiment of this invention. It is a figure which shows the FEC decompression | restoration system which concerns on embodiment of this invention. It is a figure which shows the FEC decompression | restoration system which concerns on embodiment of this invention. It is a figure which shows the FEC decompression | restoration system which concerns on embodiment of this invention. It is a sequence diagram which shows operation | movement of the video communication system which concerns on embodiment of this invention. It is a flowchart which shows operation | movement of the information communication server 2 which concerns on embodiment of this invention. It is a figure which shows the packet transmission aspect which concerns on embodiment and comparative example of this invention. It is a flowchart which shows operation | movement of the operating terminal which concerns on embodiment of this invention.

  A video communication system according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a video communication system according to the present embodiment. As shown in FIG. 1, in the video communication system according to the present embodiment, the information communication server 2 and the operation terminal 3 having a function as a playback device that receives and plays back video distributed from the server are routers 41, 43. , 44 and the switch 42 are used to transmit and receive video data packets and FEC packets.

  The network 1 is an Internet protocol communication network realized by using various network physical layers (for example, including an optical, wired, or wireless network). Data packets such as routers, switches, and hubs are transmitted from a transmission source to a transmission destination. It also includes communication control devices 41 to 44 for accurately delivering to the device.

  The information communication server 2 is a server system for providing a video transmission service by distributing video data using an RTP (Real-time Transport Protocol) protocol. The video content generated by the CPU blade 6 or the video content stored in the storage 7 is encoded as necessary by the transmission device 5 and subjected to data amount compression processing, and is converted into a predetermined data packet to be FEC. A packet is transmitted to the network 1.

  The operation terminal 3 is a video reception / playback device that receives video data and plays back video for the user. While referring to the user interface displayed on the screen of the operation terminal 3, the user inputs instructions for various functions through the touch panel interface. As shown in FIG. 1, the operation terminal 3 is connected to a local network 4, and the local network 4 is connected to an external network 1 via a router 44. Note that the operation terminal 3 does not necessarily include a display device or a speaker, and may be a receiving device (for example, a set top box (STB)) connected to an external display device or a speaker.

  Next, functional configurations of the information communication server 2 and the operation terminal 3 according to the present embodiment will be described. FIG. 2 is a diagram showing a functional configuration of the information communication server 2 according to the present embodiment. As shown in FIG. 2, the information communication server 2 includes a transmission device 5, a CPU (Central Processing Unit) blade 6, and a content storage 7. The transmission device 5 and the CPU blade 6 form a pair, and by adding a pair of the transmission device 5 and the CPU blade 6, it becomes possible to support a plurality of operation terminals 3. The content storage 7 can be referred to from a plurality of transmission devices 5.

  The transmission device 5 constituting the information communication server 2 includes a control unit 100, an internal bus 101, a memory 102, a video signal input unit 103, an audio signal input unit 104, a fiber channel connection unit 105, a fiber channel connection interface 106, and a video processing unit. 107, a video encoder 108, an audio processing unit 109, an audio encoder 110, a packetization processing unit 111, a header processing unit 112, a network connection unit 113, a network interface 114, and an error correction processing unit 115.

  Control unit 100, memory 102, network connection unit 113, fiber channel connection unit 105, video processing unit 107, video encoder 108, audio processing unit 109, audio encoder 110, packetization processing unit 111, header processing unit 112, network connection unit The error correction processing unit 115 and the error correction processing unit 115 are connected via the internal bus 101 and can communicate information with each other.

  The network connection unit 113 is connected to the network 1 via the network interface 114. The fiber channel connection unit 105 is connected to the fiber channel 115 via the fiber channel connection interface 106. A content storage 7 in which moving image content is stored is connected to the fiber channel, and the transmission device 5 can freely refer to the content via the fiber channel 123 and obtain data.

  The CPU blade 6 constituting the information communication server 2 includes a CPU 116, a memory 117, a communication unit 118, a video signal output unit 119, an audio signal output unit 120, a control communication unit 121, and an internal bus 122. The CPU 116, the memory 117, the communication unit 118, the video signal output unit 119, the audio signal output unit 120, and the control communication unit 121 are connected via an internal bus 122, and can communicate information with each other.

  The communication unit 118 is connected to the network 1 and can mutually communicate control information other than video data to the operation terminal 3 to obtain feedback as needed. As a communication protocol, a method capable of relatively reliably communicating can be used. For example, TCP (Transmission Control Protocol) can be considered. The control communication unit 121 is connected to the control unit 100 and can communicate control information with each other.

  The CPU blade 6 can generate a screen signal and an audio signal to be displayed on the operation terminal 3 on the memory 117 by the CPU 116 performing calculations according to the program loaded in the memory 117. The generated screen signal and audio signal can be output from the video signal output unit 119 and the audio signal output unit 120. The video signal output unit 119 and the audio signal output unit 120 are connected to the video signal input unit 103 and the audio signal input unit 104, respectively. A program for causing the CPU 116 to execute an operation for generating a screen signal and an audio signal is stored in a storage medium (not shown) such as an HDD (Hard Disk Drive) or an optical disk. It is loaded into the memory 117.

  The transmission device 5 has a function in which the control unit 100 controls each connection unit, encoder, and processing unit, encodes screen information input from the CPU blade 6, and sends the encoded image information to the network 1 as video data. In the encoding process, the memory 102 buffers the video signal and the audio signal input from the video signal input unit 103 and the audio signal input unit 104. For the video, the video processing unit 107 performs image quality adjustment on the video signal, and the video encoder 108 encodes the video signal and sends it to the packetization processing unit 111. For voice, the voice processing unit 109 performs voice adjustment on the voice signal, and the voice encoder 110 encodes and sends the packet to the packetization processing unit 111.

  The packetization processing unit 111 converts the encoded video and audio into multiplexed video data packets in a predetermined format in consideration of timing such as lip sync, and sends them to the header processing unit 112 and the error correction processing unit 115. send. That is, the packetization processing unit 111 functions as a video data packet generation unit. In order to generate an FEC packet, the error correction processing unit 115 buffers the necessary number of packets sent to the memory 102 to generate an FEC packet, and sends the generated FEC packet to the header processing unit 112. That is, the FEC packet is an error correction packet for correcting an error in packet transmission, and the error correction processing unit 115 functions as an error correction packet generation unit.

  The header processing unit 112 adds sequence numbers corresponding to the headers of the transmitted video data packet and FEC packet, and transmits them to the network 1 via the network connection unit 113. That is, the header processing unit 112 functions as a packet transmission unit. The error correction processing unit 115 can simultaneously generate a plurality of FEC packets for one video data packet stream. The CPU blade 6 manages and grasps what FEC method is provided.

  Next, a functional configuration of the operation terminal 3 according to the present embodiment will be described. FIG. 3 is a diagram illustrating a functional configuration of the operation terminal 3 according to the present embodiment. As shown in FIG. 3, the operation terminal 3 includes a CPU 10, a RAM (Random Access Memory) 11, a ROM (Read-Only Memory) 12, a memory card 13, a media control unit 14, a DVD drive 15, and an HDD (hard disk drive) 16. , Touch panel 17, descrambler 18, demultiplexer 19, video / audio decoder 20, video display control unit 21, display device 22, graphic / audio engine 23, audio output control unit 24, speaker 25, IP communication port 26 and IP data. A processing unit 27 is included.

  The CPU 10 performs information processing and device control in the operation terminal 3. The RAM 11 is a storage medium serving as a work area when the CPU 10 performs calculations for information processing, crisis control, and the like, and is loaded with a program for causing the CPU 10 to perform calculations. Such various programs are stored in a nonvolatile storage medium such as the ROM 12, the memory card 13, the DVD drive 15, and the HDD 16. The media control unit 14 controls input / output of video data to / from a recording medium.

  The touch panel 17 receives an operation of a user who is an operator of the operation terminal 3. The descrambler 18 restores the scrambled signal. The demultiplexer 19 separates the multiplexed signal. The video / audio decoder 20 decodes the encoded video data and audio data. The video display control unit 21 controls video display on a display device 22 such as a liquid crystal display. The graphic / audio engine 23 is browser software that controls processing for displaying a Web page screen, a still image, and the like. The audio output control unit 24 controls audio output from the speaker 25. The IP communication port 26 transmits and receives IP data packets via the network. The IP data processing unit 27 distributes RTP packets received by the IP communication port 26, HTML (HyperText Markup Language) video data received and transmitted by other HTTP protocols, and the like, and performs processing corresponding to each.

  As described above, the display device 22 and the speaker 25 are not necessarily required, and an output unit that outputs data from the video display control unit 21 and the audio output control unit 24 to the outside (for example, an S terminal, a D terminal, The component video terminal or the like may be an optical digital terminal, a coaxial digital terminal, or the like for audio and the HDMI terminal or the like for video / audio. Also, all of the nonvolatile storage media such as the DVD drive 15 and the HDD 16 are not necessarily required, and a storage area for storing necessary programs and data may be secured. A plurality of each storage device may be provided.

  The operation terminal 3 includes an IP communication port 26 for transmitting and receiving IP data packets, and an IP data processing unit 27. For example, TCP, UDP (User Datagram Protocol), DHCP (Dynamic Host Configuration Protocol), DNS (domain Name Server), HTTP. Various communication protocols such as RTP, RTSP (Real Time Streaming Protocol), MLD (Multicast Listener Discovery), IGMP (Internet Group Management Protocol) and applications such as browsers are mounted on the IP communication port 26. Connected to the external network 1 via You can receive the image data to be distributed in the distribution server 2.

  The RTP packet received by the IP communication port 26 is converted into TS (Transport Stream) data by performing data restoration processing by FEC by the IP data processing unit 27 and descrambling processing by the descrambler 18, and is converted to TS (Transport Stream) data via the demultiplexer 19. The video / audio decoder 20 decodes and decompresses the video data, and the video data is reproduced on the display device 22 and the speaker 25 via the video display control unit 21 and the audio output control unit 24.

  Of the components shown in FIG. 3, for example, the descrambler 18, the demultiplexer 19, the video / audio decoder 20, and the like may be realized by software. In that case, a software program for realizing each function is loaded onto the RAM 11, and the CPU 10 performs an operation according to each program, thereby realizing each function.

  In the present embodiment, the operation terminal 3 receives the video data transmitted from the information communication server 2 and displays it on the display device 22, and the user selects a menu or the like displayed on the touch panel 17. The operation terminal 3 can grasp in the CPU 10 what kind of FEC restoration method is provided based on information obtained by communication with the CPU blade 6. The FEC packet reception request is prepared in accordance with the instruction of the CPU blade 6, the video data and the FEC packet are received, the packet restoration process is performed, and then converted into a TS (Transport Stream) signal to be reproduced.

  The operation terminal 3 receives the video data data packet such as “A” shown in FIG. 1 and the FEC (Forward Error Correction) packets such as “A1” and “A2” shown in FIG. 1 at different ports. be able to. That is, the data packet and the FEC packet are received in order by the same physical interface, but are received in parallel as a concept of processing in information communication.

  An FEC restoration method for packet restoration is provided as shown in FIG. 4 based on the public standard Pro-MPEG, for example. Video data packets to be transmitted / received are conceptually arranged in a matrix of M rows and N columns vertically and horizontally like “1, 2, 3,... M, M + 1, M + 2,. An FEC packet is generated as a lateral FEC packet for restoring a missing packet in each M packet data for each consecutive M packet arranged, that is, for each row of packets arranged in a matrix. Is done.

  In addition, as shown in FIG. 4, for each column of packets arranged in a matrix, an FEC packet for restoring one packet loss in each N packet data is generated as a vertical FEC packet. The This vertical / horizontal FEC packet has data for restoration created by XOR (exclusive OR) of corresponding M or N data. The RTP header of the video data packet A contains a continuous sequence number. Therefore, the operation terminal 3 can detect occurrence of a packet lost by detecting a missing sequence number. These FEC packet generation processes are executed by the error correction processing unit 115 of the transmission device 5.

  The RTP header of the FEC packet contains data indicating the sequence number of the packet that is the target of restoration of the lost packet by the FEC packet, that is, the sequence number of the packet that is the target of the XOR operation. When a packet loss occurs, it can be determined which FEC packet is used for restoration.

  FIG. 5 shows an example of difference in FEC restoration method and lost packet restoration. For example, the total number of packets in the vertical direction is N + 1 (N video data packets + 1 FEC packet). Even if one of the video data packets is lost due to packet loss, a total of N received video data packets and FEC packets It is possible to restore one lost packet by performing an XOR operation. When two or more packets are lost, all lost packets cannot be recovered.

  As shown in FIG. 5, when five packet losts (marked with x in the figure) have occurred, if the FEC method provides only the FEC packet in the horizontal direction, it is possible to restore M * (N−1) +1. Is only one packet. Further, in the case of providing only the vertical FEC packet, the packet M can be restored, but other packets cannot be restored. However, if the method provides both vertical and horizontal FEC packets, it is possible to restore all packets in the order indicated by the numbers in FIG. The order shown in FIG. 5 is an example.

  In content distribution over a network, the quality of content that can be received is limited by the performance of the communication line and receiver used. The video content is transmitted as a data packet by the distribution server as a data packet, reaches the receiver via a device such as a router or a switch, and is taken into the receiver. The data transfer capacity per second at that time is called a bit rate and is a value specific to each content. The larger this value, the higher the quality such as the resolution and smoothness of the video, but the use band necessary for receiving data for display becomes larger.

  If the available bandwidth of the network or receiver used by the user is larger than the total bit rate of the received content, it is possible to receive and play content without problems, but the available bandwidth is better. If it is small, the data may not reach the receiver, or the reproduction process inside the receiver may not be in time, packet loss may occur, and the video may be distorted.

  As the number of FEC packets increases, the restoration rate increases. However, by receiving FEC packets in addition to video data, the required bandwidth as a whole increases, and the overall packet lost including FEC packets increases. is there. This problem will be described based on the FEC restoration method and the table showing the number of FEC packets in FIG.

  When six FEC restoration methods are prepared as shown in FIG. 6, for example, in the FEC restoration method A, one vertical FEC packet is received for every ten video data packets, and the vertical content packet 10 is received. A maximum of one can be restored. Since 10 FEC packets are received for 100 pieces of video data, the reception bandwidth is increased by 10%.

  On the other hand, in the FEC restoration methods B to F, the possibility that the content packet can be restored increases as described above, but the required reception band increases. For example, in the case of FEC restoration methods D and E, since 25 FEC packets are received for 100 pieces of video data, the reception band is increased by 25%. In this case, the FEC restoration methods D and E are used even if, for example, a video data stream of 16 Mbps is originally received by a local network having an allowable bandwidth of 17 Mbps and packets are dropped in a burst or the like in rare cases. Therefore, a bandwidth of 20 Mbps, an increase of 25%, is required. In this case, a bandwidth exceeding the allowable bandwidth of 17 Mbps is required, and there is a possibility that packet loss will always occur. Therefore, it is necessary to have a mechanism for selecting the best method for the local network being used.

  In FIG. 1, a data packet A, a horizontal FEC packet A1, and a vertical FEC packet A2 storing video data are distributed from the information communication server 2 via the network 1. In this example, since the data packet A, the FEC packet A1 in the vertical direction, and the FEC packet in the horizontal direction A2 are transmitted to different ports, the operation terminal 3 needs to receive the packets at the respective ports. If FEC is not used, only data packet A needs to be received. Normally, the operation terminal 3 always receives and reproduces the data packet A including the video data from the information communication server 2.

  Next, the operation of the video communication system according to this embodiment will be described. FIG. 7 is a sequence diagram showing the operation of the video communication system according to the present embodiment. For example, when the operation terminal 3 operates by receiving information from the information communication server 2 when the user operates the operation terminal 3, that is, when a response for exchanging information with the information communication server 2 is required. In the information communication server 2, it is necessary to shorten the delay until the transmission of information is started so that the user does not wait. On the other hand, the delay is not a problem in a state where it is not necessary to respond to an operation from the user, for example, a state where a moving image is displayed.

  Therefore, the information communication server 2 according to the present embodiment uses the FEC only when moving image data is transmitted. That is, in the information communication server 2, the error correction processing unit 115 generates an FEC packet only when moving image data is transmitted. In this way, information can be distributed without making the user aware of the delay. In the present embodiment, the FEC method is dynamically selected on the operation terminal 3 side. This makes it possible to ensure the quality of the moving image.

  As shown in FIG. 7, first, the operation terminal 3 connects to the information communication server 2 via the network according to the user's operation and acquires information from the information communication server 2, so that a comparison like a menu screen is performed. A static screen is displayed (S701). Also in the processing of S701, the transmission device 5 shown in FIG. 2 functions as described above, and transmits a low bit rate video data packet for displaying a menu screen. The case where a video data packet having a bit rate lower than that of a moving image is transmitted as the video data packet of the menu screen is a case where the information communication server 2 adopts an encoding method corresponding to the dynamic bit rate. .

  In addition, a screen that is a user interface such as a menu screen is required to have a low delay because a quick response to a user operation is required. Therefore, the error correction processing unit 115 does not generate the FEC packet when transmitting the video data packet of the menu screen. Therefore, when transmitting the video data packet of the menu screen, the header processing unit 112 sequentially processes the generated video data packet for transmission.

  When the user operates the operation terminal 3 and selects a function for reproducing a moving image with a high bit rate compared to a static screen such as the menu screen (S702), the operation terminal 3 responds to the operation by the user. The operation information is transmitted to the CPU blade 6 (S703). The CPU blade 6 analyzes the content of the operation, determines a moving image to be distributed to the operation terminal 3 (S704), and sends a playback preparation command to the control unit 100 of the transmission device 5 (S705). The transmission device 5 stops the input of the video data for the menu screen to the packetization processing unit 111, and sends the encoded data of the moving image reproduction waiting screen previously stored in the memory 102 to the packetization processing unit 111. .

  When the encoded data of the video playback standby screen is input to the packetization processing unit 111, the video data packet of the video playback standby screen is transmitted to the operation terminal 3 via the header processing unit 112, and the video playback is transmitted to the operation terminal 3. A waiting screen is displayed (S706).

  Next, the CPU blade 6 sends an FEC reception preparation notification to the operation terminal 3 (S707). As a result, the operation terminal 3 opens a socket for receiving FEC data packets and starts preparation for FEC reception. Then, moving image data with FEC is prepared between the CPU blade 6 and the transmission device 5 (S708). At this time, the source of the moving image data is not limited to the screen data generated by the CPU blade 6 but may be acquired from the content storage 7. Details of the process will be described later. When the moving image data with FEC is prepared, the moving image data with FEC is sent to the operation terminal 3 instead of the reproduction waiting screen. Thereby, the reproduction processing of the moving image content with FEC is executed in each of the information communication server 2 and the operation terminal 3 (S709). Note that S709 includes a process in which the operation terminal 3 determines an optimum FEC method and a process in which the information communication server 2 selects an FEC packet to be transmitted. This will be described in detail later.

  Thereafter, when the user performs an operation to stop moving image reproduction on the operation terminal 3 (S710), the operation information is notified to the CPU blade 6 in the same manner as described above (S711). The CPU blade 6 stops the video playback process (S712) and sends a playback stop command to the transmission device 5 (S713). A notification to end the FEC transmission is sent to the operation terminal 3 (S714). In response to this, the operation terminal 3 stops receiving the FEC packet.

  Next, in the transmission device 5, the header processing unit 112 stops the processing when the video data segment (Group of Picture) of the time closest to the timing at which the playback stop command is received reaches the packetization processing unit 111. The sequence number of the last transmitted packet is stored. Further, the control unit 100 switches the source data input to the video processing unit 107 and the audio processing unit 109 to the static screen data generated by the CPU blade 6 (S715). At the same time, data transmission to the error correction processing unit 115 is stopped, and the buffer on the memory 102 related to the moving image data is discarded (S716).

  Thereafter, the screen generated by the CPU blade 6 is transmitted to the operation terminal 3 via the encoder through the packetization processing by the packet processing unit 111 and the header processing by the header processing unit 112 (S717). At this time, the header processing unit 112 uses the sequence number of the recorded packet so that the sequence number does not become discontinuous. As a result, the video data packet sent from the information communication server 2 maintains the state of one stream in which the sequence numbers are continuous, so that there is no need for timing adjustment in the operation terminal 3.

  Next, details of preparation and display of moving image video data with FEC (S708, S709) will be described using the flowchart of FIG. As shown in FIG. 8, the control unit 100 first displays a screen generated by the CPU blade 6 or content source data stored in the content storage 7 based on user operation information transmitted from the operation terminal 3. The reproduction of the moving image content is started by inputting each to the processing unit 107 and the audio processing unit 109 (S801).

  At this time, the packetization processing unit 111 is input with the video source of the reproduction waiting screen for displaying the reproduction waiting screen on the operation terminal 3. The control unit 100 switches the input to the packetization processing unit 111 to the video source input to the encoder in step S801, and starts packetization processing (step S802). At this time, the reproduction waiting screen is already being transmitted to the network 1 via the header processing unit 112 in S706. Therefore, the packetization processing unit 111 once stores the generated video data packet in the buffer on the memory 102 (step S803).

  When the video data packet is stored in the buffer on the memory 102, the control unit 100 controls the error correction processing unit 115 to start the generation process of the FEC packet (step S804). At this time, FEC packets corresponding to a plurality of types of FEC schemes as shown in FIG. 6 are generated. Therefore, when the generation of the FEC packet is started, the control unit 100 waits until the video data packet necessary for the FEC packet generation processing is accumulated in the buffer, and as shown in FIG. As described above, every time an FEC packet corresponding to a plurality of types of FEC methods can be generated, the FEC packet is generated (S805 / NO).

  The standard of the number of video data packets stored in S804 and S805 is set to the number of video data packets required for FEC generation among the types of FEC to be generated. For example, in the case of the example of FIG. 6, the control unit 100 includes video data packets that can generate FEC packets according to the FEC restoration method F, that is, 100 video data packets arranged in a matrix of 25 × 4. The data is stored in the buffer, and the error correction processing unit 115 waits until all the FEC packets according to the plurality of FEC methods shown in FIG. 6 are generated.

  When the generation of the FEC packet for the plurality of FEC methods as shown in FIG. 6 corresponding to the largest number of video data packets necessary for FEC generation is completed (S805 / YES), the control unit 100 transmits the packet The conversion processing unit 111 detects a break in the video data of the screen waiting for reproduction being processed, determines whether the screen can be switched, and waits until the break is reached (NO in S806). In S806, for example, the control unit 100 sets a GOP (Group of Pictures) of video data as a switching point.

  When the point at which screen switching is possible is reached (S806 / YES), the control unit 100 controls the packetization processing unit 111 to transmit all the packets that have not been transmitted at that point and then wait for playback. The packetization processing of the video source is stopped (S807). Then, the data packet input to the header processing unit 112 and transmitted to the operation terminal 2 is converted into the video data packet of the moving image to be reproduced via the error correction processing unit 115 and the FEC packet generated based on the video data packet. Switching (S808).

  Thereafter, in accordance with the control of the control unit 100, the video data packet packetized as described above by the packetization processing unit 111 and the FEC packet generated by the error correction processing unit 115 are sent to the header processing unit 112 to perform header processing. The applied packet is sent to the operation terminal 3 (S809). Here, in S809, not all the generated FEC packets are input to the header processing unit 112, but only the FEC packets related to the FEC restoration method selected by the operation terminal 3 are input to the header processing unit 112. The header is processed and created on the network. Thereby, in the local area network on the user side including the operation terminal 3, it is not necessary to exchange extra FEC packets, and the network load of the local area network can be reduced. This will be described in detail later.

  Here, the transmission mode and operation effect of the video data packet and the FEC packet in S809 will be described with reference to FIG. As described above, at the timing when the header processing unit 112 starts transmitting the video data packet and the FEC data packet in S809, the number of video data packets most necessary for FEC generation for a plurality of FEC methods as shown in FIG. The generation of the FEC packet in accordance with the large number is completed. Therefore, the buffer area reserved in the memory 102 includes a considerable number of video data packets, horizontal FEC packets (hereinafter referred to as FEC packet 1), and vertical FEC packets (hereinafter referred to as FEC packet 2). Is stored.

  That is, as shown in FIG. 9A, packets transmitted from the respective ports of the video data packet, the FEC packet 1 and the FEC packet 2 are already prepared. Accordingly, the header processing unit 112 sequentially performs header processing on the video data packet, the FEC packet 1 and the FEC packet 2 and transmits them, so that the video data packet, the FEC packet 1 and the FEC packet 2 are transmitted to the operation terminal 3. Reach in parallel. Therefore, even if a packet loss occurs, since the FEC packet has already arrived, the operation terminal 3 can immediately restore the lost packet.

  In FIG. 9B, as in the operation of FIG. 8, the process of starting transmission of video data packets is not performed after the generation of a considerable number of FEC packets is completed, and the generated packets are sequentially transmitted. It is a figure which shows the example in the case of producing | generating the FEC packet which became possible by the video data packet being produced | generated sequentially, and also transmitting it sequentially. FIG. 9B shows an example in which a horizontal FEC packet is generated for every five packets.

  In the case of FIG. 9B, when the packet shown in FIG. 9 is lost, it is necessary to wait for the arrival of the first FEC packet 1 in order to restore the packet. Therefore, the reproduction of the moving image is stopped during the waiting time shown in FIG. The waiting time for waiting for the FEC packet in the horizontal direction as shown in FIG. 9B can be an allowable range, but when waiting for the FEC packet in the vertical direction, the waiting time becomes longer, and the moving image The quality of reproduction will be impaired.

  On the other hand, the information communication server 2 according to the present embodiment uses the FEC for all of a plurality of different FEC restoration methods for each predetermined group of video data packets such as the matrix arrangement shown in FIG. In order to start transmission of the video data packet and the FEC packet every time a packet is generated, the video data packet, the FEC packet 1 and the FEC packet 2 are started to be transmitted in parallel as shown in FIG. Even if packet loss occurs, the operation terminal 3 does not have to wait for the arrival of the FEC packet, and can maintain the quality of moving image reproduction.

  In the mode shown in FIG. 9A as well, depending on the arrival timing of the packet to the operation terminal 3 and the generation mode of the lost packet, it may be necessary to wait for the arrival of the FEC packet in order to restore the packet. . On the other hand, such a possibility can be eliminated by the header processing unit 112 transmitting the FEC packet 1 and the FEC packet 2 with priority over the video data packet. Furthermore, when transmission of a packet is started for each group of a predetermined number of predetermined video data packets as in the matrix arrangement of FIG. 4, the video data packets are transmitted after all the FEC packets are transmitted first. You may do it. Thereby, when the reproduction of the video data in which the video data packet is received is started in the operation terminal 3, all the FEC packets are already prepared, and there is no need to wait for the arrival of the FEC packet for restoration.

  Next, the operation of the operation terminal 3 will be described. FIG. 10 shows data correction in the IP data processing unit 27 that performs processing for receiving a stream of video data by applying an optimum FEC restoration method when receiving and reproducing video data from the information communication server 2 in the operation terminal 3. It is a flowchart of a process control program.

  First, the operation terminal 3 starts a video data reception process when the user performs a predetermined operation for receiving and reproducing video data from the information communication server 2 on the touch panel 17 (S1000). Then, the operation terminal 3 requests the CPU blade 6 of the information communication server 2 to start video data transmission. Thereby, in the information communication server 2, transmission processing of video data starts. Thereby, a reproduction waiting screen display process or the like as in S706 of FIG. 7 is executed, and the reproduction waiting screen is displayed on the operation terminal 3.

  Next, the IP data processing unit 27 communicates with the CPU blade 6 to obtain the provided FEC method and parameters, and the source IP address and distribution port number of the FEC packet (S1001). In this example, information indicating that the six types of restoration methods FEC restoration methods A to F shown in FIG. Then, the IP data processing unit 27 proceeds to a procedure for receiving video data.

  When starting reception for the first time, the IP data processing unit 27 first starts reception processing without restoring a lost packet by FEC. Then, while monitoring the occurrence of packet loss, the FEC restoration methods are tried in order from the smallest bandwidth increase to select the optimum method. In the present embodiment, the IP data processing unit 27 accumulates the record of the FEC restoration method that is optimally used for each service type and the bit rate of the video data stream, and when the same service is used next time, The video data is received using the method determined to be most suitable (S1002).

  Next, the IP data processing unit 27 prepares a socket for taking a new data packet from the IP communication port 26 into the apparatus, and waits for a video data packet to start to arrive. At this time, immediately after the operation terminal 3 establishes a network session with the information communication server 2 and starts to reproduce the video data, the lost packet restoration processing by FEC is not executed as described above. When the video data packet starts to reach the IP communication port 26, the video data packet is captured and converted into the TS format data from the IP packet format by the IP data processing unit 27 and sent to the decoder 14. Through this process, the user-specified video data is reproduced and displayed (S1003). That is, the IP data processing unit 27 functions as a packet receiving unit.

  When the reproduction of the video data is started, the IP data processing unit 27 calculates the packet lost rate by counting the number of lost video data packets when the IP communication port 26 captures the packets (S1004). That is, the IP data processing unit 27 functions as an error rate calculation unit. Then, it is determined whether or not the packet lost rate exceeds a predetermined allowable rate threshold (S1005). If it exceeds (S1005 / YES), data restoration by FEC is started or until then. Change the FEC restoration method used.

  Here, the threshold value in S1005 will be described. The above-described threshold for the packet lost rate is set in advance according to the reproduction quality policy of the moving image. As a most conservative policy, there is a mode in which the threshold value is set to 0 packet and FEC is applied when even one packet lost occurs. Such a policy is mainly applied when the line quality of the local area network is high.

  When a relatively large amount of packet loss occurs on a general local area network line, the packet loss rate threshold is set higher than 0 packets. As a setting method at this time, a packet lost rate corresponding to an acceptable quality is set by visually confirming in advance how much the packet lost rate affects the image disturbance.

  For example, a main frame that is hostilely important for decoding is inserted in the moving image data. If packet loss frequently occurs for moving image data corresponding to the main frame, the screen will be greatly collapsed. Therefore, it is preferable to set the threshold value so that the packet corresponding to the main frame is not lost.

  As an upper limit of the threshold value to be set, it is conceivable to set it assuming a packet lost state that can be completely restored by FEC. According to the table of FIG. 6, the upper limit of the packet lost rate that can be restored by the FEC restoration method A having the lowest restoration rate, that is, less than 10% needs to be the threshold value. This corresponds to a case where 10 packets are continuously lost in the horizontal direction out of 100 packets.

  When changing the FEC restoration method, the IP data processing unit 27 first temporarily records the FEC restoration method and the packet lost rate when restoration by FEC has been performed so far (S1006). Then, from the FEC restoration method acquired in step S1001, the method with the smallest increase in bandwidth compared to the method used so far, that is, the method with the next highest network load after the currently selected error correction method is selected. Apply (S1007).

  The process of selecting and applying the changed FEC restoration method in S1007 is a process in which the IP data processing unit 27 notifies the information communication server 2 of the changed FEC restoration method via the network. That is, the IP data processing unit 27 functions as a method change notification unit. Thereby, in the information communication server 2, the control unit 100 selects an FEC packet to be input to the header processing unit 112 based on the notified FEC restoration method. As a result, an FEC packet corresponding to the FEC restoration method selected by the operation terminal 3 is transmitted to the operation terminal 3.

  Here, the selection rule of the FEC restoration packet will be described. For example, when video data is received without restoration by FEC until then, the method with the smallest increase in bandwidth among the six methods in FIG. 6 is a method of receiving 10 FEC packets for 100 packets of video data. First, FEC restoration method A is applied. In general, since packet loss often occurs continuously, the IP data processing unit 27 according to the present embodiment has a larger M, that is, more FEC packets in the vertical direction if the bandwidth increase rate is the same. Priority should be adopted. For example, C is applied to the FEC restoration methods B and C, and E is applied to the FEC restoration methods D and E first.

  The procedure for newly receiving an FEC packet that has not been received so far is the same as the reception of a video data packet, in which a socket for taking the FEC packet into the apparatus from the IP communication port 26 is prepared, and FEC packet data starts to arrive. Then, the restoration of the lost video data packet is started using the FEC packet by the method described with reference to FIGS. 4 and 5 (S1008). That is, the IP data processing unit 27 functions as a packet restoration unit.

  When data restoration by FEC is performed, the IP data processing unit 27 calculates a packet lost rate after data restoration (S1009). That is, the statistics of the value obtained by subtracting the number of restored packets from the number of lost packets at the time of reception are taken and compared with the lost rate of the previous method recorded in S1006 (S1010), and the number of packets is more than before. If the lost rate increases (S1010 / YES), the process proceeds to step S1011 to return the FEC restoration method to the previous one. Then, the IP data processing unit 27 records the selected FEC restoration method in the FEC application method database together with the bit rate of the service and the stream of the video data as the optimum FEC restoration method for the received video data (S1013). ).

  If the comparison result in step S1010 has improved compared to before (S1010 / NO), the IP data processing unit 27 determines whether there is an FEC restoration method that has not been tried yet (S1012). This corresponds to determining the possibility that there is a method with a higher restoration rate than the current method. If there is no method with a higher restoration rate (S1012 / YES), the method is set as the optimum method, and the process of S1013 is performed.

  If there is an FEC restoration method that has not been tried yet (S1012 / NO), the IP data processing unit 27 returns to step S1005. If the packet lost rate of the currently applied method is within the threshold range (S1005 / NO), the process proceeds to step S1013, and the IP data processing unit 27 records the method as the optimum method. If it is outside the threshold (S1005 / YES), the IP data processing unit 27 repeats steps S1006 to S1010.

  When the optimum method is determined, the IP data processing unit 27 applies the method to perform video data reception processing (S1014 / NO) until the user performs an operation to stop the connection with the information communication server 2 (S1014). . When the connection is stopped (S1014 / YES), the IP data processing unit 27 stops the data restoration and the transfer to the decoder, transmits a connection stop notification to the CPU blade 6, and deletes the receiving socket. Perform (S1015).

  Even if all the provided FEC restoration methods are used, if the packet lost rate exceeds the threshold in step S1005, the IP data processing unit 27 uses the FEC restoration method with the smallest packet lost rate. Select and process. In this case, since the reception status is not good, it may be displayed on the display device 22 that there is a possibility that the image or sound may be disturbed. As a result, the user can make a determination to select the low-resolution video data or the low-bit-rate video data being received.

  In the present embodiment, the information communication server 2 that normally transmits data by RTP provides a plurality of FEC restoration methods. Therefore, the operation terminal 3 selects the provided FEC restoration methods and always selects the optimum video data. It is characterized by adjusting so that reception is possible. As described above, when video data is received and displayed via the network, it is possible to rationally adopt and receive the optimum FEC restoration method in the user network.

  Further, as described above, the information communication server 2 according to the present embodiment reduces the delay by not using the FEC in the case of a video generated from a static screen that is required to have a small delay until the video display. In order to maintain the quality of video images with a reduced bit rate and a high bit rate, insert a playback wait screen at the beginning of the video playback and wait for playback while generating FEC packets that support multiple formats. It is possible to use FECs composed of a plurality of methods, and it is possible to improve the response performance with respect to user operations and to select an optimal FEC method.

  Here, as described above, in the video communication system according to the present embodiment, the information communication server 2 generates FEC packets corresponding to a plurality of different FEC restoration methods and then selects the FEC restoration selected by the operation terminal 3. Only FEC packets corresponding to the method are selected and transmitted. As described above, the operation terminal 3 does not restore the lost packet by FEC at the beginning of the reproduction of the video data. Accordingly, the information communication server 2 does not transmit the FEC packet at the beginning of the distribution of the video data.

  This may be notified that the FEC packet is not required at first, that is, no FEC restoration method is selected every time the operation terminal 3 makes a distribution request for video data to the information communication server 2. In the information communication server 2, the control unit 100 determines in advance that the FEC packet is unnecessary at the beginning of the distribution of the video data every time, and inputs only the video data packet to the header processing unit 112. Also good.

  Further, when the information communication server 2 receives a notification for changing the FEC restoration method from the operation terminal 3, the timing for changing the FEC restoration method becomes a problem. For this, it is preferable to use a group of video data packets arranged in a matrix as described in FIG. As described above, in the present embodiment, the information communication server 2 generates FEC packets according to the number of video data packets that are most required for FEC generation. That is, in the case of the example in FIG. 6, the information communication server 2 repeats the process of generating the FEC packet by all the FEC restoration methods A to F for every 100 video data packets.

  The FEC packet is meaningful only when applied to a video data packet arranged in a matrix as shown in FIG. 4, and even if the FEC method is changed in the middle of the video data packet arranged in a matrix. meaningless. Therefore, the timing at which the information communication server 2 changes the FEC restoration method according to the notification from the operation terminal 3, that is, changes the selection of the FEC packet input to the header processing unit 112 is as shown in FIG. The timing is preferably set for each group of video data packets arranged in a matrix.

  Note that it is possible to generate FEC packets for all of the different FEC restoration methods, as shown in FIG. 6, the FEC restoration method applied in this embodiment is a packet data grouped in a matrix. This is because the same number of 100 is adopted. Thus, as described above, the timing for changing the selection of the FEC packet, that is, the timing for changing the FEC restoration method, can easily use the timing for each group of video data packets arranged in a matrix. Can do. Accordingly, it is preferable that the plurality of FEC restoration methods that can be selected in the information communication server 2 adopt the same number as the number of packet data collected in a matrix.

  On the other hand, after notifying the change of the FEC restoration method, the operation terminal 3 needs to determine the timing when the FEC packet transmitted from the information communication server 2 is changed to the notified restoration method. As described above, since the video data packet to be restored is specified in the header information of the FEC packet, the IP data processing unit 27 can determine this by referring to the information.

  In addition, as a method for determining the change of the FEC restoration method more easily, FEC restoration is performed at the timing when the information communication server 2 changes the FEC restoration method, that is, the timing when the FEC packet input to the header processing unit 112 is changed. A mode of inserting information indicating that the method is changed is conceivable. As a mode of inserting information indicating that the FEC restoration method is changed, a mode in which a predetermined number of empty packet data including neither video data nor FEC data is continuously transmitted, or after the FEC restoration method is changed, As header information of a packet transmitted first, there can be considered a mode in which information indicating that the FEC restoration method has been changed or the FEC restoration method after the change is inserted. As a result, the operation terminal 3 can easily determine that the FEC restoration method has been changed by checking continuously received empty packets and header information of the received packets.

  Although the present invention has been described in detail above, it is needless to say that the present invention is not limited to the embodiments of the video communication system described herein, and can be widely applied to other video communication systems.

1 network 2 information communication server 3 operation terminals 41, 43, 44 router,
42 Switch 5 Transmitter 6 CPU Blade 7 Content Storage 10 CPU
11 RAM
12 Memory 13 Memory Card 14 Media Control Unit 15 DVD Drive 16 HDD
17 Touch Panel 18 Descrambler 19 Demultiplexer 20 Video / Audio Decoder 21 Video Display Control Unit 22 Display Device 23 Graphic / Audio Engine 24 Audio Output Control Unit 25 Speaker 26 IP Communication Port 27 IP Data Processing Unit 100 Control Unit 101 Bus 102 Memory 103 Video signal input unit 104 Audio signal input unit 105 Fiber channel connection unit 106 Fiber channel I / F
107 video processing unit 108 video encoder 109 audio processing unit 110 audio encoder 111 packetization processing unit 112 header processing unit 113 network connection unit 114 network I / F
115 Error correction processing unit 116 CPU
117 Memory 118 Communication unit 119 Video signal output unit 120 Audio signal output unit 121 Control communication unit

Claims (15)

A video distribution device that distributes video information in response to a request from a playback device that plays back a video via a network,
A video data packet generator for generating a video data packet for transmitting the video information via a network and storing the video data packet in a storage medium based on the video information according to the request;
An error correction packet for generating an error correction packet for correcting an error in packet transmission and storing it in a storage medium for each set of a predetermined number of the video data packets based on the generated video data packet A generator,
A packet transmission unit that transmits the generated video data packet and error correction packet to the playback device via a network;
The error correction packet generation unit can generate an error correction packet corresponding to each of a plurality of different error correction methods,
The packet transmission unit generates the error correction packet every time the error correction packet is generated for all of the plurality of different error correction methods for each predetermined number of the video data packets. The transmission of the video data packet and the generated error correction packet is started, and only the error correction packet according to the error correction method designated by the playback device is transmitted among the generated error correction packets. Video distribution device.   When an error correction method is specified by the playback device, the packet transmission unit transmits among the generated error correction packets in response to switching of the predetermined number of the video data packets. The video distribution apparatus according to claim 1, wherein the error correction packet is changed.   The said packet transmission part transmits the packet which shows changing an error correction system, when changing the error correction packet transmitted according to the designation | designated by the said reproducing | regenerating apparatus, It is characterized by the above-mentioned. Video distribution device.   The packet transmission unit, when changing an error correction packet to be transmitted according to designation by the playback device, inserts information indicating that the error correction method is changed into the header information of the packet to be transmitted after the change. The video distribution apparatus according to claim 1 or 2.   The error correction packet generator generates a plurality of different error correction methods by calculating an exclusive OR for a combination of a plurality of different video data packets for each predetermined group of the predetermined number of video data packets. The video distribution apparatus according to claim 1, wherein an error correction packet corresponding to each is generated. The error correction packet generation unit does not generate the error correction packet when the video distributed to the playback device is a video of a screen serving as a user interface,
When the video data packet is generated based on the video of the screen serving as the user interface, the packet transmission unit transmits the video data packet regardless of the predetermined number of the predetermined groups of the video data packets. The video distribution device according to claim 1, wherein the video distribution device is started.   The video distribution device according to claim 1, wherein the packet transmission unit transmits the error correction packet with priority over the video data packet.   The packet transmission unit transmits the video data packet after transmitting all the error correction packets according to the specified error correction method for each predetermined number of the video data packets. The video distribution device according to claim 7. Video distribution apparatus capable of generating an error correction packet corresponding to each of a plurality of different error correction methods and transmitting a video data packet generated based on video information together with the error correction packet according to the selected one error correction method A playback device that receives and plays back video information via a network,
A packet receiver for receiving a plurality of video data packets generated based on video information to be reproduced and an error correction packet for restoring the video data packet lost due to an error in packet transmission;
Based on the received error correction packet, a packet restoration unit that restores the video data packet lost due to an error in packet transmission by the selected error correction method;
Based on the received video data packets, an error rate calculation unit that calculates an error rate that is an error occurrence rate in packet transmission;
A playback apparatus, comprising: a method change notification unit that notifies the video delivery device of a change in the error correction method when the calculated error rate exceeds a predetermined threshold value.   The threshold value is determined by setting the lowest value among restoration rates indicating a ratio of the number of lost packets that can be restored to a predetermined number of the video data packets as an upper limit in the plurality of different error correction methods. The playback apparatus according to claim 9.   When the calculated error rate exceeds a predetermined threshold value, the method change notification unit determines whether the network load is next to the currently selected error correction method among a plurality of different error correction methods. The playback apparatus according to claim 9 or 10, wherein a high error correction method is selected and notified to the video distribution apparatus. The video distribution device arranges a predetermined number of the video data packets in a matrix by arranging them in a plurality of rows in the order in which the data continues, and then performs an exclusive OR operation in each of the row direction and the column direction. The error correction packet is generated by calculating and a plurality of different error correction methods are realized by changing the arrangement of the matrix.
When there are a plurality of error correction methods having the next highest network load after the currently selected error correction method, the method change notification unit prioritizes the method having the largest number of columns in the matrix arrangement of the video data packets. The reproduction apparatus according to claim 11, wherein the video distribution apparatus is selected and notified to the video distribution apparatus.   The packet restoration unit switches the error correction method when a packet indicating that the error correction method is changed is received after the change of the error correction method is notified. 12. The playback apparatus according to any one of twelve.   When the packet restoration unit is notified of the change of the error correction method, and the header information of the received video data packet or the error correction packet includes information indicating that the error correction method is changed The playback apparatus according to claim 9, wherein the error correction method is switched.   A video communication system, comprising: the video distribution device according to any one of claims 1 to 8; and the playback device according to any one of claims 9 to 14.

Images