JP4439564B2 - IP broadcasting system and multiplexing apparatus and receiving apparatus and method used in the IP broadcasting system - Google Patents

Description

  The present invention relates to an IP broadcast system for broadcasting TS (Transport Stream) of a plurality of programs from a broadcast station to a broadcast receiving apparatus via a broadcast network on an IP (Internet Protocol) network, and multiplexing used in the IP broadcast system The present invention relates to an apparatus, a receiving apparatus, and a method.

  2. Description of the Related Art In recent years, IP broadcast systems that transmit broadcast contents to users of broadcast reception apparatuses via an IP network in addition to transmission of broadcast contents by broadcast waves from broadcast stations to broadcast reception apparatuses have become widespread. In addition, in the IP network, a large-capacity, quality-controlled wide-area broadcasting network called CDN (content distribution network) or NGN (next generation network) is being developed. In such a network, encoded data of high-definition video is transmitted, and applications include VOD (video on demand) and IP broadcast programs.

  Normally, it is transmitted as an independent UDP (User Datagram Protocol) stream for each program and is not multiplexed. In such a case, particularly when a plurality of simultaneous programs are received by an IP broadcast set-top box (playback device), it is necessary to support simultaneous reception processing of a plurality of UDP streams, and the set-top box becomes complicated accordingly. Cost.

  In addition, there is a multiple program multiplexing method based on MPEG2 (Moving Picture Experts Group 2) -TS (Transport Stream), but the synchronization process between the programs is complicated and a delay occurs.

  On the other hand, although there is a restriction on the packet length in the normal network, a CDN or NGN can transmit a packet having a certain length as it is. Long packets can improve throughput (transmission efficiency). However, the conventional packet length of about 1500 bytes cannot fully utilize the performance of the network.

Conventionally, there are communication terminal devices that reduce the network load by multiplexing a plurality of packets in one frame and transmitting them (for example, Patent Literature 1) and network relay devices (for example, Patent Literature 2). .
JP 2006-60631 A JP 2002-9832 A.

  However, the communication terminal device and the network relay device described above multiplex a plurality of packets into one frame when performing communication, and do not multiplex a TS of a plurality of programs as in the IP broadcast system. Absent.

  SUMMARY OF THE INVENTION An object of the present invention is to provide an IP broadcasting system capable of improving network throughput and suppressing processing complexity on the receiving side when broadcasting a plurality of program TSs over an IP network, and the IP broadcasting system. It is to provide a multiplexing apparatus, a receiving apparatus and a method used in the above.

  The IP broadcasting system according to the present invention converts a plurality of TSs (Transport Streams) having packet structures with different programs and reproduction clocks into RTP (Real-time Transport Protocol) packets for IP (Internet Protocol) network transmission. Receives a plurality of RTP packets from the broadcast side on the IP network and the broadcast network, extracts a reproduction clock from each RTP packet, and reproduces a program from the TS in each RTP packet based on the reproduction clock A transmission side, and a transmission side includes a multiplexing unit that multiplexes a plurality of RTP packets to generate a standardized multiplexed frame including at least a header area and an information area. A buffer that temporarily holds multiple multiplexed frames, and multiplexed from this buffer Program reproduction that captures and analyzes frames, extracts a reproduction clock shared by a plurality of programs, separates a plurality of TSs for each program from the multiplexed frame, and reproduces a program from each TS based on the reproduction clock Means.

  The multiplexing unit obtains a delay amount from a plurality of input buffers that temporarily hold a plurality of RTP packets for each program and the buffer occupancy of each of the plurality of input buffers, and multiplexes them into a multiplexed frame based on these delay amounts. A calculation unit that reads out and outputs a plurality of RTP packets from a corresponding plurality of input buffers, a header generation unit that generates data of the header area of the multiplexed frame based on the plurality of RTP packets, and a calculation unit An output buffer is provided for storing a plurality of output RTP packets and for storing and outputting data in the header area of the multiplexed frame.

  According to this configuration, the transmission side simply converts a TS of a plurality of programs into RTP packets, holds each program in the input buffer, reads the RTP packets of the plurality of programs from the corresponding input buffer, and stores them in the output buffer. In this way, it is combined with the information area of the multiplexed frame, and the data of the header area of the multiplexed frame is generated from these RTP packets and stored in the output buffer so that the multiplexed frame is generated and sent to the receiving side. Yes. That is, the delay between a plurality of programs can be suppressed to the minimum on the transmission side, so that the buffer and the clock recovery circuit can be shared between the programs on the reception side.

  Therefore, when broadcasting TSs of a plurality of programs, the throughput of the network is improved, and it is not necessary to prepare a playback device for each program on the receiving side, thereby simplifying the multiplexing device and the playback device on the receiving side. The reception processing part can be unified.

  The multiplexing unit further includes rate detection means for detecting a transmission rate of each of the plurality of RTP packets, and the arithmetic unit is configured to transmit the maximum number of programs out of the plurality of programs to be multiplexed into the multiplexed frame based on the detection result by the rate detection means. A rate program is a master program, and a plurality of corresponding RTP packets are multiplexed at a rate of the transmission rate of the master program times the number of programs. The program reproduction means on the receiving side extracts the reproduction clock of the master program and performs reproduction processing.

  According to this configuration, since a plurality of RTP packets are multiplexed in the multiplexed frame in accordance with the master program having the maximum transmission rate on the transmission side, it is only necessary to refer to the time stamp of the master program on the reception side.

  The multiplexing unit is characterized in that a combination of a plurality of programs to be multiplexed is changed according to a reception area.

  According to this configuration, since a combination of a plurality of programs is changed for each reception area and multiplexed transmission is performed, an optimal broadcast service can be realized for each reception area, and viewing is restricted for each reception area. be able to.

  As described above in detail, according to the present invention, when broadcasting a TS of a plurality of programs via an IP network, an IP broadcasting system capable of improving the throughput of the network and suppressing the processing complexity on the receiving side, and It is possible to provide a multiplexing apparatus, a receiving apparatus and a method used in this IP broadcast system.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a schematic configuration diagram showing an embodiment of an IP broadcast system according to the present invention. BS1 to BSn (n is a natural number) are transmitters TV1 to TVm installed for each broadcast station or each service provider. (M is a natural number) indicates a television receiver.

  First, the transmitters BS1 to BSn convert broadcast programs that are independent of each other from TS to RTP packets for network (CDN) transmission and send them out. These RTP packets are supplied to the multiplexer 10. Multiplexer 10 multiplexes a plurality of RTP packets to generate an Ethernet (registered trademark) frame defined for IP network transmission, and sends the Ethernet (registered trademark) frame to network NW.

  The Ethernet (registered trademark) frame is sent to a receiver (including a combination of STB (Set Top Box) + existing television receiver) 20. The receiver 20 is provided with television receivers TV1 to TVm via a wired or wireless line. The receiver 20 receives an Ethernet (registered trademark) frame from the network NW, separates TS packets of the selected program, for example, according to a selection operation by a remote controller (not shown), and each television receiver TV1. ~ Output to TVm. The television receivers TV1 to TVm can demodulate TS packets and reproduce and output them.

  FIG. 2 is a block diagram showing a configuration of the multiplexing apparatus 10.

  That is, the multiplexer 10 includes a program selector 101, a rate detector 102, an input buffer memory 103, a packet selector 104, an output buffer memory 105, a multiplexing scheduler 106, an Ethernet frame header generation unit 107, It has.

  The RTP packets output from the transmitters BS1 to BSn are arbitrarily selected by the program selector 101, then the rate is detected for each program by the rate detector 102, supplied to the input buffer memory 103, and temporarily held.

  The multiplexing scheduler 106 analyzes each RTP packet based on the rate information for each program by the rate detector 102 and the buffer occupancy for each program in the input buffer memory 103, and controls the switching of the packet selector 104. The corresponding RTP packet is transferred to the output buffer memory 105. The Ethernet frame header generation unit 107 calculates the contents of the header portion of the Ethernet (registered trademark) frame based on each RTP packet in the output buffer 105 and stores the header portion of the Ethernet (registered trademark) frame in the output buffer memory 105. Write the contents of.

  The output buffer memory 105 stores the Ethernet (registered trademark) frame header from the Ethernet (registered trademark) frame header generation unit 107 and a plurality of RTP packets transferred from the input buffer memory 103 through the packet selector 104 to store the Ethernet. A (registered trademark) frame is generated and output as appropriate. At this time, the program having the maximum transmission rate is set as a master program, and a plurality of corresponding RTP packets are multiplexed into an Ethernet (registered trademark) frame at a transmission rate of the master program × the number of programs.

  FIG. 3 is a block diagram showing the configuration of the receiver 20.

  In the receiver 20, the Ethernet® frame that arrives from the network NW is temporarily stored in the reception buffer 201. The Ethernet (registered trademark) frame held in the reception buffer 201 is taken into the packet analysis unit 202. The packet analysis unit 202 performs RTP packet analysis processing, calculates the header content of the Ethernet (registered trademark) frame, transfers the RTP packet to the TS separation unit 203, and also sets a time stamp as the header content to the clock extraction unit 204. Output information.

  The TS separation unit 203 separates a plurality of TS packets from the input RTP packets and outputs them to the television receivers TV1 to TVm.

  On the other hand, in the clock extraction unit 204, the offset addition unit 2041 holds the clock difference between the past master program and the current master program, and adds the clock difference to the new time stamp as an offset. Thereafter, the output of the offset adder 2041 passes through the LPF 2042, is supplied to the clock oscillator (OSC) 2043, and is supplied to each of the television receivers TV1 to TVm as a reproduction clock for the program.

Next, the operation in the above configuration will be described.
FIG. 4 is a configuration diagram of an IP broadcasting system that has been considered before. In FIG. 4, the same parts as those in FIG.

  Data from each of the transmitters BS1 to BSn reaches the receiver 30 as a plurality of independent streams on the network NW. The receiver 30 operates the network receiving units in parallel in order to simultaneously reproduce a plurality of programs. Therefore, if the hardware configuration is used, the circuit scale increases, and if the software configuration is used, the processing load increases, so that a faster CPU is required.

  Advantages of receiving multiple programs simultaneously include faster program switching and constant network bandwidth. Even if only the program to be viewed is received, when the program is switched, the stream of the previous program may not stop immediately. Therefore, the stream of a plurality of programs temporarily reaches the receiver 30, and the network bandwidth is reduced. Not stable.

  Therefore, in the present embodiment, the data from the transmitters BS1 to BSn are multiplexed into a single stream by the multiplexer 10 and sent to the receiver 20 as it is.

  In the multiplexing apparatus 10, each program input format is an RTP packet in which a plurality of MPEG2-TS packets of 188 bytes or 204 bytes are collected, and one packet does not exceed 1518 bytes which is the maximum frame length of the network NW. Are composed of a maximum of 7 TS packets.

  When the maximum frame length of the network NW is exceeded, it is divided into 1518 bytes or less and transmission efficiency deteriorates. In the CDN, when transmission of a frame length exceeding 1518 bytes called an Ethernet (registered trademark) frame is allowed to be performed as it is, transmission efficiency is better when the frame on the network NW is lengthened to the maximum length. Due to such characteristics, transmission up to the front of the multiplexing apparatus 10 is performed with a normal frame length, and it is possible to transmit only within the CDN with an Ethernet (registered trademark) frame.

  For example, if the maximum frame length allowed by the CDN is 9600 bytes, the original 6 frames can be stored in one Ethernet (registered trademark) frame. Since the transmission efficiency is reduced due to the collision on the network NW, if the number of transmission sources is small, the frequency of occurrence of the collision is lowered and the efficiency is improved, and the jitter of arrival time at the STB due to waiting is also reduced.

  The program selector 101 is changed based on the program configuration determined by the service provider. When changing the broadcast contents depending on the area, it is possible to change the program to be selected for each area, and to restrict viewing in units of multiple streams designated by the access list of the router in the CDN.

  The access list is a setting list of the router and determines which stream is distributed to which area. The packet selector 104 controls in units of packets so that the packet having the longest time in the input buffer memory 103 is preferentially output. In this way, the multiplexing process is performed with a minimum delay. The output buffer memory 105 is for suppressing burst (stuck at a short interval) output so that the reception buffer 201 on the receiver 20 side does not overflow.

  FIG. 5 is a frame configuration example before and after multiplexing.

  The Ethernet (registered trademark) frame is generally standardized by IEEE802.3, and has a configuration of a header portion, a data portion, and an FCS (Frame Check Sequence) from the top. The header part is composed of a preamble, a destination address, a source address, and a length. The data part is composed of RTP packets. A plurality of TS packets are stored in the payload portion of the RTP packet. The header part of the RTP packet includes fields called Payload Type and SSRC, and by setting these for each program, it is determined which program the RTP packet belongs to. When the input rates of the programs match, the programs can be arranged and multiplexed in order from the top of the RTP packet. A master program index is added to the head of the plurality of RTP packets.

  However, since the input rate error accumulates in the long run, countermeasures are necessary. Therefore, the multiplexing speed is dealt with by outputting dummy data to the missing program portion in accordance with the program having the highest input rate. The program that determines the output rate is indicated as the master program at the head of the data part, so that the receiving side can determine which time stamp should be referred to.

  FIG. 6 is a flowchart showing the processing operation of the multiplexing scheduler 106.

  The multiplexing scheduler 106 sets the index of the program of the maximum rate from the rate information for each program by the rate detector 102 as a master program index (step ST6a), and then selects an unchecked program (step ST6b). The buffer occupancy / rate information of the input buffer memory 103 holding the selected program is calculated to obtain the delay amount (step ST6c).

  Then, the multiplexing scheduler 106 determines whether or not the obtained delay amount of the selected program is larger than the target delay amount (step ST6d). Here, when the delay amount of the selected program is larger than the delay amount obtained from the target delay amount and the jitter amount of each program input (Yes), the multiplexing scheduler 106 controls the packet selector 104 to switch the corresponding RTP packet. Is transferred and written to the output buffer memory 105 (step ST6e).

  On the other hand, when the delay amount of the selected program is equal to or less than the target delay amount (No), the multiplexing scheduler 106 writes dummy data in the lacking portion of the output buffer memory 105 (step ST6f). Thereafter, the multiplexing scheduler 106 checks the selected program (step ST6g), and determines whether or not the checking is completed for all the programs held in the input buffer memory 103 (step ST6h). Here, when not completed (No), the multiplexing scheduler 106 repeatedly executes the processes of step ST6b to step ST6h until the check is completed for all the programs. If completed (Yes), the multiplexing scheduler 106 clears all checks (step ST6i), and proceeds to the process of step ST6a. Thereafter, the contents of the header portion of the Ethernet (registered trademark) frame are written in the output buffer memory 105 by the Ethernet frame header generation unit 107.

  The receiver 20 extracts the TS from the multiplexed stream in units of programs and reproduces it. If the multiplexing order of the program is fixed on the multiplexing apparatus 10 side, the processing is simple because it is only read from a fixed position. However, for reproduction, the clock must be extracted from the time stamp of the RTP header. Therefore, when the master program determined on the multiplexing apparatus 10 side is changed, the time stamp to be referred to is changed on the receiver 20 side.

  Note that the re-extraction of the clock leads to a temporary suspension of the reception operation. Therefore, the offset adder 2041 holds the clock difference between the past master program clock and the current master program, and adds it to the new time stamp. By adding it as an offset, the receiving operation is continued without re-extracting the clock. The LPF 2042 is for absorbing clock jitter and can absorb not only the fluctuation of the network but also the shock of changing the master program. The dummy data inserted on the multiplexing apparatus 10 side is discarded at the time of reception.

  As described above, in the above-described embodiment, the multiplexing apparatus 10 holds RTP packets including TSs of a plurality of programs in the input buffer memory 103 for each program, and the multiplexing scheduler 106 switches and controls the packet selector 104. An RTP packet of an arbitrary program is read from the corresponding input buffer memory 103 and stored in the output buffer memory 105 to be coupled to the data portion of the Ethernet (registered trademark) frame, and further to the Ethernet (registered trademark) frame header generator 107. Then, the data of the header part of the Ethernet (registered trademark) frame is generated from these RTP packets and stored in the output buffer memory 105 to generate the Ethernet (registered trademark) frame and send it to the receiving side. That is, the delay between a plurality of programs can be suppressed to the minimum on the transmission side, whereby the receiver 20 can share the reception buffer 201 and the clock extraction unit 204 among the programs.

  Therefore, when broadcasting TSs of a plurality of programs via the network NW, the throughput of the network NW is improved, and it is not necessary to prepare a separate playback device for each program on the receiving side. Simplification and unification of the reception processing part of the receiver 20 on the reception side are possible.

  Further, in the above embodiment, on the transmission side, a plurality of RTP packets are multiplexed in the Ethernet (registered trademark) frame in accordance with the master program that maximizes the transmission rate. You only need to refer to the program time stamp.

  Furthermore, in the above-described embodiment, the multiplexing device 10 can change the combination of a plurality of programs for each reception area and perform multiplex transmission, so that an optimum broadcast service can be realized for each reception area and each reception area can be realized. Viewing restrictions can be made for each.

  Although the present invention has been described based on the embodiments, the present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment.

1 is a schematic configuration diagram showing an embodiment of an IP broadcast system according to the present invention. FIG. 2 is a block diagram showing a specific configuration of the multiplexing apparatus shown in FIG. FIG. 2 is a block diagram showing a specific configuration of the receiver shown in FIG. The schematic block diagram which shows the IP broadcast system considered before. The figure which shows the example of a frame structure before and behind multiplexing in the same embodiment. 6 is a flowchart showing a processing procedure of a multiplexing scheduler in the embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Multiplexer, 20 ... Receiver, 101 ... Program selector, 102 ... Rate detector, 103 ... Input buffer memory, 104 ... Packet selector, 105 ... Output buffer memory, 106 ... Multiplex scheduler, 107 ... Ethernet (registration) Trademark) Frame header generation unit, 201 ... reception buffer, 202 ... packet analysis unit, 203 ... TS separation unit, 204 ... clock extraction unit, 2041 ... offset addition unit, 2042 ... LPF, 2043 ... clock oscillator, BS1 to BSn ... transmission Device, NW ... network, TV1-TVm ... television receiver.

Claims (10)

A plurality of TSs (Transport Streams) having packet structures with different programs and reproduction clocks are converted into RTP (Real-time Transport Protocol) packets for IP (Internet Protocol) network transmission, respectively, and transmitted to a broadcast network on the IP network. With the sender,
Receiving a plurality of RTP packets from the broadcast network, extracting a reproduction clock from each RTP packet, and reproducing a program from a TS in each RTP packet based on the reproduction clock,
The sender side
A multiplexing unit that multiplexes the plurality of RTP packets to generate a standardized multiplexed frame including at least a header area and an information area;
The receiving side
A buffer that temporarily holds multiplexed frames from the broadcast network;
The multiplexed frame is fetched from the buffer and analyzed, and a reproduction clock shared by the plurality of programs is extracted. A plurality of TSs are separated from the multiplexed frame for each program, and each TS is based on the reproduction clock. And a program reproducing means for reproducing the program from
The multiplexing unit includes:
A plurality of input buffers for temporarily holding the plurality of RTP packets for each program;
An arithmetic unit that obtains a delay amount from the buffer occupancy amount of each of the plurality of input buffers, reads out and outputs a plurality of RTP packets from the corresponding input buffers to be multiplexed into the multiplexed frame based on these delay amounts, and
A header generation unit that generates data of a header region of the multiplexed frame based on the plurality of RTP packets;
An IP broadcast system comprising: an output buffer that stores a plurality of RTP packets output by the arithmetic unit and stores and outputs data in a header area of the multiplexed frame . The multiplexing unit further includes rate detection means for detecting a transmission rate of each of the plurality of RTP packets,
Based on the detection result by the rate detection means, the arithmetic unit sets a program having a maximum transmission rate among a plurality of programs to be multiplexed in the multiplexed frame as a master program, and a transmission rate of the master program × a rate of the number of programs. IP broadcasting system of claim 1, wherein multiplexing the corresponding plurality of RTP packets. 3. The IP broadcast system according to claim 2 , wherein the program reproduction means extracts a reproduction clock of the master program and performs reproduction processing. The IP broadcast system according to claim 1, wherein the multiplexing unit changes a combination of a plurality of programs to be multiplexed according to a reception area. In the multiplexing apparatus used by the transmission side of the IP broadcast system of Claim 1,
Frame generating means for multiplexing the plurality of RTP packets and generating a standardized multiplexed frame including at least a header area and an information area;
The frame generation means includes
A plurality of input buffers for temporarily holding the plurality of RTP packets for each program;
An arithmetic unit that obtains a delay amount from the buffer occupancy amount of each of the plurality of input buffers, reads out and outputs a plurality of RTP packets from the corresponding input buffers to be multiplexed into the multiplexed frame based on these delay amounts, and
A header generation unit that generates data of a header area of the multiplexed frame based on the plurality of RTP packets;
A multiplexing apparatus comprising: an output buffer that stores a plurality of RTP packets output by the arithmetic unit and stores and outputs data in a header area of the multiplexed frame . The frame generation means further comprises rate detection means for detecting a transmission rate of each of the plurality of RTP packets,
Based on the detection result by the rate detection means, the arithmetic unit sets a program having a maximum transmission rate among a plurality of programs to be multiplexed in the multiplexed frame as a master program, and a transmission rate of the master program × a rate of the number of programs. 6. The multiplexing apparatus according to claim 5, wherein a plurality of corresponding RTP packets are multiplexed. 6. The multiplexing apparatus according to claim 5 , wherein the frame generation unit changes a combination of a plurality of programs to be multiplexed according to a reception area. In the receiving apparatus used on the receiving side of the IP broadcast system according to claim 1,
Receiving means for receiving, from the broadcast network, a multiplexed frame that is multiplexed with the plurality of RTP packets on the transmission side of the IP broadcast system and includes at least a header area and an information area;
A buffer for temporarily holding the multiplexed frame;
The multiplexed frame is fetched from the buffer and analyzed to extract a reproduction clock shared by the plurality of programs, and a plurality of TSs are separated from the multiplexed frame for each program, and each TS is based on the reproduction clock. And a program reproducing means for reproducing the program from the receiver. 9. The receiving apparatus according to claim 8 , wherein the program reproducing means performs reproduction processing by extracting a reproduction clock of a master program designated by the transmission side. A plurality of TSs (Transport Streams) having packet structures with different programs and reproduction clocks are converted into RTP (Real-time Transport Protocol) packets for IP (Internet Protocol) network transmission, respectively, and transmitted to a broadcast network on the IP network. In the method
A program having a maximum transmission rate among a plurality of programs to be transmitted to the broadcast network is set as a master program,
Obtain the delay amount for each of the master program and the plurality of programs following the master program,
Determine whether this delay amount is larger than the target delay amount,
For a program having a delay amount larger than the target delay amount, a plurality of corresponding RTP packets are stored in the output buffer,
For a program having a delay amount equal to or less than the target delay amount, a plurality of corresponding RTP packets and dummy data for the deficient portion are stored in the output buffer,
A method for generating and outputting a standardized multiplexed frame including at least a header area and an information area based on a plurality of RTP packets stored in the output buffer.

Images