JP2005277461A - Data transmission system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a data transmission system in which data flow is ensured with no failure by transmitting data consumption of the MPEG decoder in a receiver and following up data generation amount by controlling STC of the MPEG encoder in a transmitter. <P>SOLUTION: In order to transmit/receive an image signal, a voice signal and incidental information stably over a long time through a network (LAN 3), data consumption of the MPEG decoder 16a in a receiver 40 is transmitted to a transmitter 30, and reference frequency signal STC of the MPEG encoder 8 in the transmitter 30 is controlled by a transmission data buffer 31, a residual amount of data detector 32, and the like, to follow up transmission amount of data from the transmitter 30 to a receiver 2. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a data transmission system, and in particular, in a system in which a receiver and a transmitter are connected by LAN technology such as wired Ethernet (registered trademark) connection or radio wave wireless connection, the transmitter is synchronized with a read request from the receiver. The present invention relates to a data transmission system.

  FIG. 12 is a schematic configuration diagram of a conventional data transmission system that transmits images, sounds, and incidental information via a network.

  In this system, a transmitter 1 and a receiver 2 are connected by a LAN 3. The transmitter 1 includes a video decoder 5, a frame synchronizer 7 with A / D, an MPEG encoder 8, an A / D converter 9, a PLL unit 10, and a TCP / IP interface unit 11 (UDP / IP) of the transmitter. IPO), and a fixed reference frequency generator XO12 and the like.

  The receiver 2 includes a TCP / IP interface unit 15 (which may be UDP / IP) of the receiver, an MPEG decoder 16, an XO 17 (also referred to as a reference signal generation unit 17) which is a fixed reference frequency generator, and a PLL. Part 21 and D / A converters 18 and 20.

  The operation of the conventional system configured as described above will be described. (Operation of Transmitter 1) An image signal (video) of the NTSC system or the like sent from the outside to the transmitter 1 is input to the video decoder 5, and the image signal from the video decoder 5 is subjected to frame synchronization with A / D. The signals are input to the nizer 7 and are sequentially digitally converted to be synchronized with the frame and input to the MPEG encoder 8. The audio signal (audio) is input to the A / D converter 9 of the transmitter 1, and the audio signal is converted based on the output from the PLL unit 9 and output to the MPEG encoder 8.

  The MPEG encoder 8 inputs NTSC frame image and audio data and performs compression processing based on a reference frequency signal STC (also referred to as STC of a transmitter) from the XO 12, and converts the compressed frame image and audio data into a compressed frame image and audio data. The incidental information is synthesized and packetized and multiplexed to PS (used for DVD) or TS, and supplied to the TCP / IP interface or UDP / IP interface (hereinafter referred to as the interface unit 11 of the transmitter).

  The interface unit 11 of the transmitter 1 generates a transmission packet to be transmitted on the LAN 3 by prepending a UDP header or a TCP header and an IP header to the multiplexed TS or PS packet, and connects to the LAN 3. to manage. In the case of a TCP / IP interface, a transmission packet is transmitted on the LAN 3 to a node that has established a connection, or in the case of a UDP / IP interface, to a predetermined node.

(Receiver operation)
On the other hand, the UDP / IP interface or the UDP / IP interface (hereinafter referred to as the interface unit 15 of the receiver) of the receiver 2 takes out various packets flowing on the LAN 3 and outputs them to the MPEG decoder 16.

  For example, when the interface unit 15 of the receiver is a TCP / IP interface, a packet from the transmitter 1 that has established a connection is selected and stored in the memory of the MPEG decoder 16. In the case of the UDP / IP interface, after a predetermined packet is selected and stored in the memory of the MPEG decoder 16, the payload MPEG-TS or PS is extracted from the transmission packet, and the MPEG decoder 16 of the receiver 2 Store in memory.

  The MPEG decoder 16 decodes the packet stored in the memory or the MPEG-TS or PS packet based on the reference frequency signal (also referred to as the STC of the 27 MHz receiver) of the reference signal generator 17, and D / A A video signal is obtained via the video encoder 19 by being output to the converter 18 for analog conversion.

  Further, the MPEG decoder 16 decodes the packet stored in the memory or the MPEG-TS or PS packet based on the reference frequency signal (27 MHz, the reference frequency signal STC of the receiver) of the reference signal generator 17, Is output to the audio D / A converter 20 to output an analog-converted audio signal.

  Such a system is generally used, for example, for the purpose of transmitting camera video in real time from a remote location via the LAN 3.

  Since the transmitter 1 outputs the packet encoded by the MPEG encoder 8 to the MPEG decoder 16 of the receiver 2 and immediately decodes it, the amount of transmission data generated by the MPEG encoder 8 of the transmitter 1 and the receiver 2 are decoded. It is necessary to match the amount of transmission data consumed by the MPEG decoder 16 within the same time so that no overflow or underflow of transmission data occurs during transmission.

  For this reason, TS established by PCR information processing as means for reproducing the reference frequency signal STC (= 27 MHz) of the MPEG encoder 8 of the transmitter 1 is generally adopted in the MPEG multiplexing system. .

  At this time, the transmitter 1 transmits a transmission packet regardless of the situation of the receiver 2, and the receiver 2 reads the PCR information in the packet and sends a reference frequency signal STC for using the MPEG decoder 16 to the transmitter 1. The PLL is configured by hardware so as to obtain a frequency that matches the frequency.

  Even when the packet multiplexing method is PS, the STC playback method is provided in the same manner as in the standard, but generally PS is used for TS mainly used for broadcasting, such as DVD and HDD. This is a time of reproduction from the storage medium. At this time, since the MPEG encoder is absent, the STC synchronization function is not necessary. For this reason, the STC synchronization function has been omitted from the conventional MPEG decoder chip that uses the PS as an input, and that amount has been a factor in greatly reducing the cost of the decoder itself.

On the other hand, JP 2000-92130 A (Patent Document 1) discloses jitter removal and clock recovery for an A / V stream transmitted via a network including jitter without using a PLL circuit at the client. Processing is described.
JP 2000-92130 A

  However, in order to reduce the cost of the system to a popular price, it is desirable to use an existing chip such as a DVD playback application for the MPEG decoder in the receiver. In this case, as described above, the decoder chip includes an STC. Since the reproduction function is omitted, the STC of the MPEG encoder of the transmitter cannot be reproduced in real time transmission in which the MPEG encoder of the transmitter functions simultaneously.

  For this reason, the amount of data per unit time generated by the MPEG encoder of the transmitter is different from the amount of data per unit time consumed by the MPEG decoder of the receiver, and an overflow or underflow occurs during transmission. The loss of data occurs in one of the devices, and the processing of the MPEG decoder of the receiver is broken. That is, interference such as block noise, freeze, sound blockage, and incidental information error occurs in the image output by reception.

  As a countermeasure against these, there is a method using TS, but there is a problem that the MPEG decoder of the receiver becomes expensive as described above.

  For example, when a TS is used even in a system where a pair of transceivers is sufficient, TS is accompanied by complicated multiplex processing because the TS is originally intended for broadcasting. There was a problem that large processing became essential.

  In addition, PS can be transmitted at an arbitrary time with an arbitrary data length, whereas TS adds time information and PCR information at the time of transmission to a fixed-length packet of 188 bytes, and the receiving side starts the time from the received TS packet. Since the stamp is extracted, the delay time of the transmission path must be constant.

  In general broadcasting, the transmission path has a substantially constant delay. However, in a network such as a LAN, fluctuation of the delay time is not avoided, and a PCR with jitter added to the PLL processing of the receiver is supplied. There was a problem that stable STC reproduction was difficult.

  In other words, when the UDP / IP or TCP / IP header is simply added to the MPEG stream and transmitted to the network, there are problems with TS and PS respectively.

  The present invention has been made to solve the above problems, and realizes stable transmission and reception of image signals, audio signals, and incidental information over a network over a long period of time. The amount of data consumption is transmitted to the transmitter, and the STC of the MPEG encoder of the transmitter is controlled so that the amount of generated data can be followed and the data flow can be ensured without failure.

  The present invention is a data transmission system in which an image signal, an audio signal, and incidental information are compressed and encoded, and then packetized and transmitted between a transmitter and a receiver connected via a network.

The receiver
Separating means for receiving the packet and separating the compressed image signal, audio information, and auxiliary information from the packet, and temporarily storing the image signal, audio signal, and auxiliary information separated by the separating means A first storage means for reading, a decoding means for reading out a predetermined amount of image signal, audio signal, and supplementary information from the first storage means, and decoding and decompressing, and constant for the decoding means to decode and decompress A first reference signal generating means for generating a frequency signal and a request signal for causing the transmitter to transmit the packet when the storage capacity of the first storage means becomes smaller than a predetermined value Request signal generating means.

And the transmitter is
Encoding means for compressing and encoding the image signal, audio signal and incidental information; second reference signal generating means for generating a variable frequency signal for the encoding means to compress and encode; When the request signal is supplied by the second storage means for temporarily storing the image signal, the audio signal and the auxiliary information compressed and encoded by the encoding means, and the request signal generating means, The transmission means for transmitting a predetermined amount of packets from the two storage means and the frequency generated by the second reference signal generation means are controlled so that the storage capacity of the second storage means falls within a predetermined range. And a control means for doing so.

  In addition, the present invention is a data transmission system in which an image signal, an audio signal, and incidental information are compressed and encoded, and then packetized and transmitted between a transmitter and a receiver connected via a network.

The receiver
Separating means for receiving the packet and separating the compressed image signal, audio information, and auxiliary information from the packet, and temporarily storing the image signal, audio signal, and auxiliary information separated by the separating means A first storage means for reading, a decoding means for reading out a predetermined amount of image signal, audio signal, and supplementary information from the first storage means, and decoding and decompressing, and constant for the decoding means to decode and decompress A first reference signal generating means for generating a frequency signal and a request signal for causing the transmitter to transmit the packet when the storage capacity of the first storage means becomes smaller than a predetermined value Request signal generating means.

And the transmitter is
Encoding means for compressing and encoding the image signal, audio signal and incidental information; second reference signal generating means for generating a variable frequency signal for the encoding means to compress and encode; When the request signal is supplied by the second signal storage means for temporarily storing the data composed of the image signal, the audio signal and the incidental information compressed and encoded by the encoding means, and the request signal generating means, Requesting a predetermined amount based on the request signal to the second storage means, and transmitting means for transmitting the predetermined amount of the data packet; and a storage capacity of the second storage means is a predetermined amount When the predetermined amount is requested from the transmission means and the control means for controlling the frequency generated by the second reference signal generation means so as to be a range value, this is equally divided into arbitrary data lengths Divided amount Sequentially, wherein for the second storage means and said request signal is outputted to the gist that a means for transmitting said data from said transmitting means at a plurality of packets.

  As described above, according to the present invention, the receiver and the transmitter are connected via the network, the receiver transmits the data consumption of the MPEG decoder to the transmitter, and the transmitter encodes based on the remaining data amount of the encoding. By increasing or decreasing the reference frequency for the transmitter, the encoding of the transmitter and the decoding of the receiver are synchronized.

  Therefore, it is possible to realize a low-cost device that transmits images, sounds, and incidental information stably over a long period of time via a network.

  Further, according to the present invention, the receiver and the transmitter are connected by a network, the receiver transmits the data consumption of the MPEG decoder to the transmitter, and the transmission data buffer dividing means of the transmitter has the encoding processing capability. The data consumption requested from the receiver is divided on the basis of the above and the transmitter is sequentially sent to the transmitter in the divided amount.

  For this reason, even if the demand from the receiver is large, it is possible to realize a low-cost device that transmits images, sounds, and incidental information in a stable state over a long time with a simple configuration.

<Embodiment 1>
The first embodiment makes it possible to stably transmit and receive image signals, audio signals, and incidental information over a network (LAN 3) for a long time. Data consumption of the MPEG decoder 16a of the receiver 40 Underflow by transmitting the amount of data sent from the transmitter 30 to the receiver 40 by controlling the reference frequency signal STC of the MPEG encoder 8 of the transmitter 30. The overflow is not caused.

  FIG. 1 is a schematic configuration diagram of a data transmission system according to the first embodiment. As shown in FIG. 1, a transmitter 30 and a receiver 40 are connected via a LAN 3.

  The transmitter 30 includes a video decoder 5, a frame synchronizer 7 with A / D, an MPEG encoder 8, an A / D converter 9, a PLL unit 10, and a transmitter interface unit 11 (may be UDP / IP). A data remaining amount detector 32, a high frequency attenuator 33a, an average value calculator 33b, a D / A converter 34, an LPF 35, a VCXO 36, and the like.

  The receiver 40 includes a receiver-side interface unit 15a, an MPEG decoder 16a, a receiver-side reference signal generation unit 17 similar to FIG. 12, a PLL unit 21, D / A converters 18 and 20, It has.

(Configuration of each part of transmitter)
The video decoder 5 receives a video signal (NTSC system or the like), converts the subcarrier color signal into a color difference signal, and outputs it to the frame synchronizer 7 with A / D together with the luminance signal.

  The frame synchronizer 7 with A / D digitally converts the luminance signal and the color difference signal from the video decoder 5 and synchronizes the reference frequency signal STC to the MPEG encoder 8 and the input video signal to the MPEG encoder 8. Output (hereinafter referred to as video data).

  In other words, the frame synchronizer function described above is used when the video signal from the video decoder 5 is a non-standard frame frequency, such as a VHS reproduction signal, or a signal including jitter. A signal suitable for the MPEG encoder 8 standard is obtained by correcting to a standard frame frequency by processing including (not shown) and removing jitter. The sample frequency of the A / D conversion described above is an output signal (also called a clock) based on the reference frequency signal STC (27 MHz; VCXO 36) of the MPEG encoder 8.

  The A / D converter 9 digitally converts the audio signal and outputs it to the MPEG encoder 8. Here, the sampling frequency of the A / D converter 9 for the audio signal is, for example, a frequency that is an integral multiple of 44.1 KHz generated by the dedicated PLL 10 from the same 27 MHz.

  The MPEG encoder 8 compresses and encodes the video data and audio data from the A / D-attached frame synchronizer 7 with the reference frequency signal STC of the VCXO 36, and adds a program stream PS (audio data and video data are added with accompanying information). The data Di is output to the transmission data buffer 31 at the subsequent stage at a time when compressed data of a certain unit amount, for example, 256 kilobytes, is generated.

  The transmission data buffer 31 stores the data Di from the MPEG encoder 8 and sequentially transmits the data Di based on a transmission request (including the requested data amount) from the interface unit 11 of the transmitter.

  When the data from the transmission data buffer 31 is supplied, the interface unit 11 of the transmitter 30 divides the data into data lengths suitable for the frequency of packet traffic on the LAN 3, and the TCP header (in the case of UDP, UDP header) The packet is added with an IP header. Then, a receiver (also simply referred to as a node) or an arbitrary receiver that has established a connection is identified and transmitted to the LAN 3.

The remaining data amount detector 32 monitors the transmission data buffer 31 and detects the remaining amount di (byte amount) of the data Di in the transmission data buffer 32. For example, the difference between the previous data remaining amount d _i−1 and the current data remaining amount di is obtained, and the degree of increase / decrease Qi (consisting of a sign indicating increase or decrease and the amount ki) is determined from this difference. .

  When the degree Qi indicates an increase, the reference frequency control signal for decreasing the reference frequency signal STC of the transmitter 30 is expressed by an amount ki (difference amount ki between the previous and current data remaining amount di). Generate based on

  The high frequency attenuator 33a and the average value calculator 33b remove the relatively high frequency fluctuation of the time axis of the detection result of the data amount remaining amount detector 32 as jitter, and the average value calculator 33b An abnormal detection value is removed, and negative feedback is performed to the VCXO 36 (27 MHz) via the D / A converter 34 and the LPF 35.

  That is, the remaining data di of the transmission data buffer 31 is fed back to the VCXO 36 that generates the reference frequency signal STC (27 MHz) of the MPEG encoder 8 after removing high-frequency jitter and removing abnormal detection values. Will be.

(Receiver configuration)
The TCP / IP interface unit 15a of the receiver 40 (hereinafter referred to as receiver interface unit: UDP / IP may be used) from a transmitter or an arbitrary transmitter 30 with which a connection is established from a packet traveling on the connected LAN 3. A packet which is generated and whose destination is specified by the receiver itself is selected, and the MPEG-PS which is a payload is separated from the packet and supplied to the MPEG decoder 16a. In the case of the UDP / IP interface, after a predetermined packet is selected and stored, MPEG-PS as a payload is taken out from the packet and supplied to the MPEG decoder 16a of the receiver. Further, the transmission request (including the data amount) from the MPEG decoder 16a is transmitted as a packet to the transmitter 30.

  The MPEG decoder 16a performs a decoding operation by a clock (reference frequency signal STC of the receiver) from the reference signal generator 17 (XO17) which is a 27 MHz fixed frequency transmitter, and is output from the interface unit 15a on the receiving side. Is stored in a buffer (not shown), and each time a certain amount is stored, the compressed and encoded video data and audio data are separated, respectively decompressed and decoded, and then the D / A converter 18. And a desired amount (a certain amount, for example, 256 kBytes) is determined for each decompression decoding process and transmitted to the transmitter 30 via the interface unit 15 of the receiver and the LAN 3.

  The D / A converter 18 converts the video data from the MPEG decoder 16a into analog and outputs it to an external device (not shown) via the video decoder 19.

  The D / A converter 20 converts the audio data from the MPEG decoder 16a to analog based on the reference signal (44.1 KHz) from the PLL 21, and outputs the analog data to an external device (not shown).

(Description of operation)
The operation of the system according to the first embodiment configured as described above will be described below.

  A video signal (NTSC system or the like) supplied to the transmitter 30 is input to the video decoder 5, and the subcarrier color signal is converted into a color difference signal by the video decoder 5, and a frame synchronizer with A / D conversion along with the luminance signal. 7 is output.

  Then, the A / D conversion-equipped frame synchronizer 7 performs A / D conversion based on the reference frequency signal STC (27 MHz) from the VCOX 36.

  The audio signal is digitally converted by the A / D converter 9 and input to the MPEG encoder 8.

  Next, the MPEG encoder 8 compresses and encodes these digitized video signal and audio signal to generate a program stream PS, and each time the generated data has a unit amount, for example, a data amount of 256 kilobytes is generated. The data Di is sequentially supplied to the transmission data buffer 31 and a write request is supplied to the transmission data buffer 31.

  The configuration of the transmission data buffer 31 will be described with reference to FIG. As shown in FIG. 2, the transmission data buffer 31 is a memory having two ports, a write data input port (Din, write) and a read data port (Dout, read). However, the transmission data buffer 31 necessarily has two ports as hardware. It doesn't have to be memory. For example, any functionally identical DRAM or the like that can be written and read asynchronously with a CPU controlled by the CPU may be used.

  The transmission data buffer 31 reads out a certain unit amount, for example, every 256 kilobytes, and sends it to the interface unit 11 of the transmitter in response to a program stream PS per fixed time and a transmission data read request described later. When the interface unit 11 of the transmitter sends out the above-mentioned unit amount (256 kbytes) of data Di from the transmission data buffer 31, the transmitter interface unit 11 divides the data Di into data lengths suitable for the frequency of packet traffic on the LAN 3 and transmits the data Di. A header is added, or a UDP header is added after being divided into an arbitrary data length, and a receiver or an arbitrary receiver with which connection is established is specified and transmitted on the LAN 3.

  On the other hand, the interface unit 15a of the receiver selects a packet that is generated from a transmitter or any transmitter that has established a connection and has a destination specified by the receiver itself, from a packet that travels on the connected LAN. The data Di is separated from the packet and sent to the MPEG decoder 16a of the receiver.

  The MPEG decoder 16a is operated by the clock (27 MHz) of the reference signal generator 17 (XO17) which is a fixed frequency transmitter that becomes the reference frequency signal STC of the receiver, and the compressed and encoded video that is the supplied data Di. The signal and the audio signal are separated, each is decompressed and decoded, converted into an analog signal by the D / A converters 18 and 20, and output from the receiver through the video encoder 19.

  The video encoder 19 is a multiplier that converts a color difference signal into a subcarrier color signal. However, a digital data multiplier may be provided between the D / A converter 18 and the MPEG decoder 16a.

  In addition to the above-described packets (including video and audio signals), it is also possible to transmit auxiliary information from the transmitter 30 to the receiver 40. For example, the aspect ratio of the image, the number of audio channels, etc. Information set by 8 and required by the MPEG decoder 16a of the receiver may be transmitted as data Di in a predetermined storage location in the PS.

  Alternatively, incidental information such as the production date of the image / audio signal may be supplied from the outside of the transmitter to the MPEG encoder 8 of the transmitter, stored in a position similar to the PS user area, and transmitted to the receiver 40. .

  Here, when the MPEG decoder 16a of the receiver decodes a certain amount of data in the buffer, for example, 256 kBytes of data and consumes it from the buffer, it continuously sends a request for data to the transmitter 30 via the interface unit 15a of the receiver. To do.

  Here, the time change of the remaining data amount di remaining in the transmission data buffer 31 is shown in FIG. FIG. 3 shows a state where encoding and decoding are balanced.

  3, times tw1 to tw3 indicate times when the MPEG encoder 8 generates a data write request and the program stream PS is written to the transmission data buffer 31.

  Tr1 to tr3 are times when the data Di of the program stream PS is read by a data read request from the MPEG decoder 16a of the receiver.

  In FIG. 3, when the remaining data di of the program stream PS in the transmission data buffer 31 is xkByte, the data Di of the program stream PS is written in the transmission data buffer 31 at times tw1, tw2, and tw3, and the remaining data di is It becomes x + nkByte, and is read by the data read request from the receiver 40 at times tr1, tr2, tr3, and becomes the remaining amount of xkByte again. That is, when the amount of encoded data and the amount of decoded data are balanced, the data is reciprocated by a data amount of nkBytes at the timing of reading and writing.

  Therefore, in order to stabilize the remaining amount of data in the transmission data buffer for a long time and avoid overflow or underflow, the unit of data generation and consumption is made the same, and the amount per time is determined. It is only necessary to synchronize 27 MHz of the reference frequency signal STC used by the MPEG encoder and decoder of each transceiver.

  For example, an example in which the cycle of the reference frequency signal STC is not taken is shown in FIGS. 4 and 5 (reading from the receiver is fast).

  When the reference frequency signal STC of the receiver from the reference signal generation unit (XO 17) of the receiver 40 is higher than the reference frequency signal STC of the receiver of the VCXO 36 of the transmitter 30, data writing is generated as shown in FIG. The time from tw1 to tr1 (tw1 to tr1, tw2 to tr2, tw3 to tr3) decreases, and finally data read occurs before the data write occurs as shown in FIG. If the amount di decreases by 1 unit amount (nkByte), and this decrease is repeated, it eventually leads to underflow.

  If the frequency difference between the respective reference frequency signals STC is 60 ppm corresponding to the allowable limit of 27 MHz ± 30 ppm of the MPEG standard, the frame period of the video signal generated by decoding with the MPEG decoder is MPEG. If it is about 2 microseconds shorter than that determined by the encoder 8 and is 30 frames / second, a shift corresponding to one frame is reached after 9.26 seconds.

  The time when the underflow is reached is determined by the frequency difference of the reference frequency signal STC, the amount of buffer built in the MPEG decoder 16a and the transmission rate of the stream, but since the built-in buffer is finite, the frequency of each reference frequency signal STC As long as there is a difference, underflow is inevitable.

  In order to solve such a flow failure, in the present embodiment, as shown in FIG. 1, the transmitter 30 includes a transmission data buffer 31, a remaining data detector 32, and a relatively high frequency fluctuation of the detection result. The high frequency attenuator 33a to be removed, the average value calculator 33b to eliminate abnormal detection values, and the LPF 35 are provided, and the result of the remaining data di in the transmission data buffer 31 is obtained as the reference frequency signal STC of the MPEG encoder 8. Negative feedback is applied to the VCXO 36 (27 MHz) that generates.

  For example, when the remaining data amount decreases, the reference frequency signal STC of the transmitter is lower than the reference frequency signal STC of the receiver as described above, so the frequency of the reference frequency signal STC from the VCXO 36 (27 MHz) increases. Thus, a control voltage (increasing the voltage) is generated from the DA converter 34.

  That is, since the data remaining amount detector 32 of the transmitter 30 can stably monitor the transmission data buffer 31 is tw shown in FIG. 3, the data Di of the program stream PS is transmitted from the MPEG encoder 8. Each time data is written to the data buffer 31, the data remaining amount di of the transmission data buffer 31 is detected, and the previous data remaining amount is subtracted.

  If the subtraction result is positive, the remaining data di of the data Di from the MPEG encoder 8 is increased, so that a frequency control signal for lowering the reference frequency from the VCXO 36 is generated, and the control voltage from the PLL 35 to the VCOX 36 is generated. Lower.

  On the other hand, if the value is negative, the remaining data di is decreased, so a reference frequency control signal for increasing the control voltage from the PLL 35 to the VCOX 36 is transmitted. In other words, if the detection value subtraction result is inverted and transmitted to the VCXO 36, data transmission can be performed without overflow and underflow of the buffer.

  Here, since the data transmission request generated from the receiver 40, that is, the data read signal to the transmission data buffer 31, is transmitted via the LAN 3, the delay time may not be constant, and the time shown in FIG. The interval between tr and the next tr may increase or decrease.

  At this time, since the detection value of the remaining data di in the transmission data buffer 31 includes an error and jitter occurs in the frequency of the VCXO 36, for example, a high frequency attenuator 33a that attenuates a frequency of several millihertz or more is provided to influence the jitter. Is eliminated. Also, an average value calculation circuit that accumulates detection values two times before or after and subtracts them with the average value with them, or accumulates the subtraction values in the same manner and uses the average value as a subtraction result. By providing 33b, an abnormal value of the detected value can be eliminated and stable control of the VCXO 36 can be obtained.

  As a result, there is no difference between STC transceivers, and the remaining amount of the transmission data buffer 31 is stabilized for a long time, so that the occurrence of a data flow failure can be eliminated. That is, stable STC reproduction is possible.

<Embodiment 2>
FIG. 6 is a schematic configuration diagram of a system according to the second embodiment. As shown in FIG. 6, the system according to the second embodiment is configured by connecting a receiver 40 similar to the first embodiment and a transmitter 50 having the transmission data buffer 51 according to the second embodiment via a LAN 3. It is.

  The transmitter 50 includes a video decoder 5 similar to that in the first embodiment, a frame synchronizer 7 with A / D, an MPEG encoder 8, an A / D converter 9, a PLL unit 10, and an interface unit of the transmitter. 11 (which may be UDP / IP), a remaining data detection unit 32, a high frequency attenuator 33a, an average value calculator 33b, a D / A converter 34, a VCXO 36, and the like. In the second embodiment, the transmission data buffer 51 having the determination unit 53 is provided.

  As shown in FIG. 7, the transmission data buffer 51 includes a memory 54 (RAM) having two ports, a write data input port (Din, write), a read data port (Dout, read), and a determiner 53. doing.

  The determination unit 53 determines the amount of read data every time there is a read request transmitted from the receiver side, determines the number of transmissions of data Di for one read from this read data amount, The number of determination results is sent from the memory 54 to the receiver 40 side.

  In addition, the receiver 40 includes a TCP / IP interface 15a (which may be UDP / IP), an MPEG decoder 16a, an XO 17 (27 MHz), and a D / A converter 18 similar to those in FIG. , 20, PLL 21, video encoder 19, etc., data is separated from the packet from the transmitter 50, and the amount of data in the decoding / decompression is constant while decoding / decompressing the image, audio and incidental information When reaching, a packet transmission request is sent to the transmitter.

  A system according to the second embodiment configured as described above will be described.

  A video signal (NTSC system or the like) supplied to the transmitter 50 is input to the video decoder 5, and the subcarrier color signal is converted into a color difference signal by the video decoder 5, and a frame synchronizer with A / D conversion is added together with the luminance signal. 7 is output.

  Then, the A / D conversion-equipped frame synchronizer 7 performs A / D conversion based on the reference frequency signal STC (27 MHz) from the VCOX 36.

  The audio signal is digitally converted by the A / D converter 9 and input to the MPEG encoder 8.

  Next, the MPEG encoder 8 compresses and encodes these digitized video signal and audio signal to generate a program stream PS, and each time the generated data has a unit amount, for example, a data amount of 256 kilobytes is generated. The data (PS) is sequentially supplied to the transmission data buffer 51.

When the interface unit 11 of the transmitter sends a read request from the receiver 40 to the transmission data buffer 51, the determiner 53 reads the write data amount n indicated by the write request from the MPEG encoder 8 and is set in advance. The number of times of reading m is obtained based on the amount of reading d per time. In this embodiment,
Number of readings m = n / d
Asking.

  For example, when the data request amount of the transmission data read request from the receiver 40 is 256 kilobytes and the write data amount n is 2 kilobytes, it is determined that 2 kilobytes are transmitted 128 times.

  Further, when the write data amount n from the MPEG encoder 8 of the transmitter 50 is not an integral multiple of the aforementioned single read amount d, the determiner 53 determines that the value of d * m is n or more. A small integer is determined as m. That is, the data of 2 kBytes is sent out from the memory 54 128 times.

  Therefore, as shown in FIG. 8, when the data Di of the MPEG encoder 8 is written to the transmission data buffer 51 at the timing of tw1, tw2, tw3, a read request (256 kBytes of 256 kBytes) is issued at the time of tr1, tr2, tr3. When there is a request), as described above, the determiner 53 determines that 2 kBytes are divided into 128 outputs and outputs them, and outputs 2 kBytes for 128 times to the read of the memory 54. As shown in FIG. In response to the read request for tr1, tr2, and tr3, data of 2 kBytes is sequentially output (in FIG. 8, it is an inclined portion).

  As shown in FIG. 9, when the frequency of the reference frequency signal STC of the receiver is faster than the frequency of the reference frequency signal STC of the transmitter, the intervals between tw1 and tr1, tw2 and tr2, and tw3 and tr3 are shortened. .

  Even at this timing, if the determiner 53 receives a read request (request of 256 kBytes) from the receiver 40 at the times tr1, tr2, and tr3, the determiner 53 divides 2 kBytes into 128 times as described above. Since it is determined to be output and 2 kBytes are output to the read of the memory 54 for 128 times, as shown in FIG. 8, 2 kByte data is sequentially output in response to a read request of tr1, tr2, tr3 (FIG. 8). Is an inclined portion).

  Further, when the frequency of the reference frequency signal STC of the receiver is faster than the frequency of the reference frequency signal STC of the transmitter, tr3 is input before tw3 as shown in FIG.

However, as described above, the determiner 53 is
Number of readings m = n / d
Since d is output m times from the memory 54, as shown in FIG. 10, at tr1, the difference between x + nkbyte and xkByte is sequentially read at intervals of 2 kBytes (inclined portion in FIG. 10). Next, when tr2 enters before tw2, there is no nkByte, so a certain amount d is read between xkByte and tr2 to tw3 (for example, 2 kByte).

  When tw2 is entered, nkByte from x-akByte is written. That is, the remaining amount is x + n-akByte.

  Further, when tr3 enters before tw3 more quickly, there is no nkByte, so d is sequentially read from trk and tw3 from xkByte at the timing of tr3 (inclination part). When the amount read from tr3 to tw3 is b, x-bkByte to nkByte are written at tw3. That is, the write amount at tw3 is x + n−bkByte.

  Here, the description of the processing of the determination unit 53 will be supplemented using the flowchart of FIG. As shown in FIG. 11, the determination unit 53 determines whether or not there is a data read request from the receiver side (S1).

  If it is determined in step S1 that there is a data read request, the requested amount is read (S2), and the number of reads based on the above equation is obtained for this requested amount (S3).

  Next, a predetermined read amount d set in advance is read and output to the memory 54 (S4). Next, with the output of the read amount d, the number of outputs is counted (S5), and it is determined whether or not the count number ri = m has been reached (S6). In step S6, when the count number ri has not reached m, the process returns to step S4.

  In step S6, if the count number ri has reached m, it is determined whether or not the process is finished (S7). If not, the process returns to step S1.

  On the other hand, a data remaining amount detector 32, a high frequency attenuator 33a for removing a relatively high frequency fluctuation of the detection result as jitter, an average value calculator 33b for eliminating abnormal detection values, and an LPF 35 are provided. The result of the remaining data is negatively fed back to the 27 MHz VCXO 36 (27 MHz) that generates the reference frequency signal STC of the MPEG encoder 8.

  For example, when the remaining data amount decreases, the reference frequency signal STC of the transmitter is lower than the reference frequency signal STC of the receiver as described above, so the frequency of the reference frequency signal from the VCXO 36 (27 MHz) increases. Thus, a control voltage (increasing the voltage) is generated from the DA converter 34.

  That is, each time the data Di of the program stream PS is written from the MPEG encoder 8 to the transmission data buffer 51, the remaining data detector 32 detects the remaining data di and subtracts the previous remaining data.

  Here, in the second embodiment, the unit amount of change in the remaining amount of data that can be detected is d, which is smaller than the same unit amount (= n) in the first embodiment, so that the detection accuracy is improved. The follow-up operation of STC can be made faster.

  If the subtraction result is positive, the remaining data amount of the MPEG encoder 8 is increased, so that a frequency control signal for lowering the reference frequency from the VCXO 36 is generated, and the control voltage from the PLL 35 to the VCOX 36 is lowered.

  On the other hand, if the value is negative, the remaining amount of data is reduced, so a reference frequency control signal for increasing the control voltage from the PLL 35 to the VCOX 36 is transmitted. That is, if the detection value subtraction result is inverted and transmitted to the VCXO, data transmission can be performed without overflow and underflow of the transmission data buffer 51.

  In the first and second embodiments, the hardware device is used. However, if the processing is the same, software processing can also be realized. For example, the MPEG decoding process is an off-the-shelf decoder chip that is advantageous in terms of cost, but may be a so-called software MPEG decoder on a personal computer.

It is a schematic block diagram of the compressed data transmission system of this Embodiment 1. It is a block diagram of a transmission data buffer. It is explanatory drawing explaining reading and writing when an encoder and a decoder are balanced. It is explanatory drawing explaining the reading and writing in case reading of a decoder is quick. It is explanatory drawing explaining the reading and writing in case reading of a decoder is faster. FIG. 3 is a schematic configuration diagram of a data transmission system according to a second embodiment. 6 is a schematic configuration diagram of a transmission data buffer according to a second embodiment. FIG. FIG. 10 is an explanatory diagram for explaining reading and writing by an encoder and a decoder according to a second embodiment. FIG. 10 is an explanatory diagram for explaining reading and writing by an encoder and a decoder according to a second embodiment. FIG. 10 is an explanatory diagram for explaining reading and writing by an encoder and a decoder according to a second embodiment. 10 is a flowchart for explaining the operation of the determiner according to the second embodiment. It is a schematic block diagram of the conventional transmission / reception system.

Explanation of symbols

5 Video Decoder 7 Frame Synchronizer with A / D 8 MPEG Encoder 9 A / D Converter 10 PLL Unit 11 Transmitter TCP / IP Interface Unit 16a MPEG Decoder 30 Transmitter 32 Data Remaining Data Detection Unit 33a High Frequency Attenuator 33b Average Value calculator 34 D / A converter 36 VCXO
40 Receiver

Claims (2)

A data transmission system that compresses and encodes an image signal, an audio signal, and incidental information, and then packetizes and transmits between a transmitter and a receiver connected via a network,
The receiver
Separating means for receiving the packet and separating the compressed image signal, audio information and incidental information from the packet;
First storage means for temporarily storing the image signal, the audio signal and the incidental information separated by the separation means;
Decoding means for reading out a predetermined amount of image signal, audio signal and auxiliary information from the first storage means, and decoding and decompressing;
First reference signal generating means for generating a constant frequency signal for the decoding means to decode and decompress;
Request signal generating means for generating a request signal to cause the transmitter to transmit the packet when the storage capacity of the first storage means becomes smaller than a predetermined value;
With
The transmitter is
An encoding means for compressing and encoding the image signal, the audio signal, and the accompanying information;
Second reference signal generating means for generating a variable frequency signal for compression encoding by the encoding means;
Second storage means for temporarily storing the image signal, the audio signal, and the auxiliary information compressed and encoded by the encoding means;
A transmission means for transmitting a predetermined amount of packets from the second storage means when the request signal is supplied by the request signal generating means;
Control means for controlling the frequency generated by the second reference signal generating means so that the storage capacity of the second storage means becomes a value within a predetermined range;
A data transmission system comprising: A data transmission system that compresses and encodes an image signal, an audio signal, and incidental information, and then packetizes and transmits between a transmitter and a receiver connected via a network,
The receiver
Separating means for receiving the packet and separating the compressed image signal, audio information and incidental information from the packet;
First storage means for temporarily storing the image signal, the audio signal and the incidental information separated by the separation means;
Decoding means for reading out a predetermined amount of image signal, audio signal and auxiliary information from the first storage means, and decoding and decompressing;
First reference signal generating means for generating a constant frequency signal for the decoding means to decode and decompress;
Request signal generating means for generating a request signal to cause the transmitter to transmit the packet when the storage capacity of the first storage means becomes smaller than a predetermined value;
With
The transmitter is
An encoding means for compressing and encoding the image signal, the audio signal, and the accompanying information;
Second reference signal generating means for generating a variable frequency signal for compression encoding by the encoding means;
Second storage means for temporarily storing data composed of an image signal, an audio signal and incidental information compressed and encoded by the encoding means;
When the request signal is supplied by the request signal generation means, a request for requesting a predetermined amount based on the request signal to the second storage means, and transmission for transmitting the predetermined amount of the data packet Means,
Control means for controlling the frequency generated by the second reference signal generating means so that the storage capacity of the second storage means becomes a value within a predetermined range;
When the predetermined amount is requested from the transmission unit, the division amount obtained by equally dividing the predetermined amount into an arbitrary data length is sequentially output to the second storage unit, and the request signal is transmitted from the transmission unit to the second storage unit. Means for sending data in multiple packets;
A data transmission system comprising:

Images