Data communication system

Description: TECHNICAL FIELD [Detailed description of the invention]

【0001】

[Technical field to which the invention belongs]

The present invention is a data that realizes a service for communicating data having a predetermined periodic structure such as moving image data with high quality via a communication network such as the Internet that employs a datagram type communication protocol. Regarding communication systems. There is an IEEE1394 standard by the Institute of Electrical and Electronics Engineering (IEEE) as a standard for synchronous communication of moving image data such as digital video on a local bus. In recent years, with the spread of devices such as digital video cameras and high-performance personal computers to general users, local networks configured according to the IEEE1394 standard are connected to each other via, for example, the Internet, and comply with the IEEE1394 standard. There is an increasing need for technology to relay data streams.

【0002】

[Conventional technology]

For example, digital video data is formed by a huge number of frames, and each frame contains image data and audio data for one video frame. As shown in FIG. 14 (a), one frame of digital video data is composed of 10 DIF sequences (shown with serial numbers in the figure), and each DIF sequence is shown in FIG. 14 (a). As shown in b), it is composed of 150 DIF blocks.

【0003】

Here, in FIG. 14, the DIF blocks represented by the reference numerals f, the reference numerals a, and the reference numerals v are information units including frame header information, audio information, and image information, respectively, and each DIF block is shown in FIG. 14 (c). As shown in, the data for 77 bytes is added with the identification information represented by 3 bytes. The data structure shown in FIG. 14 represents a logical format of digital video data, and when actually transferring data in a network, a format corresponding to the characteristics of the network is adopted.

【0004】

For example, when transferring digital video data via a local bus that complies with the IEEE1394 standard, as shown in Fig. 15 (a), predetermined headers (CIP header and ISO header) and CRC are added to the six DIF blocks. The IEEE1394 packet is formed, and the transfer operation is performed using this IEEE1394 packet as a transfer unit. In this case, each DIF sequence is divided into 25 IEEE1394 packets, as shown in FIG. 14 (b) separated by vertical bars, so one frame of digital video data is shown in FIG. 15 (b). As such, it is forwarded as 250 IEEE1394 packets.

【0005】

In the IEEE1394 standard, in order to adjust the synchronization communication timing, an empty packet for timing adjustment may be inserted every 15 to 16 synchronization packets.

【0006】

This empty packet is a packet formed only from the synchronization header and the CIP header, and the actual amount of data transferred by the synchronization packet for one frame is fixed regardless of the number of empty packets inserted in one frame. .. In addition, the IEEE1394 standard stipulates that 8000 IEEE1394 packets are transferred per second in the synchronous transfer mode, which enables the transfer of video / audio devices such as display devices and speakers connected to the local bus. It enables playback of 30 frames per second of image and audio information.

【0007】

Therefore, by using a gateway connected to both a local bus that supports the IEEE1394 standard and another type of network such as the Internet, it is the same as the first network that is connected by the local bus that supports the IEEE1394 standard. In principle, it is possible to connect to the second network of the above via a third network of another type and exchange the above-mentioned IEEE1394 packets with each other.

【0008】

In addition, a data communication system using such a gateway has also been proposed. For example, there is a data communication system proposed by Keio University Graduate School of Media and Governance and the Faculty of Environment and Information Studies of the same university ("Internet DV Transfer Technology Using Frame Exclusive Technology" by Kazunori Sugiura et al. Information TECHNICAL REPORT OF IEICE.CPSY99-33 (1999-05) PP77-81).

【0009】

As shown in FIG. 16, this data communication system includes a gateway 410 that relays between the first network and the Internet corresponding to the IEEE1394 standard, and a gateway 420 that relays between the same second network and the Internet. And have. In FIG. 16, the IEEE1394 packet sent to the first network by the digital video camera 401 is passed to the digital video transmission unit (DV transmission unit) 412 via the IEEE1394 adapter 411.

【0010】

The digital video transmitter 412 performs frame thinning processing, and as shown in FIG. 17, header information consisting of an IP header, a UDP (User Datagram Protocol) header, and an application header is added to the IEEE1394 packet to be transmitted. The added, IP-encapsulated packet is sent to the Internet via the Internet adapter 413.

【0011】

The packet thus IP-encapsulated is passed to the digital video receiver (DV receiver) 421 via the Internet adapter 413, decapsulated by the digital video receiver 421, and then the IEEE1394 adapter 411. It is sent to the second network via. In this way, for example, the video and audio captured by the digital video camera 401 connected to the first network is passed to the second network via the Internet, and the digital video deck 402 connected to the second network. Allows recording / playback.

【0012】

In this data communication system, in the encapsulation process by the digital video transmission unit 412, an application header including an adjustment parameter indicating information about frame thinning and a sequence number is added, and in the decapsulation process in the digital video reception unit 421. , Restoring the IEEE1394 packet stream according to the information in this application header.

【0013】

For details on the frame thinning process and the IEEE1394 packet stream restoration process in this data communication system, refer to the above-mentioned documents.

【0014】

[Problems to be Solved by the Invention]

In the above-mentioned data communication system, UDP is adopted as a communication protocol of the transport layer of the Internet, and each IEEE 1394 packet constituting the IEEE 1394 packet stream in the first network is sent to the second network via the Internet as a datagram. Has been passed.

【0015】

Since UDP is a connectionless communication protocol, it is possible to perform broadcast-style communication in which the same data is broadcast from one sender to a large number of receivers. It is suitable for the purpose of exchanging data between the 1st network and the 2nd network via the Internet. On the other hand, UDP only has an error detection mechanism as a mechanism for maintaining transfer quality, and for loss and joke of IP-encapsulated packets, fluctuations in the order of arrival at the receiving side, etc., the sending and receiving side You need to deal with it in your application.

【0016】

Further, for example, when the packet stream in the above-mentioned IEEE1394 synchronous communication mode is synchronously communicated between the sender and the receiver, it is necessary to keep the transmission delay from the sender to the receiver constant. However, in a "best effort" type network such as the Internet, both the transfer speed and the transfer delay fluctuate depending on the network traffic, and the transmission delay on the receiving side is caused by the loss or junk of IP-encapsulated packets. It fluctuates significantly.

【0017】

Moreover, the probability of packet loss or junking varies depending on network traffic. Therefore, in order to realize the transmission of packet streams in the IEEE1394 synchronous communication mode via a network that employs a datagram communication protocol, such as the Internet, along with a technology that compensates for the deterioration of transfer quality due to the relay network, A technology that suppresses large fluctuations in transmission delay is required.

【0018】

The present invention provides a data communication system that transmits data having a known data structure between a plurality of physically independent independent networks via a relay network that employs a datagram-type communication protocol.

【0019】

[Means for solving problems]

FIG. 1 shows a principle block diagram of the data communication system according to claims 1 to 4. The invention of claim 1 is provided corresponding to a first network, which is one of a plurality of physically independent independent networks that transfer structural data having a predetermined regular structure in predetermined transfer units. The transmitting side relay means 110 transmits the transfer unit transferred in the first network to another relay network, and the receiving side relay means provided corresponding to the second network which is at least one of a plurality of independent networks. In the data communication system that relays the transfer unit transmitted to the relay network to the second network by 120, the transmitting side relay means 110 includes the transfer unit according to the input of the transfer unit transferred in the first network. The receiving side relay means 120 is configured to include a forming means 111 for forming a datagram format relay packet destined for the receiving side relay means 120 and a transmitting means 112 for sending the relay packet to the relay network. , Obtained by the decomposition means 121, which decomposes the relay packet received via the relay network and separates the transfer units, and the decomposition means 121 based on the information about a predetermined regular structure that the structural data should have. The configuration is characterized by including a restoring means 122 for restoring structural data using a transfer unit and an output means 123 for outputting structural data to a second network for each predetermined transfer unit.

【0020】

The invention of claim 1 is to relay a relay packet including a transfer unit transferred in the first network by the forming means 111 and the transmitting means 112 provided in the transmitting side relay means 110 via the relay network. The original structural data is restored from a series of relay packets by the disassembling means 121 and the restoring means 122 provided in the receiving side relay means 120, and the original structural data is restored by the means 120 and subjected to the output processing by the output means 123. Structural data transferred for each transfer unit in the first network can be relayed to the second network.

【0021】

According to the second aspect of the present invention, in the data communication system according to the first aspect, the plurality of independent networks transfer digital video data in predetermined transfer units, and the restoration means provided in the receiving side relay means 120. 122 is a detection means 124 that detects information indicating the beginning of a video frame included in the input transfer unit, and from the detection of the beginning of the video frame by the detection means 124 to the detection of the beginning of the next video frame. Reform means 125 that reshapes the structural data for one video frame using the transfer unit received from the decomposition means 121, and reshapes by the reshape means 125 based on the transfer unit input to the reshape means 125. The completeness of the structural data reformed by the reforming means 125 based on the collecting means 126 that collects the reforming information regarding the amount of information contained in the structural data to be performed and the detection result and the reforming information by the detecting means 124. It is characterized by having an evaluation means 127 for evaluating the property and a result output means 128 for outputting the structural data reformed by the reforming means 125 as a restoration result according to the evaluation result by the evaluation means 127. And.

【0022】

According to the second aspect of the present invention, the structural data is reformed by the reforming means 125 according to the detection result by the detecting means 124 provided in the restoring means 122 of the receiving side relay means 120, and the collecting means 126 and the evaluation means 127 are formed. By operating, the integrity of this structural data can be evaluated. Further, by operating the result output means 128 according to this evaluation result, for example, only the structural data evaluated to have a sufficiently high degree of perfection is selectively output as the restoration result, and the processing of the output means 123 is performed. Can be offered to.

【0023】

According to the invention of claim 3, in the data communication system according to claim 2, the collecting means 126 detects the beginning of a video frame until the next beginning is detected according to the detection result by the detecting means 124. A counting means 131 for counting the number of transfer units input to the reforming means 125 is provided, and the reforming information including the counting value is sequentially output. The evaluation means 127 is included in the reforming information. The first determination means 132 that monitors the count value, determines that the structural data reformed by the reforming means 125 is incomplete when the count value exceeds a predetermined threshold, and outputs the evaluation result. It is characterized by having a provided configuration.

【0024】

According to the third aspect of the present invention, the number of transfer units included in the structural data being reformed is abnormal due to the operation of the first determination means 132 according to the count value by the counting means 131 provided in the collecting means 126. When the number becomes large, it can be evaluated that this structural data is incomplete. As a result, for example, the transfer unit containing the information indicating the beginning of the video frame is missing in the transmission process in the relay network, and the transfer unit belonging to the two video frames is combined as one structural data by the reforming means 125. If it does, it can be reliably eliminated as this structural data is incomplete.

【0025】

The invention of claim 4 is the data communication system according to claim 2, wherein the collecting means 126 includes the detection result by the detecting means 124 and the information regarding the data length included in each transfer unit input to the reforming means 125. Based on this, a measuring means 133 for measuring the amount of information included in the structural data reformed by the reforming means 125 is provided, and the reforming information including the measurement result is output, and the evaluation means 127 is the reforming means 127. Based on the measurement result included in the information, the estimation means 134 that estimates the number of transfer units that contributed to the reform processing by the reforming means 125, and the estimation result by the estimating means 134 and the comparison result with a predetermined threshold value. , A configuration including a second determination means 135 that determines whether or not the amount of information missing when reforming the structural data by the reforming means 125 is within the permissible range, and outputs this determination result as an evaluation result. The result output means 128 outputs the structural data reformed by the reforming means 125 as a valid restoration result when the evaluation result indicating that the missing information amount is within the permissible range is input. It is characterized in that it is configured to be used for output processing to the second network by the output means 123.

【0026】

The invention of claim 4 is a transfer unit included in the structural data reformed by the reforming means 125 by the operation of the measuring means 133 provided in the collecting means 126 and the estimating means 134 provided in the evaluation means 127. Information on the number of data is obtained, and based on this information, the second determination means 135 determines whether or not the lack of information in the transmission process via the relay network is within the permissible range, and according to this determination result. , Controls the output operation of structural data by the result output means 128.

【0027】

In this way, by subjecting the structural data whose information loss is within a predetermined allowable range to the processing of the output means 123 as the restoration result, the structural data to be sent to the second network as the restoration result corresponding to a large number of video frames. Can be secured. FIG. 2 shows a principle block diagram of the data communication system according to claim 5.

【0028】

The invention of claim 5 is the data communication system according to claim 1, wherein a plurality of independent networks transfer digital video data for each predetermined transfer unit, and a restoration means provided in the receiving side relay means 120. 122 is a configuration in which structural data is restored in video frame units using the transfer unit received from the decomposition means 121, and output means 123 holds the structural data for one video frame restored by the restoration means 122. The first holding means 136, the second holding means 137 holding the structural data for one video frame to be transmitted, and each transfer unit held by the second holding means 137 are transmitted to the second network according to a predetermined procedure. Each time the transmission means 138 and the transmission means 138 transmit the structural data for one video frame, the first holding means 136 is referred to, and a third determination for determining whether or not new structural data is held is made. It is characterized by having a configuration including means 139 and an input means 140 for inputting structural data held in the first holding means 136 into the second holding means 137 according to a judgment result by the third judging means 139. To do.

【0029】

According to the fifth aspect of the present invention, when the input means 140 operates according to the determination result by the third determination unit 139 and the transmission of the restoration result corresponding to each video frame is completed, the first holding means 136 is used. If the restoration result corresponding to the next video frame is not prepared, the transmitted structural data held in the second holding means 137 is input to the transmitting means 138 instead of the structural data corresponding to the next video frame. can do.

【0030】

FIG. 3 is a principle block diagram of the data communication system according to claims 6 to 7. According to the invention of claim 6, in the data communication system according to claim 1, the plurality of independent networks transfer digital video data in predetermined transfer units, and the transmitting side relay means 110 is in the first network. Based on the frame counting means 141 that counts the video frames transferred as structural data, the counting value by the frame counting means 141, and the thinning rate specified by the thinning instruction, the packet forming operation by the forming means 111 is stopped and restarted. It is characterized in that the configuration is provided with the formation control means 142 for controlling.

【0031】

According to the invention of claim 6, in the transmitting side relay means 110, the formation control means 142 operates according to the count value by the frame counting means 141 and the input thinning information, and the packet formation process by the forming means 111 is videotaped. By stopping and resuming on a frame-by-frame basis, a part of the structural data transferred in the first network is selectively converted into a series of relay packets according to the thinning rate indicated by the above-mentioned thinning information, and is passed through the relay network. And can be sent to the second network side.

【0032】

In this way, frame thinning is realized by the operation of the formation control means 142 and the formation means 111, and a part of the structural data is selectively relayed by the transmission side relay means 110 to be transmitted via the relay network. Since the amount of information can be reduced, the burden on the relay network can be reduced and the probability of information loss can be reduced.

【0033】

According to the invention of claim 7, in the data communication system according to claim 6, the transmitting means 112 provided in the transmitting side relay means 110 is a packet holding means 143 that holds a series of packets formed by the forming means 111. , The interval calculation means 144 that calculates the transmission interval at which packets should be sent to the relay network according to the thinning rate specified in the thinning instruction, and the packets held by the packet holding means 143 are sequentially taken out for each transmission interval. It is characterized in that the configuration is provided with the packet output means 145 to be transmitted to the relay network.

【0034】

According to the invention of claim 7, when the frame thinning is performed by the operations of the formation control means 142 and the formation means 111, the packet output means 145 operates according to the transmission interval obtained by the interval calculation means 144. As a result, a series of relay packets held in the packet holding means 143 can be sent to the relay network at a transmission interval according to the thinning rate, regardless of the transfer timing of each transfer unit in the first network.

【0035】

As a result, the traffic in the relay network can be averaged, the load on the relay network can be reduced, and the probability of information loss can be reduced.

【0036】

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

【0037】

FIG. 4 shows an embodiment of the data communication system of the present invention. In the data communication system shown in FIG. 4, the two relay devices 210s and 210r, together with the IEEE1394 adapter 411 and the Internet adapter 413, respectively, and the digital video (DV) transmitter 211 that converts digital video data into a relay packet described later. , It is provided with a digital video (DV) receiver 221 that converts relay packets into digital video data as described later.

【0038】

In these relay devices 210s and 210r, the IEEE1394 adapter 411 is connected to the serial bus corresponding to the first network and the second network, respectively, and is defined by the IEEE1394 synchronization mode in which the transfer is performed using a predetermined synchronization channel. A packet in this format (hereinafter, simply referred to as a synchronous packet) is extracted by the IEEE1394 adapter 411 and input to the digital video transmission unit 211.

【0039】

In this digital video transmission unit 211, the packet formation unit 212 operates in response to an instruction from the transmission control unit 213, adds an appropriate header to the synchronization packet received from the above-mentioned IEEE1394 adapter 411, and transmits it via the Internet. It converts to a suitable relay packet. Further, the transmission buffer 214 has a function of receiving a series of relay packets from the packet forming unit 212 and temporarily holding the packet, and the packet reading unit 215 receives an instruction from the transmission control unit 213 and temporarily holds the transmission buffer. It has a function to sequentially read the relay packets held in 214 and input them to the Internet adapter 413.

【0040】

On the other hand, in the digital video receiving unit 221 shown in FIG. 4, the capsule disassembling unit 222 has a function of decomposing the relay packet described above and separating the synchronous packet, and the obtained synchronous packet is sequentially received. It is held in buffer 223. Further, the integrity evaluation unit 224 has a function of evaluating the integrity of the digital video data composed of the synchronized packets held in the reception buffer 223 as described later.

【0041】

Further, according to the evaluation result, the result output unit 225 shown in FIG. 4 reads out a series of synchronization packets held in the reception buffer 223 and inputs them to the IEEE1394 adapter 411. Further, FIGS. 5 and 6 show a detailed configuration of the digital video transmitter and a detailed configuration of the digital video receiver.

【0042】

In the transmission control unit 213 shown in FIG. 5, the frame detection unit 231 has a function of detecting a synchronization packet including information indicating the beginning of a video frame from a series of synchronization packets input to the packet formation unit 212. There is.

【0043】

Further, in FIG. 5, the frame counter 232 has a function of counting the number of video frames to be composed of a series of synchronized packets input to the packet forming unit 212 based on the detection result by the frame detecting unit 231. The count value by the frame counter 232 is input to the formation control unit 233. The formation control unit 233 has a function of creating a formation instruction for stopping or restarting the relay packet formation process by the packet formation unit 212 based on the above-mentioned count value and the thinning information held in the control information holding unit 234. This formation instruction is input to the packet forming unit 212 as an instruction from the transmission control unit 213.

【0044】

Further, the interval calculation unit 235 has a function of calculating an appropriate packet transmission interval based on the above-mentioned thinning information, and the packet transmission interval obtained by the interval calculation unit 235 is notified to the transmission control unit 236. Has been done. The transmission control unit 236 has a function of outputting a read instruction for each time notified as a packet transmission interval, and this read instruction is input to the packet read unit 215 as an instruction from the transmission control unit 213. There is.

【0045】

On the other hand, in the completeness evaluation unit 224 shown in FIG. 6, the frame detection unit 241 sets the synchronization packet containing the information indicating the beginning of the video frame from the series of synchronization packets input to the reception buffer 223 as the boundary of the video frame. The detection result is input to the packet counter 242 and the data length counter 243.

【0046】

The packet counter 242 and the data length counter 243 shown in FIG. 6 perform counting operations according to the detection result by the frame detection unit 241 and the input of the synchronization packet to the reception buffer 223, respectively, and the boundary indicated by the detection result described above. It has a function to calculate the total number of synchronization packets contained in one video frame separated by and the number of bytes of the data part contained in these synchronization packets.

【0047】

Further, in the integrity evaluation unit 224 shown in FIG. 6, the comparators 244a and 244b have a function of comparing the count value by the packet counter 242 or the data length counter 243 with the predetermined thresholds Tha and Thb, respectively. The comparison results by these comparators 244a and 244b are input to the integrity determination unit 245. The integrity determination unit 245 is held in the reception buffer 223 based on the comparison result by the comparator 244a described above and the comparison result by the comparator 244b at the time when the boundary of the video frame is detected by the frame detection unit 241. It has a function of determining the completeness of the video frame to be reformed by the set of synchronous packets, and the determination result is notified to the result output unit 225 as the evaluation result by the completeness evaluation unit 224.

【0048】

In the result output unit 225 shown in FIG. 6, the data reading unit 246 sequentially outputs synchronization packets for one video frame held in the reception buffer 223 according to the evaluation result notified from the integrity evaluation unit 224 described above. Read to and input to the send queue 247. Further, the data output unit 248 sequentially takes out a series of synchronization packets held in the output buffer 249 and sends them to the IEEE1394 adapter 411.

【0049】

On the other hand, the rewrite control unit 250 shown in FIG. 6 refers to the transmission queue 247 described above each time the data output unit 248 outputs a synchronization packet for one video frame, and a new synchronization packet is added to the transmission queue 247. Controls the rewriting operation of the output buffer 249 depending on whether or not is held. Further, the output buffer 249 receives and holds a series of synchronization packets held in the transmission queue 247 in response to the instruction from the rewriting control unit 250 described above.

【0050】

Hereinafter, the correspondence between each part shown in FIGS. 4 to 6 and each means described in claims 1 to 7 will be shown. In FIG. 4, the transmitting side relay means 110 described in claim 1 is formed of an IEEE1394 adapter 411 provided in the relay device 210s, a digital video transmitting unit 211, and an Internet adapter 413, while the receiving side. The relay means 120 is composed of an IEEE1394 adapter 411 provided in the relay device 210r, a digital video receiver 221 and an Internet adapter 413.

【0051】

The packet forming unit 212 shown in FIG. 4 corresponds to the forming means 111 described in claim 1, and the transmission buffer 214, the packet reading unit 215, and the Internet adapter 413 are the transmission described in claim 1. It corresponds to means 112. On the other hand, the capsule disassembling unit 222 shown in FIG. 4 corresponds to the disassembling means 121 described in claim 1, and the receiving buffer 223, the integrity evaluation unit 224, and the result output unit 225 correspond to the restoring means 122. It is something to do.

【0052】

Further, the result output unit 225 and the IEEE1394 adapter 411 shown in FIG. 4 fulfill the functions of the output means 123 described in claim 1. Further, the frame detection unit 231 and the frame counter 232 shown in FIG. 5 correspond to the frame counting means 141 described in claim 6, and the thinning information held by the formation control unit 233 in the control information holding unit 234. The function of the formation control means 142 is fulfilled by operating according to the above.

【0053】

Further, the transmission buffer 214 shown in FIGS. 4 and 5 corresponds to the packet holding means 143 described in claim 7, and is billed by the interval calculation unit 235 operating in response to the above-mentioned thinning information. The function of the interval calculation means 144 described in item 7 is fulfilled. Further, the transmission control unit 236 controls the output operation of the packet reading unit 215 to the Internet adapter 413 according to the calculation result by the interval calculation unit 235, so that the function of the packet output means 145 described in claim 7 can be obtained. It has been fulfilled.

【0054】

On the other hand, the frame detection unit 241 shown in FIG. 6 corresponds to the detection means 124 described in claim 2. Further, the packet counter 242 and the data length counter 243 correspond to the counting means 131 described in claim 3 and the measuring means 133 described in claim 4, respectively, and in FIG. 6, the collection described in claim 2 The means 126 is formed from the packet counter 242 and the data length counter 243 described above.

【0055】

Further, by operating the integrity determination unit 245 according to the comparison result by the comparator 244a shown in FIG. 6, the function of the first determination means 132 described in claim 3 is fulfilled, while the comparison is performed. By operating the integrity determination unit 245 according to the comparison result by the device 244b, the functions of the estimation means 134 and the second determination means 135 described in claim 4 are fulfilled.

【0056】

Further, according to the determination result by the integrity determination unit 246, the data reading unit 246 inputs the synchronization packet held in the reception buffer 223 into the transmission queue 247 and uses it for the processing of the data output unit 248. As a result described in 4, the function of the output means 128 is fulfilled. Further, the above-mentioned comparators 244a and 244b and the integrity determination unit 245 together form the evaluation means 127 described in claim 2.

【0057】

Further, the transmission queue 247 and the output buffer 249 shown in FIG. 6 correspond to the first holding means 136 and the second holding means 137 described in claim 5, respectively, and the rewriting control unit 250 makes a third determination. It corresponds to means 139. Further, the output buffer 249 operates in response to an instruction from the rewrite control unit 250 to fulfill the function of the input means 140 described in claim 5, while the transmission means 138 described in claim 5. The function of is performed by the data output unit 248 and the IEEE1394 adapter 411 shown in FIG.

【0058】

Next, FIG. 4 shows an example of relaying a synchronization packet sent to the serial bus by the digital video camera 401, which is a node belonging to the first network, to the digital video deck 402, which is a node belonging to the second network. The operation of the data communication system will be described. FIG. 7 shows a flow chart showing a packet forming operation and a relay packet sending operation by the digital video transmitter.

【0059】

Each time a synchronization packet is input by the IEEE1394 adapter 411, the frame detection unit 231 shown in FIG. 5 matches the first 3 bytes of the data unit of the input synchronization packet with the bit pattern indicating the beginning of the video frame. It is determined whether or not it is the beginning of a new video frame based on whether or not it is performed (steps 301 and 302 in FIG. 7A).

【0060】

In the case of the affirmative determination of step 302, the frame counter 232 increments the count value C1 indicating the number of input video frames (step 303), and then proceeds to step 304, while the negative determination of step 301. In that case, the process proceeds to step 304 as it is. In step 304, the formation control unit 233 determines whether or not to convert the input synchronization packet into a relay packet based on the count value C1 of the frame counter 232 and the thinning information held in the control information holding unit 234. judge.

【0061】

For example, when the control information holding unit 234 holds the thinning information to the effect that the video frames are thinned out at a thinning rate of 1/2, the formation control unit 233 determines that the count value C1 by the frame counter 232 is an odd number. It may be determined that the synchronization packet input to the packet forming unit 212 is the conversion target to the relay packet. In the case of the affirmative determination in step 304, the packet forming unit 212 adds a predetermined header to the input synchronization packet and converts it into a relay packet in response to the instruction from the forming control unit 233 (step 305).

【0062】

At this time, the packet forming unit 212 may add an IP header and a UDP header destined for the network address of the relay device 210r shown in FIG. 4 to the beginning of the input synchronization packet. The relay packets thus obtained are sequentially held in the transmission buffer 214 (step 306).

【0063】

After that, it is determined whether or not the input of all the synchronization packets is completed (step 307), and in the case of the negative determination, the process returns to step 301 and the processing for the new synchronization packet may be continued. On the other hand, in the case of the negative determination in step 304, the packet forming unit 212 may discard the input synchronous packet in response to the instruction from the forming control unit 233 (step 308), and then proceed to step 307.

【0064】

In this way, the formation control unit 233 controls the operation of the packet formation unit 212 according to the count value of the frame counter 232, so that a part of the video frame output by the digital video camera 401 to the serial bus according to the thinning information. It is possible to selectively convert the synchronous packet constituting the above into a relay packet and use it for the transmission processing of the relay packet via the transmission buffer 214.

【0065】

As a result, the amount of information transmitted to the Internet can be reduced, and the transmission load on the Internet can be reduced. The thinning information described above may be input to the control information holding unit 234 prior to starting the relay operation of the digital video data, for example. Further, the format of the thinning information is not particularly limited, and any information may be used as an index for determining whether or not to thin out the corresponding video frame by comparing with the number of input frames.

【0066】

Next, the operation of sending the relay packet held in the transmission buffer 214 to the Internet as described above will be described. The interval calculation unit 235 shown in FIG. 5 calculates the transmission interval of the relay packet based on the thinning information held in the control information holding unit 234 prior to the start of transmission of the relay packet (FIG. 7 (b). ) Step 311).

【0067】

At this time, the interval calculation unit 235 may calculate the transmission interval in consideration of the transfer time corresponding to the video frames thinned out according to the above-mentioned thinning information. For example, as described above, when video frames are thinned out at a thinning rate of 1/2, as shown in FIG. 8 (a), in addition to the transfer time corresponding to the video frame selected as the relay target, the thinning out is performed. The transfer time corresponding to the video frame (shown by the dotted line in FIG. 8A) can be used to send out each relay packet constituting the video frame to be relayed.

【0068】

Therefore, the interval calculation unit 235 is the reciprocal of the value obtained by subtracting the thinning rate δ from the numerical value 1 to the transfer interval T1 of the synchronous packets on the serial bus constituting the first network (that is, 1 / (1-δ)). To obtain the transmission interval TS by multiplying by, and notify the transmission control unit 236. After that, the packet reading unit 215 sequentially reads relay packets from the transmission buffer 214 in response to an instruction from the transmission control unit 236 and inputs the relay packets to the Internet adapter 413 (step 312). The packet is sent to the Internet.

【0069】

Next, it is determined whether or not a relay packet that has not been transmitted remains in the transmission buffer 214 (step 313), and in the case of an affirmative determination, the transmission control unit 236 repeats step 314 to elapse the transmission interval TS described above. When the transmission time TS has elapsed, the process returns to step 312 as an affirmative determination in step 314, and the transmission operation of the next relay packet may be started.

【0070】

In this way, the transmission interval of each relay packet constituting the relay target video frame is determined in consideration of the transfer time corresponding to the video frame excluded by the thinning process, and is shown in FIG. 8 (b). As described above, it is possible to distribute the timing of sending the relay packet to the Internet.

【0071】

In this case, the band used for transmitting the relay packet can be reduced according to the amount of information reduced by thinning out the video frames, so that the transmission load on the Internet can be effectively reduced. In addition, the average reduction of traffic on the Internet reduces the probability of packet loss on the transmission path over the Internet, resulting in more complete video on the receiving second network. There is a high probability that the frame can be restored.

【0072】

In a "best effort" network such as the Internet, the correlation between an increase in transmission load and a decrease in communication quality is clear. Therefore, by reducing the transmission load as described above, it is used for relay. The probability of occurrence of missing packets is suppressed, and the effect of suppressing deterioration of communication quality can be expected. Next, the operation of the digital video receiver will be described.

【0073】

FIG. 9 shows a flow chart showing the operation of restoring digital video data. As shown in FIG. 9, in response to the input of the relay packet, the capsule decomposition unit 222 shown in FIG. 6 first removes the UDP header and the IP header from the relay packet to separate the synchronization packet (as shown in FIG. 9). Steps 321, 322). At this time, the frame detection unit 241 shown in FIG. 6 determines whether or not the synchronization packet separated by the capsule decomposition unit 222 includes the head portion of the video frame in the same manner as the frame detection unit 231 described above. (Step 323).

【0074】

In the case of the negative determination in step 323, the packet counter 242 increments the count value PC (step 324), and the data length counter 243 refers to the synchronization header of the synchronization packet separated by the capsule separator 222 and data. The value DM obtained by subtracting the data length of the CIP header from the data length DL of the part is added to the count value DS (step 325).

【0075】

Next, the integrity determination unit 245 determines whether or not the count value PC by the packet counter 242 exceeds the threshold value Tha based on the comparison result by the comparator 244a (step 326). In the case of the negative determination in step 326, the reception buffer 223 sequentially accumulates the synchronization packets received from the capsule decomposition unit 222 (step 327).

【0076】

After that, it is determined whether or not the input of all the synchronization packets is completed (step 328), and in the case of the negative determination of this step 328, the process returns to step 321 and the reception operation of the new relay packet may be started. .. In this way, the synchronization packets of the first network included in the relay packets received via the Internet are sequentially accumulated in the reception buffer 223, and the synchronization packets including the beginning part of the new video frame are input. At that time, the affirmative judgment of step 323 is obtained.

【0077】

At this time, the count value DS of the data length counter 243 indicates the total data length of the data part included in the synchronization packet held in the reception buffer 223, and is a series of synchronization packets held in the reception buffer 223. One video frame of digital video data consisting of synchronized packets transferred in the first network is reconstructed.

【0078】

Here, in the transmission path in the relay network (for example, the Internet), as shown in FIG. 10A, a part of the relay packet sent to the transmission path as one video frame (shaded in the figure). If) arrives missing, the digital video data for one video frame reconstructed in the receive buffer 223 will be incomplete.

【0079】

However, if the number of missing relay packets is small, the video and audio played using the incompletely reconstructed digital video data will retain acceptable quality, albeit with some noise. doing. In consideration of this, in response to the affirmative determination in step 323 described above, whether the integrity determination unit 245 exceeds the threshold Thb in the count value Ds by the data length counter 243 based on the comparison result by the comparator 244b. If it is determined whether or not (step 329), and in the case of an affirmative determination, the data reading unit 246 inputs the contents of the receive buffer 223 into the transmission queue 247 in response to an instruction from the integrity determination unit 245 (step 330). ), Digital video data can be selectively passed to the IEEE1394 adapter 411 by the result output unit 225 for video frames that meet the quality criteria indicated by the threshold Thb.

【0080】

In this case, a complete video frame with 250 synchronous packets, as well as a video frame that is incompletely reconstructed due to the lack of relay packets in the transmission path, will be sent to the second network, and will be sent to the second network individually. If you look at video frames, their quality deteriorates. However, as described above, by allowing video frames that are incomplete but can be expected to meet certain criteria, a sufficient number of video frames are sent to the second network regardless of the transmission quality in the transmission path. It is possible to guarantee the quality of video / audio software consisting of a huge number of video frames.

【0081】

The above-mentioned threshold Thb value is based on the result of an experiment for examining the quality of the reproduced video obtained when some synchronization packets are missing, and the probability of missing relay packet expected in the transmission path. Then, an appropriate value may be obtained. In this way, after the synchronization packet for one video frame held in the reception buffer 223 is output, the reception buffer 223 holds the synchronization packet received from the capsule decomposition unit 222 as the synchronization packet at the beginning of the video frame. (Step 331).

【0082】

Further, the packet counter 242 and the data length counter 243 may set the initial value 1 and the initial value DL to the respective count values, respectively (step 332), and proceed to step 328. On the other hand, in the case of the negative determination in step 329 described above, the receive buffer 223 discards the series of synchronized packets it holds in response to the instruction from the integrity determination unit 245 (step 333), and then in step 331. You can proceed to.

【0083】

By the way, as shown in FIG. 10 (b), when a relay packet (shown with shading in the figure) including the beginning of a video frame is missing and arrives in the transmission path, the packet counter 242 Count value by: The PC continues to be incremented in response to the input of sync packets that would otherwise be counted as another video frame, exceeding the total number n of sync packets for one video frame.

【0084】

In such a case, it is considered that the two video frames (shown as frame 1 and frame 2 in FIG. 10B) are held in the receive buffer 223 in an inseparable state, and thus are held in the receive buffer 223. It is not possible to reconstruct digital video data using a series of synchronized packets. At this time, the integrity determination unit 245 may proceed to step 333 as an affirmative determination in step 326 described above.

【0085】

Here, in the synchronous transfer mode according to the IEEE1394 standard, since the transmitting node inserts empty packets for adjusting the transmission timing, the total number n of the synchronous packets constituting one video frame is the synchronous packet having a data unit (hereinafter, hereinafter, The total number of valid packets) is the sum of Ps and the number of empty packets inserted, and the total number is not constant. However, since the insertion interval of empty packets is limited, 20 or more empty packets are not inserted in one video frame.

【0086】

Therefore, for example, if the value obtained by adding the total number Pe of empty packets that may be inserted to the total number Ps of valid packets constituting one video frame is input to the comparator 244a as the threshold value Tha, the value of the video frame can be input. It is possible to reliably detect the omission of the synchronization packet including the first part. As a result, it is possible to eliminate an invalid video frame in which synchronous packets belonging to two video frames are combined, and to reduce the influence of low transmission quality on the relay network on the second network.

【0087】

If the number of packets excluding empty packets is counted according to the detection result by the frame detection unit 241 and the input of the synchronization packet, the evaluation process equivalent to the above-mentioned integrity evaluation unit 224 is performed based on this count value. It is possible to do. For example, as shown in FIG. 11, instead of the data length counter 243 shown in FIG. 6, a packet discriminating unit 251 is provided to form a completeness evaluation unit 224, and the synchronous packet input by the packet discriminating unit 251 is provided. It may be determined whether or not is a valid packet, and the packet counter 242 may be configured to increment the count value PC according to the determination result.

【0088】

The packet determination unit 251 may determine whether or not the synchronization packet is an empty packet, for example, based on the information regarding the data length of the data unit shown in the synchronization header of each synchronization packet. In this case, the total number of valid packets for one video frame (Ps = 250) is input to the comparator 244a as the threshold value Tha, and the allowable number of missing packets Pd is subtracted from the total number of valid valid packets Ps described above. The obtained value may be input to the comparator 244b as the threshold value Thb.

【0089】

Next, the operation of outputting the reconstructed digital video data to the serial bus via the IEEE1394 adapter 411 as described above will be described. FIG. 12 shows a flow chart showing the output operation of digital video data. Further, FIG. 13 shows a diagram illustrating an output operation of digital video data. At the output start timing of each video frame (described later), the rewrite control unit 250 refers to the transmission queue 247 and determines whether or not the digital video data to be output to the serial bus is input to the transmission queue 247 (described later). Steps 341, 342).

【0090】

As shown in FIG. 13 (a), when the output of the k-1st video frame (shown as frame (k-1) in the figure) is completed, the kth video frame (in the figure, the frame (in the figure) When all the synchronization packets constituting (shown as k)) have been received and the digital video data for one video frame having an acceptable quality is reconstructed from these synchronization packets, the above-mentioned As described above, the data reading unit 246 shown in FIG. 6 inputs a series of synchronization packets held in the reception buffer 223 to the transmission queue 247.

【0091】

In this way, when the transmission queue 247 is updated with new digital video data, the rewrite control unit 250 instructs the output buffer 249 to rewrite the contents as a positive determination in step 342 described above (step 343). In response, output buffer 249 receives a series of synchronization packets held in transmit queue 247 and holds them for output (step 344).

【0092】

In this case, the data output unit 248 executes the output operation to the IEEE1394 adapter 411 in step 345, so that the synchronization packets constituting the kth video frame (shown as frame (k) in the figure) are sequentially generated. Is output to the serial bus. In this way, when the output of the digital video data for one video frame is completed, the data output unit 248 notifies the rewrite control unit 250 of the arrival of the output start timing of the new video frame (step 346). The output operation for this video frame may be terminated.

【0093】

On the other hand, for example, when the transmission side adjusts the transmission interval of the relay packet by thinning out the video frame, or the arrival of the relay packet is delayed due to the fluctuation of the transmission delay in the transmission path via the Internet. In this case, as shown in Fig. 13 (b), when the operation to output the previous video frame (indicated as frame (k-1) in the figure) to the serial bus is completed, the next video frame (Fig. In, there is a possibility that the reception of all the synchronization packets constituting (shown as frame (k)) has not been completed.

【0094】

If a large number of relay packets are missing in the transmission path, it is determined that the quality of the digital video data reconstructed in the receive buffer 223 shown in FIG. 6 is below the permissible range, and the data is accumulated in the receive buffer 223. A series of synchronization packets are discarded without being output. In such a case, since the transmission queue 247 has not been updated by the new digital video data, the rewrite control unit 250 skips the rewrite process of the output buffer 249 as a negative determination in step 342 described above.

【0095】

In this case, since the data of the immediately preceding video frame is left as it is in the output buffer 249, a series of synchronizations output in the immediately preceding video frame by the data output unit 248 as shown in FIG. 13 (b). The packet (shown as frame (k-1) in Figure 13 (b)) is output again to the IEEE1394 adapter 411 (step 345).

【0096】

In this way, when the new digital video data input to the transmission queue 247 is output to the IEEE1394 adapter 411 and new digital video data cannot be obtained in accordance with the transmission timing of the new video frame, A series of sync packets transmitted in the previous video frame can be reused.

【0097】

This makes it possible to reliably send a specified number of video frames according to the IEEE1394 standard to the serial bus regardless of the thinning process on the transmitting side and the transmission quality on the transmission path, and it is possible to reproduce high quality video and audio as a whole. Digital video data can be reconstructed.

【0098】

As described above, it consists of a series of synchronization packets transferred in the first network according to the IEEE1394 synchronization mode by the operation of the digital video transmission unit 211 provided in the relay device 210s provided as a node belonging to the first network. Digital video data is relayed via a relay network (for example, the Internet) to a relay device 210r provided as a node belonging to the second network as a series of relay packets, and digital video reception provided in the relay device 210r is performed. The operation of unit 221 makes it possible to reconstruct digital video data from this sequence of relay packets and send it to the second network.

【0099】

In this way, by relaying the relay packet between the physically independent first network and the second network via the relay network, the above-mentioned first network and the second network are virtually virtual. It is possible to realize a data communication system in which synchronization packets sent by a node belonging to the first network are transmitted to a node belonging to the second network as if it were a single network.

【0100】

The data communication system of the present invention is not limited to networks connected by a serial bus according to the IEEE1394 standard, and is applied to mutual relay between networks that transfer digital data having a known structure for each transfer unit of a predetermined format. be able to. Further, the relay network is not limited to the Internet, and a network that follows a communication protocol in the datagram format can be used.

【0101】

[Effect of the invention]

As described above, according to the invention of claim 1, structural data having a known structure transferred in the first network is relayed to the second network side via a relay network, and information on the structure is transmitted. It can be used and restored.

【0102】

As a result, for example, the structural data transmitted by the node belonging to the first network according to the synchronous communication procedure can be transmitted to the node belonging to the second network while maintaining the structure of the original structural data. It is possible to provide various services regardless of the scale and physical distance. In the present invention, the information used for restoring the structural data in the receiving side relay means is known information regarding the structure of the structural data, and it is necessary to add special information to the structural data by the transmitting side relay means. Absent.

【0103】

In particular, according to the invention of claim 2, in the receiving side relay means, the digital video data is restored for each video frame which is a component thereof, and the quality of the restored digital video data is evaluated according to the second result. Since it can be sent to the network, the quality of the digital video data of each video frame sent to the second network can be guaranteed.

【0104】

Further, according to the invention of claim 3, by setting an upper limit on the number of transfer units included in the reformed structural data, the structural data reshaped including an abnormally large number of transfer units is invalid. It is possible to reduce the influence of low transmission quality on the transmission path via the relay network on the second network.

【0105】

Further, according to the invention of claim 4, a lower limit is set for the number of transfer units included in the reformed structural data, and the number of transfer units indicated by this lower limit is allowed to be lost, whereby a sufficient number is obtained. It is possible to send video frames to the second network, and it is possible to improve the quality of digital video data as a whole, which consists of a chain of video frames.

【0106】

On the other hand, according to the invention of claim 5, when the structural data to be transmitted cannot be reformed by the transmission start timing of the video frame, the structural data transmitted in the immediately preceding video frame is re-used as the second. By sending to the network, the number of video frames to be sent to the second network can be guaranteed. On the other hand, according to the invention of claim 6, the transmitting side relay means can thin out the digital video data in video frame units and send it to the relay network in response to the thinning instruction. Therefore, the information to be sent to the relay network. The amount can be reduced and the transmission load on the relay network can be reduced.

【0107】

In particular, according to the invention of claim 7, by adjusting the transmission interval for transmitting the relay transfer unit to the relay network according to the thinning rate, the traffic in the relay network is averaged and the transmission of the relay transfer unit is performed. Since the required band can be reduced, the probability of occurrence of loss of transfer units in the transmission path can be suppressed, and transmission quality can be improved.

[Simple explanation of drawings]

[Figure 1]

It is a principle block diagram of the data communication system of claims 1 to 4.

[Figure 2]

It is a principle block diagram of the data communication system of claim 5.

[Fig. 3]

It is a principle block diagram of the data communication system of claim 6 and claim 7.

[Fig. 4]

It is a figure which shows the embodiment of the data communication system of this invention.

[Fig. 5]

It is a figure which shows the detailed structure of the digital video transmission part.

[Fig. 6]

It is a figure which shows the detailed structure of the digital video receiver.

[Fig. 7]

It is a flow chart which shows the packet formation operation and the packet transmission operation by a DV transmission part.

[Fig. 8]

It is a figure explaining the transmission interval adjustment operation.

[Fig. 9]

It is a flow chart which shows the operation which restores a digital video data.

[Fig. 10]

It is a figure explaining the restoration operation of digital video data.

[Fig. 11]

It is a figure which shows another embodiment of the digital video receiving part.

[Fig. 12]

It is a flow chart which shows the output operation of digital video data.

[Fig. 13]

It is a figure explaining the output operation of digital video data.

[Fig. 14]

It is a figure which shows the format of a digital video data.

[Fig. 15]

It is a figure which shows the data structure at the time of transferring DV data in the IEEE1394 synchronous mode.

[Fig. 16]

It is a figure which shows the system configuration used for the moving image communication experiment.

[Fig. 17]

It is a figure explaining the format at the time of transmitting DV data using UDP.

[Explanation of symbols]

110 Sender relay means 111 Forming means 112 Sending means 120 Receiving side relay means 121 Disassembly means 122 Restoration means 123 Output means 124 Detection means 125 Reshaping means 126 Collection means 127 Evaluation means 128 Result output means 131 Counting means 132 First judgment means 133 Measuring means 134 Estimating means 135 Second judgment means 136 First holding means 137 Second holding means 138 Transmission means 139 Third judgment means 140 Input means 141 Frame counting means 142 Formation control means 143 Packet holding means 144 Interval calculation means 145 Packet output means 210 Relay device 211, 412 Digital video transmitter 212 Packet forming part 213 Transmission control unit 214 transmit buffer 215 Packet reader 221,421 Digital video receiver 222 Capsule disassembly part 223 Receive buffer 224 Integrity Evaluation Department 225 Result output section 231, 241 Frame detector 232 frame counter 233 Formation control unit 234 Control information holding unit 235 Interval calculation unit 236 Transmission control unit 242 packet counter 243 Data length counter 244 Comparator 245 Integrity judgment unit 246 Data reading section 247 Send queue 248 Data output section 249 output buffer 250 Selection control unit 251 Packet discriminator 401 Digital Video Camera 402 Digital VCR 410,420 gateway 411 IEEE1394 adapter 413 internet adapter