JP4714615B2 - Transmission error correction method - Google Patents

Description

  The present invention relates to a technique for transmitting data. In particular, the present invention relates to a data transmission technique that is highly recoverable even if there is loss due to noise.

  Conventionally, in order to cope with errors in continuous data transmission such as moving images, data is divided into a plurality of blocks at the time of transmission, and each block is subjected to error correction coding processing and transmitted. When an error occurs in a single block, transmitted data such as an image can be reproduced by performing interpolation between frames using block data around the block at the time of decoding on the receiving side.

  However, when a burst-type error in which an error occurs continuously over a plurality of block data due to periodic noise occurs, reproduction may not be possible by interpolation between frames.

  Even in an environment where a bursty error occurs, there is a technique that can cope with the error correction coding process and inter-frame interpolation (for example, see Patent Document 1).

  The technique disclosed in Patent Document 1 discloses an arrangement direction of block data examples to be subjected to error correction coding processing and an arrangement direction of block data strings to be subjected to transmission processing with respect to block data sequences constituting a frame image. Are different from each other. Even if a burst error occurs in the transmitted block data string, it does not occur as a continuous error in the sorted block data string subjected to the error correction processing. Thereby, reproduction by data interpolation becomes possible.

JP-A-10-23417

  The technique disclosed in Patent Document 1 still restores error data by data interpolation. As long as restoration is performed by interpolation, restoration may not be possible depending on the degree of error. Regardless of the degree of noise that is the cause of the error, the encoding process and the conversion process of the alignment direction at the time of transmission are always performed, and the processing amount cannot be adjusted.

  The present invention has been made in view of such circumstances, and an object thereof is to realize error correction processing with high accuracy in an environment in which noise periodically occurs for a predetermined period.

  According to the present invention, data is multiplexed by duplicating, each of which is transmitted with a time difference, and the data in which an error has occurred is reproduced using the data of another multiplexed data string. Further, the time difference at the time of transmission is adjusted to an optimum time difference based on the error occurrence period of the usage environment.

For example, a data transmission device that transmits continuous data (hereinafter referred to as a data string),
A transmission error correction method in a data transmission / reception system including a data receiving device that receives the data string and performs error correction,
A data encoding step for encoding the input data string;
A data multiplexing step of duplicating the encoded data sequence to generate a plurality of data sequences;
A data transmission step of transmitting each of the multiplexed data strings with a predetermined time difference;
A data receiving step of receiving and holding the data strings transmitted with the time difference, respectively;
Data decoding that performs error correction by decoding received data using one series of the data string and, when there is data loss due to a transmission error, decoding the data using data of another series Steps,
A time difference determining step for determining a time difference in the data transmission step;
With
In the data decoding step, when there is data loss due to a transmission error in one series of the data, the time difference determining step includes:
A second data encoding step for receiving an input of a data string for determining a time difference and encoding the data string;
A second data multiplexing step of duplicating the data sequence encoded in the second data encoding step to generate a plurality of data sequences;
A second data transmission step of transmitting the data sequences multiplexed in the second data multiplexing step with a minimum time difference, and
A second data receiving step for receiving each of the data strings transmitted with a time difference in the second data transmitting step;
The received data is decoded using one series of data strings received in the second receiving step, and when there is data loss due to a transmission error, the subsequent data is decoded using other series data. If the error correction is performed and decoding is not possible using other series data, a second time difference is calculated based on a period during which decoding was not possible using other series data. A second data decoding step to transmit;
A transmission error correction method comprising: a time difference setting step of setting the second time difference calculated in the second data decoding step to the time difference in the data transmission step .

  A highly accurate error correction process can be efficiently realized in an environment where noise periodically occurs for a predetermined period.

  Embodiments to which the present invention is applied will be described below with reference to the drawings.

  First, an outline of the moving image transmission system of the present embodiment will be described.

  FIG. 1 is a schematic diagram of a bidirectional moving image transmission system in the moving image transmission system of the present embodiment.

  As shown in the figure, the bidirectional moving image transmission system of the present embodiment includes moving image transmitting / receiving devices 100a and 100b, a transmission path 300, image input devices 400a and 400b such as a digital video (DV) camera, and a microphone. Audio input devices 500a and 500b, image display devices 600a and 600b such as monitors, and audio output devices 700a and 700b such as speakers.

  The moving image transmitting / receiving apparatus 100a performs compression and error correction coding processing on the image and audio data sequentially input from the image input apparatus 400a and the audio input apparatus 500a, and then duplicates the data by duplicating the data. The data is sequentially output to the transmission line 300 with a time difference.

  The moving image transmission / reception device 100b sequentially receives the duplicated data transmitted from the moving image transmission / reception device 100a via the transmission path 300, and reconstructs the data while reconstructing the error location, that is, compensating for the loss due to noise. Correction decoding and expansion processing are performed, and the result is output to the image display device 600b and the audio output device 700b.

  The moving image transmission / reception devices 100a and 100b basically have the same configuration, but in this embodiment, the moving image transmission / reception device 100a is used as a transmission side device, and the moving image transmission / reception device 100b is used as a reception side device. Taking the configuration as an example, the operation outline, processing flow, processing sequence, etc. will be described.

  The transmission line 300 includes general DSL (Digital Subscriber Line) that superimposes a data string on a telephone line, PLC (Power Line Carrier) that superimposes a data string on a power line, and the like, and radio waves and light. In general, a wireless transmission system that modulates the above signal with a data string is applicable. In the present embodiment, the transmission line 300 is an xDSL line.

  FIG. 2 is a diagram for explaining the outline of moving image transmission / reception and data recovery in the present embodiment.

  In the present embodiment, the data lost in one data string is compensated with the other data by duplicating the data and transmitting with a time difference.

  As shown in FIG. 2A, the transmitting-side moving-image transmitting / receiving apparatus 100a transmits two identical moving-image data strings with a time difference to one xDSL line (transmission path 300). Here, it is assumed that A column and B column. At this time, a sequence number is assigned to each data string for each frame for data order management. The same sequence number is assigned to the same frame.

  When the moving image transmitting / receiving apparatus 100b on the receiving side decodes one of the received data strings and detects the loss of a frame, the moving picture transmitting / receiving apparatus 100b performs decoding from the corresponding frame of the other data string. In this way, every time the disappearance is detected, the data string to be decoded is alternately switched to perform decoding.

  Taking the case shown in FIG. 2B as an example, the data recovery method will be specifically described. As shown in FIG. 2B, the loss due to noise occurs in a frame that exists in the transmission line 300 that is an xDSL line at the same time in both the A column and the B column.

  The moving image transmitting / receiving apparatus 100b on the receiving side performs decoding while confirming the sequence number of the A column that is the previously received data column. When it is detected that the frame with the sequence number 5 has disappeared due to noise, the decoding target is switched to the data in the B column, and decoding is started from the frame with the sequence number 5 in the B column. Next, when the loss of data in the 9th frame in the B column is detected, the decoding target is switched to the A column again, and decoding is started from the frame in which the sequence number in the A column is the 9th. Repeat this.

  In the present embodiment, optimization is performed so that the time difference until the B sequence is transmitted is a time difference corresponding to periodic burst noise generated in an environment in which the moving image transmitting / receiving apparatuses 100a and 100b are used. To do. Hereinafter, this optimization process is called training. In the moving image transmitting / receiving apparatuses 100a and 100b, a state in which training processing is performed is referred to as a training mode, and a processing state in which normal moving image transmission / reception and reproduction processing is performed is referred to as a normal mode. In the present embodiment, which mode the processing state is in is determined by setting a flag indicating that the training mode is in the training mode.

  Next, an outline of training according to the present embodiment will be described.

  FIG. 3 is a diagram for explaining an outline of training according to the present embodiment. The training mode is entered upon receiving a training start request that is a training instruction. When shifting to the training mode, the moving image transmitting / receiving apparatus 100a on the transmission side sets the minimum difference of one frame as the time difference for transmitting the sequence B as shown in FIG.

  The moving image transmitting / receiving apparatus 100b on the receiving side tries to reconstruct the data by using the received data in the A and B columns by the above method. Here, in the example shown in FIG. 3A in which the B column is transmitted with a time difference of one frame, after the disappearance of the frame 5 is detected in the A column, the reconstruction is performed by switching to the B column. However, when switching to the B column, the frame number 7 can be extracted next. Therefore, frames with frame numbers 5 and 6 cannot be reproduced. The moving image transmitting / receiving apparatus 100b on the receiving side determines that two frames cannot be reproduced.

  Next, the moving image transmission / reception device 100b on the reception side generates training data instructing transmission by adding a time difference of two frames to the current time difference, and transmits the training data to the moving image transmission / reception device 100a on the transmission side.

  When receiving the training data, which is an instruction to add a time difference of 2 frames, the moving image transmitting / receiving device 100a on the transmission side, as shown in FIG. Control to transmit as 3 frames of time difference. As a result, the transmitting-side moving image transmitting / receiving apparatus 100a starts transmission with a time difference of 3 frames, and the receiving-side moving image transmitting / receiving apparatus 100b receives transmission with a time difference of 3 frames. When the moving image transmitting / receiving apparatus 100b on the receiving side detects the disappearance of the frame 5 in the A column and switches to the B column, it can extract the frame 5 in the B column. As a result, there are no missing frames, and all frames can be reproduced.

  Next, the moving image transmission / reception apparatuses 100a and 100b of the present embodiment that realize the above data transmission and training will be described below.

  FIG. 4 is a hardware block diagram of the bidirectional moving image transmission system according to the present embodiment.

  As shown in the figure, the moving image transmitting / receiving apparatuses 100a and 100b according to the present embodiment process images input from the image / audio compression / decompression units 105a and 105b and the image input apparatuses 400a and 400b, respectively. The image input units 101a and 101b output to the compression / decompression units 105a and 105b, and the audio input unit 102a that processes the audio data input from the audio input devices 500a and 500b and outputs the processed data to the image audio compression / decompression units 105a and 105b , 102b, the image output units 103a and 103b for outputting the images processed by the image / sound compression / decompression units 105a and 105b to the image display devices 600a and 600b, and the sound processed by the image / sound compression / decompression units 105a and 105b. Are output to the audio output devices 700a and 700b, The main control units 110a and 110b that control the entire image transmission / reception devices 100a and 100b, the memories 111a and 111b, the video / audio compression / decompression units 105a and 105b, and the transmission path 300, an xDSL interface ( I / F) sections 112a and 112b.

  The video / audio compression / decompression units 105a and 105b perform compression, error correction coding processing, data string duplication processing, and the like on the input video and audio data, and transfer lines via the xDSL I / F units 112a and 112b. The data received from the transmission line 300 via the xDSL I / F units 112a and 112b are subjected to data reconstruction, error correction decoding and decompression processing, and output.

  FIG. 5 shows a functional block diagram of the moving image transmitting / receiving apparatuses 100a and 100b of the present embodiment.

  Since the moving image transmitting / receiving apparatuses 100a and 100b have the same functions, the moving image transmitting / receiving apparatus 100a will be described here.

  The moving image transmission / reception apparatus 100a of this embodiment includes an image data duplex control unit 11, an image data reconstruction control unit 12, a transmission / reception data configuration control unit 13, an image training setting control unit 14, an image training detection control unit 15, and an image compression control. Unit 16, image expansion control unit 17, analog / digital (A / D) conversion units 18, 20B, digital / analog (D / A) conversion units 19, 20A, image input control unit 21, image output control unit 22, xDSL input / output And a control unit 23.

  The image input control unit 21 receives image data sequentially input from the image input device 400 a and transmits it to the A / D conversion unit 18.

  The A / D conversion unit 18 converts image data that is an analog signal into image data that is a digital signal, and transmits the image data to the image compression control unit 16.

  The image compression control unit 16 applies a compression process such as MPEG to the digitized image data.

  The image data duplication control unit 11 assigns a sequence number to the data from the image compression control unit 16 and transmits the data in duplicate with a time difference.

  The transmission / reception data configuration control unit 13 converts the image data to be transmitted into an IP packet and transmits the IP packet to the xDSL input / output control unit 23. Further, image data is extracted from the IP packet received from the xDSL input / output control unit 23 and transmitted to the image data reconstruction control unit 12.

  The xDSL input / output control unit 23 performs conversion between the IP packet and the xDSL packet.

  The D / A conversion unit 20A and the A / D conversion unit 20B convert digital data transmitted from the xDSL input / output control unit 23 into analog data, respectively, through an interface between the transmission line 300 and the xDSL input / output control unit 23. The analog data received from the transmission path 300 is converted to digital data and transmitted to the xDSL input / output control unit 23.

  The image data reconstruction control unit 12 receives the image data transmitted in a duplex manner via the transmission path 300, the A / D conversion unit 20B, the xDSL input / output control unit 23, and the transmission / reception data configuration control unit 13, and receives the received image data. The sequence numbers of the processed image data are checked, rearranged in the order of the sequence numbers, and transmitted to the image expansion control unit 17 and the image training detection control unit 15.

  The image data expansion control unit 17 restores the received image data and transmits it to the D / A conversion unit 19.

  The D / A conversion unit 19 converts image data that is a digital signal into image data that is an analog signal, and transmits the image data to the image output control unit 22.

  The image output control unit 22 sequentially outputs images to the image display device 600a.

  The image training detection control unit 15 checks the sequence numbers of the image data rearranged in the sequence order received from the image data reconstruction control unit 12, and detects the number of data loss due to noise. When there is data loss, the number of data loss and the number of lost data at each occurrence are stored.

  In the training mode, when the occurrence of data loss is greater than or equal to a predetermined number of times, the transmission / reception data configuration control unit 13 uses the time difference obtained from the maximum value of the stored number of lost data as training data, The data is transmitted to the moving image transmitting / receiving apparatuses 100a and 100b on the transmission side via the xDSL input / output control unit 23, the D / A conversion unit 20A, and the transmission path 300. At this time, the image training detection control unit 15 also notifies the image data reconstruction control unit 12 that reconstructs the received image data of the time difference instructed to be newly set.

  On the other hand, in the normal mode, when the occurrence of data loss is greater than or equal to a predetermined number of times, a training start request is transmitted to the moving image transmitting / receiving apparatuses 100a and 100b on the transmission side.

  Further, the image training detection control unit 15 deletes the recorded number of times and the number of lost data when the training data or the training start request is transmitted.

  The image training setting control unit 14 is sent from the moving image transmission / reception devices 100a and 100b on the reception side via the transmission path 300, the A / D conversion unit 20B, the xDSL input / output control unit 23, and the transmission / reception data configuration control unit 13. The training data or the training start request is received, the time difference of duplex transmission is adjusted, and the adjustment result is transmitted to the image data duplex control unit 11.

  Next, the frame format of a packet transmitted from the moving image transmitting / receiving apparatus 100a will be described.

  FIG. 6 is a diagram for explaining a frame format.

  As shown in this figure, the frame 330 of this embodiment is an ether frame, and the transport layer selects the UDP protocol. The frame of this embodiment includes an IP header portion 330a, a UDP header portion 330b, an MPEG data portion 330c, a sequence number portion 330d (4 bytes) for storing the sequence number of the frame, and whether the frame is of the A column or B A data string identifying unit 330e (4 bytes) that stores information for identifying whether it is a string.

  FIG. 7 is a diagram for explaining how frames are transmitted to the transmission path 300.

  In FIG. 2 and FIG. 3, the frame 330 is described as an image transmitted to the transmission line 300 in parallel. However, as shown in FIG. 7, the frame 330 is actually transmitted from the moving image transmitting / receiving apparatuses 100a and 100b on the transmission side. A frame and B frame are transmitted alternately. Then, in the moving image transmitting / receiving apparatuses 100a and 100b on the receiving side, the data string identification unit 330e is referenced to identify which data is transmitted as the A column or the B column.

  Next, the processing operation of the moving image transmitting / receiving apparatuses 100a and 100b according to the present embodiment will be described.

  FIG. 8 shows a processing flow of the moving image transmitting / receiving apparatus 100a of the present embodiment when image data is input from the image input apparatus 400a.

  When an image is input from the image input control unit 21 (step 401), the A / D conversion unit 18 performs A / D conversion on the image (step 402). Then, the image compression control unit 16 compresses the digitized image data (step 403) and transmits it to the image data duplexing control unit 11.

  The image data duplexing control unit 11 frames the compressed image data, and assigns a sequence number by storing the sequence number in the sequence number unit 330d for each frame (step 404). Then, the data is duplicated, and information indicating the A column and the B column is stored in the data column identification unit 330e, so that the A column and the B column are obtained, respectively, and are duplicated (step 405). Thereafter, the duplexed image data is transmitted to the transmission / reception data configuration control unit 13.

  The transmission / reception data configuration control unit 13 converts the received image data into IP packets (step 406), the xDSL input / output control unit 23 converts the IP packets into xDSL packets (step 407), and the D / A conversion unit 20B The digital signal after packet conversion is converted into an analog signal and transmitted to the transmission line 300 (step 408).

  FIG. 9 is a processing flow of the moving image transmitting / receiving apparatus 100a when training data is received.

  When the training data is received, the A / D conversion unit 20B converts the training data into a digital signal (step 501), and the xDSL input / output control unit 23 converts the converted data from an xDSL packet to an IP packet (step 502). ). When the transmission / reception data configuration control unit 13 receives an IP packet of training data from the xDSL input / output control unit 23, the transmission / reception data configuration control unit 13 transfers the training data to the image training setting control unit 14 (step 503). The image training setting control unit 14 changes the setting of the time difference between the data strings in accordance with the instruction in the training data (step 504), and notifies the image data duplexing control unit 11 of the change result.

  Note that, when a training start request is received instead of training data, the same processing as described above is performed. However, in this case, in step 504, the time difference between the data strings is set to the minimum (time for one frame in this embodiment).

  FIG. 10 is a processing flow of the moving image transmitting / receiving apparatus 100b when image data is received from the transmission path 300.

  Upon receiving the image data, the A / D conversion unit 20B converts the image data into a digital signal (step 601), and the xDSL input / output control unit 23 converts the digital signal after the A / D conversion into an IP packet (step 602). Upon receiving the IP packetized data, the transmission / reception data configuration control unit 13 transmits the data to the image data reconstruction control unit 12 (step 603).

  The image data reconstruction control unit 12 has the data in the A column or the B column according to the information stored in the data column identification unit 330e and the sequence number unit 330d of each frame. And attempts to arrange the frames of the column currently employed as the processing target in the order of the sequence number (step 604).

  The absence of the sequence number, that is, the loss of data is confirmed (step 605). If there is no loss of data (step 606), the frame of the image data is reconstructed in the order of the sequence number (step 608). On the other hand, if there is a missing sequence number, that is, if there is data loss (step 606), the processing target is switched to a column that is not the current processing target (step 607), and frame reconstruction is performed in the order of the sequence number. Perform (step 608).

  At this time, when frame reconstruction is succeeded in sequence numbers in the order of sequence numbers (step 609), the reconstructed data is output to the image expansion control unit 17.

  Then, the image expansion control unit 17 expands the received image data (step 611), the D / A conversion processing unit 19 converts the expanded data into an analog signal (step 612), and the image output control unit 22 And output to the image display device 600a (step 613).

  On the other hand, if there is a frame with a sequence number that cannot be acquired by switching to either column in the above step 609, that is, if the frame cannot be reconstructed in sequence number order, the image data reconstruction control unit 12 This is notified to the image training detection control unit 15 (impossible to reconstruct).

  Upon receiving the notification, the image training detection control unit 15 investigates the frame that has been processed by the image data reconstruction control unit 12 and detects the number of lost frames. Then, the counter for storing the number of occurrences of disappearance (number counter) is incremented by 1, and the detected number of disappearances is recorded (step 631). Next, the image training detection control unit 15 determines whether or not the number recorded in the number counter is greater than or equal to a predetermined number (step 632).

  If the number is smaller than the predetermined number, the image training detection control unit 15 advances the process to step 611.

  On the other hand, if the number is greater than or equal to the predetermined number in step 632, the image training detection control unit 15 determines whether the processing is performed in the normal mode or the training mode (step 633).

  If it is determined that the processing in the training mode is performed, the image training detection control unit 15 calculates the minimum necessary time difference between the two data strings based on the maximum number of recorded erasures. (Step 621). Then, using the calculation result, training data instructing to change the time difference is generated and transmitted to the transmission / reception data configuration control unit 13 (step 622).

  The received transmission / reception data configuration control unit 13 converts the received training data into IP packets (step 623). The xDSL input / output control unit 23 converts the IP packetized training data into an xDSL format (step 624). The D / A converter 20A converts the digital signal after the packet conversion into an analog signal and transmits the analog signal to the transmission line 300 (step 625).

  On the other hand, when it is determined in step 633 that the processing in the normal mode is being performed, the image training detection control unit 15 generates a training start request (step 634) and proceeds to step 623.

  Next, a procedure for switching between the normal mode and the training mode when moving images are transmitted / received between the moving image transmitting / receiving apparatuses 100a and 100b will be described. In the present embodiment, the instruction to shift to the training mode is given when a non-reproducible state occurs more than a predetermined number of times when starting the transmitting side device of the moving image transmitting / receiving device and during moving image transmission. (When an error occurs during transmission).

  FIG. 11 is a sequence in a case where the moving image transmitting / receiving apparatus 100a, which is a transmission-side apparatus, shifts to the training mode upon activation. Here, the moving image transmitting / receiving apparatus 100b, which is the receiving apparatus, operates in the normal mode.

  A training start request packet is transmitted from the moving image transmitting / receiving apparatus 100a to the moving image transmitting / receiving apparatus 100b (step 701). In response to this, the moving image transmitting / receiving apparatus 100b transmits a training start acceptance packet and responds (step 702). In addition, the time difference data used when composing the received moving image is minimized (hereinafter referred to as time difference A), and the mode is shifted to the training mode. In this embodiment, the minimum time difference is a time difference corresponding to one image frame to be transmitted.

  When receiving the response, the moving image transmitting / receiving apparatus 100a sets the time difference for transmitting the A column and the B column to the time difference A, and transmits the image data to the moving image transmitting / receiving apparatus 100b (step 703).

  When the moving image transmitting / receiving apparatus 100b detects that the state in which the reconstruction is impossible has occurred for a predetermined number of times or more, the moving image transmission / reception device 100b is based on the case where the reconstruction cannot be performed for the longest time among the states in which the reconstruction is impossible. Then, an optimum time difference (hereinafter referred to as time difference B) which is a time difference to be newly set is determined (step 704). Then, the time difference B is notified to the moving image transmitting / receiving apparatus 100a as training data (step 705).

  When receiving the training data, the moving image transmitting / receiving apparatus 100a stops transmitting the moving image (step 706), and transmits a training result confirmation packet indicating that the training data has been received to the moving image transmitting / receiving apparatus 100b (step 707). ).

  Receiving the training result confirmation packet, the moving image transmission / reception device 100b transmits a training completion notification packet to the moving image transmission / reception device 100a as a reception response to the training result confirmation packet, and sets the time difference at the time of reception to the time difference B. (Step 708) and the mode is changed to the normal mode.

  The moving image transmitting / receiving apparatus 100a that has received the training completion notification packet sets the time difference B as the time difference at the time of duplex transmission (step 709), shifts to the normal mode, and moves to the A and B columns with the time difference B. Is started (step 710).

  Next, a sequence for shifting to the training mode when an error occurs during transmission will be described with reference to FIG.

  Both the moving image transmission / reception devices 100a and 100b operate in the normal mode, and the moving image is transmitted from the moving image transmission / reception device 100a to the moving image transmission / reception device 100b with a time difference B (step 801). The time difference B is a time difference set during the previous training.

  When the moving image transmission / reception device 100b detects that the reconfigurable state (reception error) exceeds the predetermined number of times during processing in the normal mode, the moving image transmission / reception device 100b sends a training start request packet to the moving image transmission / reception device 100a. Transmit (step 802).

  The moving image transmission / reception device 100a that has received the training start request packet stops data transmission (step 803). Then, the time difference between the moving image transmissions in the A column and the B column is set to the minimum (here, the time difference of one frame; hereinafter referred to as the time difference A), the mode is shifted to the training mode, and the training start reception is a moving image. It transmits to the image transmission / reception apparatus 100b (step 804).

  The moving image transmitting / receiving apparatus 100b sets the time difference used for reconstruction when receiving a moving image to the time difference A (step 805), shifts to the training mode, and waits for the reception of the training image.

  The moving image transmitting / receiving apparatus 100a starts transmitting moving image data (step 806).

  When the moving image transmitting / receiving apparatus 100b detects that the state in which the reconstruction is impossible has occurred for a predetermined number of times or more, the moving image transmission / reception device 100b is based on the case where the reconstruction cannot be performed for the longest time among the states in which the reconstruction is impossible. Then, an optimum time difference (hereinafter referred to as time difference C), which is a time difference to be newly set, is determined (step 807). Then, the time difference C is notified to the moving image transmitting / receiving apparatus 100a as training data (step 808).

  When receiving the training data, the moving image transmission / reception device 100a stops transmitting the moving image (step 809), and transmits a training result confirmation packet indicating that the training data has been received to the moving image transmission / reception device 100b (step 810). ).

  Receiving the training result confirmation packet, the moving image transmission / reception device 100b transmits a training completion notification packet to the moving image transmission / reception device 100a as a reception response to the training result confirmation packet, and sets the time difference upon reception to the time difference C. (Step 811) and the mode is changed to the normal mode.

  The moving image transmitting / receiving apparatus 100a that has received the training completion notification packet sets the time difference C as the time difference at the time of duplex transmission (step 812), shifts to the normal mode, and moves to the A and B columns with the time difference C. Starts to be transmitted (step 813).

  In the above, when the mode is shifted to the training mode when an error occurs during transmission, the training process is started after setting to the time difference A which is the minimum time difference. However, for example, the training process may be performed while the time difference is set to the time difference B. In this case, the moving image transmitting / receiving apparatus 100b calculates the number of frames that cannot be reconstructed in the time difference B, and calculates the optimal time difference.

  As described above, according to the present embodiment, it is possible to improve the accuracy of error correction in data transmission in an environment where noise is periodically generated for a predetermined period. In addition, it is possible to realize an optimum processing amount according to the environment where the moving image transmitting / receiving apparatus is installed.

  For example, when the state inside the elevator is continuously photographed and transmitted to a monitor in a management room as a moving image, noise may be generated periodically for a predetermined period as the elevator moves up and down. And since the installation environment of each elevator differs, the generation | occurrence | production period and period of noise differ with each elevator. In such a use environment, the system using the moving image transmitting / receiving apparatus of the present embodiment can achieve error correction with an optimum processing amount.

  In the above embodiment, the moving image transmission / reception devices 100a and 100b, which are bidirectional moving image transmission / reception devices having both the moving image transmission function and the reception function, are used as the transmission side device and the reception side device, respectively. Although the case has been described as an example, the above-described embodiment can be realized by an apparatus having only a transmission function and only a reception function.

  FIG. 13 shows a schematic diagram of a one-way moving image transmission system that transmits moving images from a moving image transmitting device 100c having only a moving image transmitting function to a moving image receiving device 100d having only a moving image receiving function.

  As shown in the figure, the one-way moving image transmission system of the present embodiment includes a moving image transmitting device 100c, a moving image receiving device 100d, a transmission path 300, and an image input device 400c such as a digital video (DV) camera. And an audio input device 500c such as a microphone, an image display device 600d such as a monitor, and an audio output device 700d such as a speaker.

  The moving image transmitting apparatus 100c is an apparatus having only the moving image transmitting function of the above-described moving image transmitting / receiving apparatuses 100a and 100b, and compresses and compresses image and audio data sequentially input from the image input apparatus 400c and the audio input apparatus 500c. After performing error correction coding processing, data string duplication processing, and the like, the data is sequentially transmitted to the transmission line 300.

  The moving image receiving apparatus 100d is an apparatus having only the moving image receiving function of the moving image transmitting / receiving apparatuses 100a and 100b described above, and sequentially receives the data transmitted from the moving image transmitting apparatus 100c via the transmission path 300, Data reconstruction, error correction decoding, and decompression processing are performed and output to the image display device 600d and the audio output device 700d.

  FIG. 14 is a hardware block diagram of the one-way moving image transmission system.

  As shown in the figure, the moving image transmitting apparatus 100c includes an image / audio compression unit 105c, an image input unit 101c that processes an image input from the image input device 400c and outputs the processed image to the image / audio compression unit 105c, and an audio input device. An audio input unit 102c that processes audio data input from 500c and outputs the processed audio data to the image / audio compression unit 105c, a main control unit 110c that controls the entire moving image transmitting apparatus 100c, a memory 111c, and an image / audio compression unit 105c. And an xDSL interface (I / F) unit 112c, which is an interface between the transmission line 300 and the transmission line 300.

  The moving image receiving apparatus 100d includes an image / audio expansion unit 105d, an image output unit 103d that outputs an image processed by the image / audio expansion unit 105d to the image display device 600d, and an audio processed by the image / audio expansion unit 105d. Is an interface between the audio output unit 104d that outputs to the audio output device 700d, the main control unit 110d that controls the entire moving image receiving device 100d, the memory 111d, the image / audio expansion unit 105d, and the transmission path 300. , An xDSL interface (I / F) unit 112d.

  In the moving image transmitting apparatus 100c and the moving image receiving apparatus 100d having the above-described configuration, the same processing as that described with respect to the above-described moving image transmitting / receiving apparatuses 100a and 100b is performed, and appropriately according to the noise generation cycle and generation period. Perform training and optimally adjust the transmission time difference of the moving images to be transmitted in duplicate.

  In the above embodiment, the case where moving images and audio are transmitted and received has been described as an example. However, the data to be transmitted and received is not limited thereto. It may be a still image, only audio, or data other than an image.

  In the above embodiment, the case where data is duplexed and transmitted / received has been described as an example. However, the processing of data to be transmitted / received is not limited to duplexing, and multiplexing is possible.

  In the above embodiment, the xDSL line has been described as an example, but the line is not limited to this. For example, a wired or wireless LAN line is also possible.

FIG. 1 is a schematic diagram of a bidirectional moving image transmission system according to the present embodiment. FIG. 2 is a diagram for explaining the outline of moving image transmission / reception and data recovery in the present embodiment. FIG. 3 is a diagram for explaining an outline of training according to the present embodiment. FIG. 4 is a hardware block diagram of the bidirectional moving image transmission system according to the present embodiment. FIG. 5 is a functional block diagram of the moving image transmitting / receiving apparatus of the present embodiment. FIG. 6 is a diagram for explaining the frame format of the present embodiment. FIG. 7 is a diagram for explaining a state of transmission of a frame to a transmission path according to the present embodiment. FIG. 8 is a processing flow at the time of image data input of the moving image transmitting / receiving apparatus of this embodiment. FIG. 9 is a processing flow at the time of training data reception of the moving image transmission / reception apparatus of this embodiment. FIG. 10 is a processing flow at the time of image data reception from the transmission path of the moving image transmission / reception apparatus of this embodiment. FIG. 11 is a sequence of training mode transition at the time of activation of the present embodiment. FIG. 12 is a training mode transition sequence when an error occurs during transmission according to this embodiment. FIG. 13 is a schematic diagram of the one-way moving image transmission system of the present embodiment. FIG. 14 is a hardware block diagram of the one-way moving image transmission system of this embodiment.

Explanation of symbols

100a: moving image transmission / reception device, 100b: moving image transmission / reception device, 300: transmission path, 400a: image input device, 400b: image input device, 500a: voice input device, 500b: voice input device, 600a: image display device, 600b : Image display device, 700a: Audio output device, 700b: Audio output device, 105a: Image audio compression / decompression unit, 105b: Image audio compression / decompression unit, 101a: Image input unit, 101b: Image input unit, 102a: Audio Input unit, 102b: audio input unit, 103a: image output unit, 103b: image output unit, 104a: audio output unit, 104b: audio output unit, 110a: main control unit, 110b: main control unit, 111a: memory, 111b : Memory, 112a: xDSL interface (I / F) section, 112b: xDSL interface (I / F) section, 11: Image data duplex control unit, 12: Image data reconstruction control unit, 13: Transmission / reception data configuration control unit, 14: Image training setting control unit, 15: Image training detection control unit, 16: Image compression control unit, 17: Image expansion control unit, 18: analog-digital (A / D) conversion unit, 19: digital-analog (D / A) conversion unit, 20A: digital-analog (D / A) conversion unit, 20B: analog-digital (A / D) Conversion unit, 21: Image input control unit, 22: Image output control unit, 23: xDSL input / output control unit

Claims (1)

A data transmission device that transmits continuous data (hereinafter referred to as a data string);
A transmission error correction method in a data transmission / reception system including a data receiving device that receives the data string and performs error correction,
A data encoding step for encoding the input data string;
A data multiplexing step of duplicating the encoded data sequence to generate a plurality of data sequences;
A data transmission step of transmitting each of the multiplexed data strings with a predetermined time difference;
A data receiving step of receiving and holding the data strings transmitted with the time difference, respectively;
Data decoding that performs error correction by decoding received data using one series of the data string and, when there is data loss due to a transmission error, decoding the data using data of another series Steps,
A time difference determining step for determining a time difference in the data transmission step,
In the data decoding step, when there is data loss due to a transmission error in one series of the data, the time difference determining step includes:
A second data encoding step for receiving an input of a data string for determining a time difference and encoding the data string;
A second data multiplexing step of duplicating the data sequence encoded in the second data encoding step to generate a plurality of data sequences;
A second data transmission step of transmitting the data sequences multiplexed in the second data multiplexing step with a minimum time difference, and
A second data receiving step for receiving each of the data strings transmitted with a time difference in the second data transmitting step;
The received data is decoded using one series of data strings received in the second receiving step, and when there is data loss due to a transmission error, the subsequent data is decoded using other series data. It performs error correction by, if not also decoded using data of another sequence, the second to calculate a time difference based on the period of time could not be decoded even by using the data of the other series, the transmission device A second data decoding step to transmit;
A transmission error correction method comprising: a time difference setting step of setting the second time difference calculated in the second data decoding step to the time difference in the data transmission step.

Images