JP5978710B2 - Digital broadcast receiver and digital broadcast receiving method - Google Patents

Description

  The present invention relates to a stream repair device that detects an error in a transport stream of a digital broadcast and repairs the transport stream.

  In a digital broadcast receiver, when a radio wave reception state is insufficient, a received TS packet may be missing or erroneous. If the TS packet is missing or erroneous, the decoded video becomes incomplete and normal reproduction cannot be performed. Therefore, a method of selecting and reproducing one having a smaller error from the digital broadcast transport streams received by a plurality of antennas is disclosed in order to repair the error of the received digital broadcast transport stream. In such a digital broadcast receiving apparatus, a freeze process for stopping reproduction is performed when normal reproduction is impossible, and video reproduction is resumed when a sufficient reception state is secured.

  In a digital broadcast receiver mounted on a mobile body, the reception state of radio waves changes frequently, so that missing or erroneous TS packets frequently occur, and the time for freeze processing as described above, that is, reproduction becomes impossible. More time. Therefore, in order to reduce the time during which such reproduction becomes impossible, an error in the transport stream of the received digital broadcast has been repaired.

  For example, a digital broadcast receiver compatible with the Japanese digital broadcasting system usually reproduces a 12-segment MPEG2 format transport stream included in the digital broadcast, and there is an error in the MPEG2 format transport stream. Discloses a method of reproducing by complementing an error area with one-segment MPEG4 format transport stream data transmitted simultaneously (see, for example, Patent Document 1).

Japanese Patent Laying-Open No. 2005-260606 (page 9-10, FIG. 6)

  However, in the conventional digital broadcast receiver, an error in the transport stream is detected based on time information of several bytes. Therefore, the error can be detected only when an error occurs in the time information. Therefore, when the time information is correct and other data has an error, there is a problem that the transport stream cannot be repaired.

  In the conventional digital broadcast receiver, two streams (full segment and one segment) having the same content are decoded and errors occurring in the full segment stream are complemented by the one segment stream. There is also a problem that only two ISDB digital broadcasts are available.

  The present invention has been made to solve the above-described problems. Even when an error occurs in data other than time information, or when there are not two or more streams of the same content, the transport stream error is repaired. The purpose is to reduce the time during which playback is impossible.

A digital broadcast receiver according to the present invention receives a digital broadcast, stores a data related to a data stream after decoding by an error correction inner code, and performs error correction on the decoded data stream. A decoding unit that performs error correction decoding using an outer code, an estimation unit that estimates the type of decoded data output from the decoding unit, and a detection unit that detects an error in the decoded data based on an estimation result of the estimation unit And a correction unit that corrects an error in the decoded data based on a detection result of the detection unit, and a second storage unit that stores corrected data corrected by the correction unit. the relative first the corrected data stored in the second storage unit based on the accumulated error correcting outer code contained in the data stream after the decoding in the storage unit, also performs error correction again It is.

The present invention, the received data corrected by the auxiliary Tadashibu, again error corrected using the error correction outer code, since the re-corrected after complement Tadashibu thereof, repair performance of reception data finally obtained it is possible to improve.

1 is a block diagram of a stream repair device showing a first embodiment of the present invention. FIG. It is a flowchart which shows operation | movement of the classification estimation part in Embodiment 1 of this invention. It is a figure which shows the syntax of TVCT in Embodiment 1 of this invention. It is a flowchart which shows operation | movement of the error detection part in Embodiment 1 of this invention. It is an example of TS packet which the error correction part in Embodiment 1 of this invention handles. It is a block diagram of the stream repair apparatus which shows Embodiment 2 of this invention.

Embodiment 1 FIG.
FIG. 1 is a block diagram of a stream repair apparatus according to Embodiment 1 for carrying out the present invention. As shown in FIG. 1, the stream restoration device 100 according to the present embodiment includes a demodulation unit 11, a separation unit 5, and an error correction unit 12. The demodulation unit 11 includes a synchronous reproduction unit 2 that receives a broadcast wave from the antenna 1, a waveform equalization unit 3, and an error correction unit 4. The error correction unit 12 includes a type estimation unit 6, an error detection unit 7, a stored data determination unit 8, a correction memory 9, and a correction unit 10. Further, the error correction unit 4 includes a trellis decoding unit 41, an RS decoding unit 42, and a correction memory 43.

  Next, each configuration of the stream restoration device 100 will be described. In the following description, the received digital broadcast wave will be described as an ATSC (Advanced Television Systems Committee) terrestrial wave adopted in North America.

  The antenna 1 receives an ATSC digital broadcast and sends a broadcast wave of the received digital broadcast to the demodulator 11. In the present embodiment, the demodulator 11 will be described as an 8-VSB demodulator in order to explain using an ATSC broadcast wave. The demodulator 11 demodulates the input broadcast wave, converts it into a transport stream, and outputs it to the separator 5. The separation unit 5 separates the transport stream into 188-byte TS packets and outputs them to the error correction unit 12. The error correction unit 12 detects an error from the TS packet input from the separation unit 5, corrects the error, and outputs the error. Here, the TS packet error is a bit inversion occurring in the transmission path. For example, if a bit corresponding to PID (Packet ID) is inverted, a TS packet that transmits video data cannot be recognized as video data, so a part of the video is lost or a TS packet that is a PAT is recognized as PAT. It is a problem that video and audio cannot be played because it is not possible.

  The demodulation unit 11 includes a synchronous reproduction unit 2, a waveform equalization unit 3, and an error correction unit 4. The synchronous reproduction unit 2 performs synchronization for generating a symbol from the input broadcast wave, and sends the symbol to the waveform equalization unit 3. The waveform equalization unit 3 corrects the effects of attenuation, rotation, and delay of the broadcast wave received by the symbol generated by the synchronous reproduction unit on the transmission path, and then returns the waveform from the symbol to a waveform close to that at the time of transmission. A portion corresponding to the data is extracted and a signal converted into a bit string is output to the error correction unit 4. In the case of the ATSC system, this bit string is a trellis code. The broadcast wave is encoded and transmitted, and the bit string output from the waveform equalization unit 3 is also encoded.

  The error correction unit 4 decodes the encoded bit string. Encoding is performed twice. First, an error correction outer code is attached to the bit string, and then the entire bit string with the error correction outer code is converted into an error correction inner code. Decoding of the error correction unit 4 is performed in the reverse order to encoding. The error correction unit 4 first decodes the error correction inner code to form a bit string with an error correction outer code, and then decodes the bit string in bytes by the error correction outer code (error correction) as a transport stream. Output to the separation unit 5. In addition, the error correction unit 4 outputs the corrected portion subjected to the error correction to the error repair unit 12.

  The error correction unit 4 will be described in more detail. The error correction unit 4 includes a trellis decoding unit 41, an RS decoding unit 42, and a correction memory 43. The ATSC broadcast wave is encoded with two types of an inner code trellis code and an outer code RS code. Trellis decoding unit 41 performs trellis decoding, and RS decoding unit 42 performs RS decoding.

  The trellis decoding unit 41 trellis-decodes the bit string of the trellis code to obtain a new bit string, and outputs the new bit string to the RS decoding unit 42 and the correction memory 43. The new bit string after trellis decoding is 1656 bits (207 bytes), of which 1596 bits (187 bytes) are transport stream data, and the remaining 160 bits (20 bytes) are RS codes. The RS code is error information of transport stream data.

  The RS decoding unit 42 performs error correction on the transport stream data in units of bytes based on the RS code, and outputs the corrected transport stream data to the separation unit 5. Further, the RS decoding unit 42 outputs the corrected part subjected to the error correction to the error repairing unit 12.

  The correction memory 43 is divided into a first storage unit and a second storage unit. The decoded bit string input from the trellis decoding unit 41 is stored in the first storage unit, and the repaired TS packet input from the error recovery unit 10 described later is stored in the second storage unit. The correction memory 43 outputs both the decoded bit string and the repaired TS packet to the RS decoding unit 42. When receiving an input from the correction memory 43, the RS decoding unit 42 uses the RS code of the decoded bit string input from the correction memory 43 to convert the repaired TS packet input from the correction memory 43. Correct the error.

  Here, the error correction unit 4 includes a trellis decoding unit 41 and an RS decoding unit 42 because the ATSC digital broadcast signal uses a trellis code and an RS code. . For digital broadcasts employing other systems, a decryption unit is provided for deciphering the encryption (code) employed in each system. For example, in the case of ISDB (Integrated Services Digital Broadcasting) system, which is a broadcasting system for Japanese digital broadcasting, the trellis decoding unit is a convolutional decoding unit.

  The error correction unit 12 includes a type estimation unit 6, an error detection unit 7, a stored data determination unit 8, a correction memory 9, and a correction unit 10. The type estimation unit 6 estimates the type of the TS packet input from the separation unit 5 and outputs the TS packet and the type of the TS packet to the error detection unit 7. The type of TS packet will be described later.

  The error detection unit 7 detects an error of the TS packet from the correction portion input from the RS decoding unit 42, the TS packet input from the type estimation unit 6, and the type of the TS packet, and the TS packet and the error Information is output to the correction unit 10. In addition, the error detection unit 7 outputs the TS packet input from the type estimation unit 6 to the storage data determination unit 8 and receives a determination result described later from the storage data determination unit 8. The TS packet output from the error detection unit 7 to the stored data determination unit 8 specifies, for example, each PID of a video TS packet, an audio TS packet, and a PCR (Program Clock Reference: time information) TS packet. TS packets (PAT and PMT, TVCT, etc.). Further, the error detection unit 7 extracts storage data from the TS packet input from the type estimation unit 6 based on the determination result received from the storage data determination unit 8, and outputs it to the correction memory 9. Here, the storage data is the 188-byte TS packet itself, or data other than the adaptation field extracted from the TS packet.

  The stored data determination unit 8 analyzes the TS packet input from the error detection unit 7, determines data to be stored in the correction memory 9, and outputs the determination result to the error detection unit 7.

  The correction memory 9 accumulates data to be stored input from the error detection unit 7 and outputs this data to the correction unit 10.

  The correction unit 10 corrects the error part of the TS packet by using the TS packet and error information input from the error detection unit 7 and the storage data input from the correction memory 9. The correction unit 10 outputs the corrected TS packet from the error correction unit 12. Further, the correction unit 10 outputs the corrected TS packet to the correction memory 43 based on the TS packet correction method.

  The TS packet output from the error correction unit 12 is output to, for example, a decoder (not shown). The decoder decodes video and audio from the input TS packet, and reproduces video / audio with AV synchronization based on PCR information. Alternatively, the TS packet output from the error correction unit 12 is output to a multiplexer (not shown). The multiplexer extracts PSIP (Program & System Information Protocol) from the input TS packet, creates a database for managing video and audio using this PSIP, and inputs the TS packet. The video and audio data is output to the decoder, and for example, the video of the channel specified by the user is output, or the electronic program guide and data broadcast are displayed.

  Next, the operation of the stream restoration device in this embodiment will be described. First, the antenna 1 receives an ATSC broadcast wave. The received broadcast wave is synchronized by the synchronous reproduction unit 2 to clarify a group of symbols corresponding to one data. Subsequently, the waveform equalization unit 3 generates a waveform from the symbols clarified by the synchronous reproduction unit 2. At this time, the symbol remains affected by the transmission path, and is affected by attenuation, rotation, and delay with respect to the waveform during transmission. Therefore, in order to return to a waveform close to that at the time of transmission, the waveform equalizer 3 corrects the waveform. The corrected symbol is output to the error correction unit 4.

  The error correction unit 4 decodes the trellis code applied to the broadcast wave, error-corrects the decoded data in byte units with the RS code, outputs the corrected data to the separation unit 5 as a transport stream, and The corrected portion subjected to the error correction is output to the error repair unit 12.

  The error correction unit 4 includes a trellis decoding unit 41, an RS decoding unit 42, and a correction memory 43. The trellis decoding unit 41 decodes the symbols and generates a 1656-bit bit string. This is output to the RS decoding unit 42 and the correction memory 43. The RS decoding unit 42 performs error correction on the 1656-bit (207 bytes) bit string input from the trellis decoding unit 41 in units of bytes. The last 20 bytes of the 207 bytes are RS codes, and the remaining 187 bytes of data are corrected using the 20-byte RS code. The RS code is a parity (error code) attached to 207 bytes, and an error may exist in 20 bytes of the RS code. The RS decoding unit 42 outputs the corrected 187-byte data to the separation unit 5 as a transport stream.

  Here, there is a limit to the correction capability of the RS decoding unit 42 that corrects an error. When the number of bytes with errors exceeds 10 bytes, correction becomes impossible. When there is no error, the transport stream output from the RS decoding unit 42 is not subjected to correction processing and becomes 187 bytes of data from the front of 207 bytes. When the number of bytes with errors is 10 bytes or less, the error is corrected. If there are more than 11 bytes with errors, the correction process is not performed and 187 bytes of data from the previous 207 bytes are displayed. The first bit of the third byte indicates that there is an error. Is set to 1. When the byte having an error is 11 bytes or more, the first bit of the third byte of the transport stream output from the RS decoding unit 42 is TRANSPORT_ERROR_INDICATOR of the transport packet header, and an error occurs in the transport packet on the receiving side. It is a bit for recognizing a possibility. Usually, it is sent as 0 in a broadcast wave.

  The correction capability of the RS decoding unit 42 depends on the demodulator. Since 10 bytes is currently the maximum, in this embodiment, the correction capability of the RS decoding unit has been described as 10 bytes. However, the present invention is not limited to this.

  Further, the RS decoding unit 42 outputs the corrected part subjected to the error correction to the error repairing unit 12. Since error correction is processed in units of bytes, for example, when the 1st, 15th, 63rd, 148th and 170th bytes are corrected, the corrected locations are output as numbers 1, 15, 63, 148 and 170. To do.

  The separation unit 5 adds 1 byte to the head of the 187-byte transport stream input from the RS decoding unit 42 to obtain 188 bytes, and outputs 188 bytes to the type estimation unit 6 as a TS packet. The 1-byte data added to the head is a fixed value of 0x47. Since the antenna 1 continues to receive broadcast waves, the transport stream input from the RS decoding unit 42 to the separation unit 5 is transmitted 187 bytes at a time without interruption. The separation unit 5 has a buffer inside, accumulates the inputted 187 bytes, and starts processing after a certain period of time. This is because the processing of the next TS packet is performed after the processing of the TS packet output immediately before is completed.

  The type estimation unit 6 estimates what the TS packet input from the separation unit 5 is a TS packet to transmit (the type of TS packet). There are 15 types of ATSC TS packets. The three types are PAT, CAT, and PMT defined in ISO / IEC13818-1. Nine types are TS packets that transmit data (PSIP) for managing video unique to ATSC, and are MGT, TVCT, CVCT, RRT, EIT, ETT, STT, DCCT, and DCCSCT. The remaining three types are TS packets that transmit PES packets of video ES, audio ES, and data ES. The type estimation unit 6 estimates which one of these.

  The operation of the type estimation unit 6 will be described. First, the PID and Table_ID of the TS packet input from the separation unit 5 are extracted, and the type of the TS packet is estimated from these values. The PID is 13-bit data from the 12th bit to the 24th bit from the beginning of the TS packet, and indicates the type of data stored in the payload portion of the TS packet. Table_ID indicates the ID of data transmitted in the section format. Note that the value of PID and the value of Table_ID are determined by the MPEG-2 system and ATSC standard.

  FIG. 2 shows a flowchart of the operation in which the type estimation unit 6 in this embodiment estimates the type. In S1, it is determined whether or not the PID is 0x0000. If the PID is 0x0000, the TS packet type is estimated as PAT, and then the process proceeds to S9. If the PID is not 0x0000, the process proceeds to S2. In S2, it is determined whether or not the PID is 0x0001. If the PID is 0x0001, the TS packet type is estimated to be CAT, and then the process proceeds to S9. If the PID is not 0x0001, the process proceeds to S3. In S3, it is determined whether or not the PID is 0x1FFF. If the PID is 0x1FFF, the TS packet type is assumed to be a NULL packet, and then the process proceeds to S9. If the PID is not 0x1FFF, the process proceeds to S4. In S4, it is determined whether or not the PID is 0x1FFB. If it is 0x1FFB, the process proceeds to S8, and if it is not 0x1FFF, the process proceeds to S5.

  In S5, it is determined whether Table_ID is 0xCB. If Table_ID is 0xCB, the TS packet type is estimated to be EIT, and then the process proceeds to S9. If Table_ID is not 0xCB, the process proceeds to S6. In S6, it is determined whether Table_ID is 0xCC. If Table_ID is 0xCC, it is estimated that the type of TS packet is ETT, and then the process proceeds to S9. If Table_ID is not 0xCC, the process proceeds to S7. In S7, it is determined whether Table_ID is 0x02. If Table_ID is 0x02, the TS packet type is estimated as PMT, and then the process proceeds to S9. If Table_ID is not 0x02, the TS packet type is estimated as ES (video ES, audio ES, or data ES). Then, the process proceeds to S9.

  In S8, the type of TS packet is estimated from Table_ID. When Table_ID is 0xC7, the type of TS packet is estimated as MGT. When Table_ID is 0xC8, the TS packet type is estimated as TVCT. When Table_ID is 0xC9, it is estimated that the type of TS packet is TVCT. When Table_ID is 0xCA, the TS packet type is estimated to be RRT. When Table_ID is 0xCD, the type of TS packet is estimated as STT. When Table_ID is 0xD3, the type of TS packet is estimated as DCCT. When Table_ID is 0xD4, the type of TS packet is estimated as DCCSCT. S9 outputs the type of TS packet as an estimation result.

  Here, the TS packet type is expressed as “estimated” instead of “determined” because the TS packet input to the type determining unit 6 may contain an error, and the location where the error occurred identifies the type. This is because it is not possible to make a correct determination if the bit is. For example, although the PID of the PAT is defined as 0x0000, if this PID is 0x0000 as a result of an error in the PID of a TS packet other than the PAT, the type of this TS packet is determined to be PAT. There is a case.

  As shown in FIG. 2, PID and Table_ID are used to estimate the type of TS packet. However, the method for estimating the type of TS packet is not limited to this. For example, the type can be estimated using an adaptation field or syntax. This is because of the 15 types of TS packets, TS packets that may have an adaptation field are two types, PAT and PMT. This is because PSIP has a unique syntax, and the value is determined by the type of TS packet, or there are bits that can be estimated.

  FIG. 3 shows the syntax of TVCT, which is one of PSIPs in this embodiment. The notation in the Format column in the figure represents the following contents.

uimsbf: Integer value starting from upper bit bslbf ... Bit string starting from left bit rpchof ... Remainder polynomial coefficient starting from upper order

  When there is an adaptation field, the value of adaptation_field_control is 0x10 or 0x11. Only when there is a stuffing byte in the adaptation field and there is a payload after the adaptation field, 0xFF follows between the 4-byte header and the payload. Therefore, it is possible to determine whether or not there is an adaptation field.

  The type estimation unit 6 outputs the type of TS packet as an estimation result, but there is an error in the TS packet, and the type of TS packet may not be specified by the operation shown in FIG. In that case, all possible TS packet types may be output. Furthermore, priorities may be assigned in order of the highest possibility, and the types and priorities may be combined and output to the error detection unit 7. The priority order can be determined, for example, by the number of items that meet the determination criteria among PID, Table_ID, and syntax.

  If there is an error in the TS packet and the TS packet type cannot be specified, a memory is provided in the type estimation unit 6 to store the data of the PAT, PMT, and all PSIP TS packets received in the past. The TS packet input from the separation unit 5 may be compared with the TS packet stored in the internal memory, and the type may be estimated based on the degree of coincidence with other data areas even if there is a difference in PID or Table_ID. .

  Note that the type estimation unit 6 may estimate which of the 15 types of TS packets, and is necessary for outputting video and audio only for the purpose of outputting video and audio. You may limit to estimation of whether it is a TS packet. The minimum TS packets required for video and audio output are a TS packet for transmitting video ES, a TS packet for transmitting audio ES, and a TS packet (PAT for transmitting data for managing video PID and audio PID). , PMT, TVCT).

  In the case of the ATSC system, the TS packet for transmitting data for managing the video PID and the audio PID differs depending on the contents of the TVCT. If the TVCT has a Service Location Descriptor, it is a TVCT. In the Service Location Descriptor, the PID of the TS packet that transmits the video ES and the PID of the TS packet that transmits the audio ES exist. On the other hand, if the TVCT does not have a Service Location Descriptor, the video PID and audio PID are managed in three types: TVCT, PAT, and PMT. In this management method, the PMT has a video PID and an audio PID for the program number. TVCT designates Program number and PAT indicates PID of PMT corresponding to Program number.

  The type of TS packet and the TS packet estimated in this way are output from the type estimation unit 6 to the error detection unit 7.

  The error detection unit 7 detects whether there is an error in the TS packet by using the error location input from the RS decoding unit 42 and the TS packet and the type of TS packet input from the type estimation unit 6, If there is an error, identify the error location.

  FIG. 4 is a flowchart showing the operation of the error detection unit 7 in the present embodiment. With reference to FIG. 4, a method for detecting whether or not there is an error in the error detection unit 7 and a method for specifying the error location will be described.

  First, in S1, it is determined whether Transport_error_indicator indicating that the RS decoding unit 42 has corrected the TS packet error is 0 or not. Here, Transport_error_indicator is the first bit of the second byte of the TS packet and is a signal rewritten by the error correction unit 4. In broadcast waves, Transport_error_indicator is always transmitted as 0. The error correction unit can determine the criterion for setting Transport_error_indicator to 1. In the present embodiment, the RS decoding unit 42 rewrites to 1 when there is an error in the TS packet regardless of whether or not error correction has been performed. If the Transport_error_indicator is 0 in S1, “no error” is determined and the process proceeds to S6. If Transport_error_indicator is not 0, the process proceeds to S2. In S2, the correction part input from the RS decoding unit 42 is determined. When there is a correction part, it means that the TS packet has an error before the RS decoding unit 42, but the error has already been corrected by the RS decoding of the RS decoding unit 42. In this case, the process proceeds to S3 as “no error”. When there is no correction part, it means that the TS packet cannot correct the error by the RS decoding unit 42. In this case, “error is present” and the process proceeds to S4.

  In S3, it is determined from the syntax and the bit data that can be estimated from the syntax and CRC (in the case of a TS packet with CRC) whether or not the correction portion is correct. If the correction location is correct, the TS packet is determined as “no error” and the process proceeds to S6. If the correction part is not correct, the TS packet has “error”, and the correction part is regarded as an error part and the process proceeds to S6. Since the TS packet having no correction portion in S2 includes an error, the type of the TS packet output from the type estimation unit 6 may be incorrect. Therefore, S4 re-estimates the type of TS packet. For example, the type of TS packet with the smallest difference is used as the estimation result compared with all 15 types of syntax. Alternatively, the bit as an estimation key is matched, and the type of TS packet having a high matching ratio is used as the estimation result. Then, it progresses to S5. In S5, the error location is specified, and the process proceeds to S6. In S5, the TS packet type syntax estimated in S4 is compared with bit data that can be estimated from the syntax, and a different location is determined as an error location. In S6, the presence / absence of an error in the TS packet and the error location in bit units when there is an error are output as error information.

  In addition, when the TS packet input from the type estimation unit 6 is a TS packet that specifies the PID of a video TS packet, an audio TS packet, or a PCR TS packet, the error detection unit 7 stores the TS packet as stored data. The data is output to the determination unit 8 and the determination result is received from the stored data determination unit 8. For example, when the type of TS packet is TVCT and there is no error as a result of error detection, the error detection unit 7 outputs the TVCT TS packet to the storage data determination unit 8. The saved data determination unit 8 analyzes the TVCT input from the error detection unit 7, and when the TVCT has a Service Location Descriptor, decides to save the TVCT and outputs it to the error detection unit 7. If the TVCT does not have a Service Location Descriptor, the TVCT, PMT, and PAT TS packets are determined to be stored and output to the error detection unit 7.

  The error detection unit 7 extracts data to be stored from the TS packet input from the type estimation unit 6 based on the determination result input from the storage data determination unit 8 and outputs the data to the correction memory 9. For example, if the determination result is to save the TVCT, the TVCT TS packet 188 bytes is output to the correction memory 9.

  The correction memory 9 outputs the stored data to the correction unit 10. The correction unit 10 uses the error information (presence / absence of error and error location), the TS packet input from the error detection unit 7 and the data input from the correction memory 9 to correct the error location of the TS packet. The correction is performed by replacing the data input from the memory 9 or rewriting the data estimated from the data input from the correction memory 9. Here, a method of generating the estimated data will be described. Among TS packets, there is data that changes based on a certain rule, such as continuity_counter (TS packets of the same type have 16 modulo values of consecutive numbers). This part generates data obtained by changing the data input from the correction memory 9 based on the law. The correction unit 10 outputs the corrected TS packet from the error correction unit 12, and the correction unit 10 outputs the corrected TS packet to the correction memory 43 based on the TS packet correction method and the number of corrections.

  For example, when the number of times of repair exceeds a predetermined number, the repair unit 10 does not return the repaired TS packet to the correction memory 43 but outputs it from the error repair unit 12 to the outside. The predetermined number may be determined from the operation clock of the stream restoration device and the frequency of the broadcast wave. For example, if the operation clock of the stream repair device is 176 MHz and can be processed at 352 Mbps, a TS packet of ATSC (transmitted at 19 Mbps) can be repaired up to 18 times. Further, the predetermined number is set to 5, and the correction unit 10 repairs TS packets as TVCT, PAT, PMT, video ES TS packets and audio ES TS packets, and only these 5 types can be repaired correctly. It may be. In this case, the storage data determining unit 8 determines five types of TVCT, PAT, PMT, video ES TS packets and audio ES TS packets as storage data.

  Note that the correction method of the correction unit 10 may replace the entire TS packet instead of correcting only the error part.

  FIG. 5 is an example of a TS packet handled by the error correction unit 13. Here, a specific example of the TS packet will be described with reference to FIG. 5, and operations of the error detection unit 7 and the correction unit 10 will be described. FIG. 5A assumes a TVCT TS packet at the time of transmission, and is a TS packet with no error. FIG. 5B shows the data type of each byte in FIG. Data types are classified into the following three types.

(1) What has a fixed value by syntax (2) What can be estimated from syntax and data around TS packet and TS packet received in the past (3) Information that is necessary for reproduction

  In addition, the part enclosed with the square like b1 shows (1), it encloses with the broken line like b2, shows the thin shaded part shows (2), and the thick part shaded like b4 shows (3 ).

  Here, an estimation method corresponding to the above (2) will be described in the case of b2, b5 and b7 in FIG. b2 is continuity_counter. Since this b2 is assigned a continuous number in TS packets having the same PID, the error detection unit 7 counts the number of PIDs, and b2 in TS packets having the same PID. The value can be estimated. Further, b5 is a byte indicating the number of bytes up to b6 in which “f” appears next to b5, and therefore the value of b5 can be estimated. Further, the region filled with “ff” shown in b7 can be estimated from the value of b3 as follows. b3 is a section length, and the value (hexadecimal number) of b3 represents the length from the beginning of the TS packet to the end of the section data (here, the byte immediately before the area indicated by b7). Therefore, the area indicated by b7 is an area unnecessary as data, and is entirely filled with “ff”. In this way, the byte corresponding to the above (2) is corrected by the error correction unit 7.

  (C) in FIG. 5 assumes a received TS packet, and it is assumed that a portion shown in bold is an error. After the TVCT is estimated by the type estimation unit 6, the data is restored according to the data type (b). By comparing the data fixed by the syntax with the data estimated by the syntax, a different location is detected as an error location. The detected error location is output to the correction unit 10. The correction unit 10 rewrites the error location input from the error detection unit 7 with data corresponding to the error location stored in the correction memory 9. For example, c1 “55” shown in FIG. 5C is converted to d1 “00”. This is because d1 has a fixed value by the syntax corresponding to the above (1). FIG. 5D shows the corrected TS packet corrected in this way.

  When a TS packet for transmitting a video ES is corrected, the digital broadcast video ES is compliant with MPEG2, and therefore has a bit having a fixed value and a bit that can be estimated by the syntax. These can be used to correct TS packets that transmit video ES. In this case, the error detection unit 7 outputs the TS packet for transmitting the video ES to the storage data determination unit 8, and the storage data determination unit 8 estimates the bit that can be estimated from the syntax of the video ES and the input TS packet. Is returned to the error detection unit 17 as a determination result. For example, a bit having a fixed value by syntax includes a start synchronization code of each layer. The bits that can be estimated include Picture Coding Type, a frame number in the GOP, and a location indicating a slice number.

  As described above, according to the stream restoration device of the present embodiment, the TS packet corrected by the error correction unit is error-corrected again by the demodulation unit, and then re-corrected by the error correction unit, so that it is output from the error correction unit. This makes it possible to improve the repair performance of TS packets, and to reduce the time during which playback is impossible.

  Note that the TS packet that has been error-corrected again by the demodulating unit may be output to a decoder or multiplexer provided at the subsequent stage of the error correcting unit, instead of being output to the error correcting unit again.

  Further, according to the stream restoration device of the present embodiment, it is possible to control the processing of one TS packet by determining whether or not to correct the TS packet corrected by the error correction unit again. Thus, the error repair process can be reliably realized with the TS packet output from the error correction unit, and the time during which reproduction cannot be performed can be reduced.

  In addition, since the error correction unit estimates the type of TS packet using syntax and estimates the data of the error location using syntax, TS packets can be corrected even when there are errors exceeding the error correction capability. is there. Furthermore, the error correction unit can only correct up to about 10 bytes per packet. However, since the error correction unit can estimate data of the error location by syntax, the error in the TS packet after correction can be about 10 bytes. For example, by performing error correction again, it becomes possible to perform almost complete correction, improving the repair performance of the TS packet and reducing the time during which reproduction becomes impossible.

  In addition, according to the stream repair device of the present embodiment, error correction is performed using an outer code uniquely determined by a TS packet, so that a byte with an error is determined regardless of the contents of the transport packet. However, the error correction unit detects the error based on the syntax and the bit value that can be estimated from the syntax, so it is possible to identify the error location according to the contents of the transport stream, and the error location more accurately. Can be specified. As a result, it is possible to improve the repair performance of the TS packet output from the error correction unit, and it is possible to reduce the time during which reproduction is impossible.

  Further, according to the stream repair device of the present embodiment, the error correction unit uses the correction part output by the error correction unit in order to specify the error part of the transport stream, so that the error part is specified with higher accuracy. This makes it possible to improve the performance of repairing TS packets output from the error correction unit, thereby reducing the time during which reproduction is impossible.

  Also, according to the stream repair device of the present embodiment, transport packet errors can be repaired using data of TS packets received in the past. It is possible to reliably correct, it is possible to improve the repair performance of the TS packet output from the error correction unit, and it is possible to reduce the time during which reproduction is impossible.

  Further, according to the stream repair device of the present embodiment, since the demodulation unit has an error correction unit having a storage function, it is possible to share the error correction unit included in the demodulation unit, with a smaller circuit scale. Repair of the TS packet becomes possible.

  Further, according to the stream restoration device of the present embodiment, it is possible to limit the transport packet for correcting the error, and it is possible to correct the TS packet with a smaller processing amount.

  In this embodiment, the error correction unit outputs the corrected data as a transport stream to the separation unit, and the separation unit separates the transport stream into TS packets. However, the error correction unit is not necessarily separated from the error correction unit. There is no need to make up parts. The error correction unit may have a function of separating the transport stream into TS packets, and may be configured to exclude the separation unit.

Embodiment 2. FIG.
FIG. 6 is a block diagram of a stream repair apparatus according to Embodiment 2 for carrying out the present invention. As shown in FIG. 6, the stream restoration device 200 according to the present embodiment includes a demodulation unit 13, a separation unit 15, an error correction unit 16, and a second error correction unit 18. Similarly to the first embodiment, the demodulation unit 13 includes a synchronous reproduction unit 2 that receives broadcast waves from the antenna 1, a waveform equalization unit 3, and a first error correction unit 14. Similarly to the first embodiment, the error correction unit 16 includes a type estimation unit 6, an error detection unit 17, a stored data determination unit 8, a correction memory 9, and a correction unit 10. However, unlike the first embodiment, the first error correction unit 14 includes a trellis decoding unit 41 and an RS decoding unit 42, and does not include a correction memory.

  The stream repair apparatus 200 according to the present embodiment is significantly different from the first embodiment in that the second error correction unit 18 is provided. The second error correction unit 18 includes an RS decoding unit 45 and a correction memory 46 and is outside the demodulation unit 13. The demodulator 13 is an 8-VSB demodulator as in the first embodiment.

  Components having the same names as those of the first embodiment perform the same operations as those of the first embodiment except for the demodulator 13 and the RS decoder 44. Since the demodulating unit 13 does not have a correction memory, the repaired TS packet is not input from the repairing unit 10, and thus the RS decoding unit 44 has no input from the correcting memory. In addition, constituent elements having the same names and different numbers as those of the first embodiment are different in data input source and output destination from those of the first embodiment. Only the differences will be described below.

  The separation unit 15 receives the transport stream after error correction from the RS decoding unit 44 in the first error correction unit 14 or from the RS decoding unit 45 in the second error correction unit 18. The error detection unit 17 is input with a corrected part subjected to error correction from the RS decoding unit 44 or the RS decoding unit 45.

  In the stream repair device configured as described above, it is not necessary to add a memory for storing broadcast waves before error correction to the demodulator, and the second error correction provided with a correction memory separately from the demodulator Therefore, the TS packet can be restored by using a conventional demodulator for digital broadcasting. As a result, similar to the first embodiment, the TS packet corrected by the error correction unit is corrected again by the demodulation unit and then re-corrected by the error correction unit, so that the TS packet output from the error correction unit is restored. Can be improved, and the time during which reproduction cannot be performed can be reduced.

DESCRIPTION OF SYMBOLS 1 Antenna 2 Synchronous reproduction part 3 Waveform equalization part 4 Error correction part 5, 15 Separation part 6 Type estimation part 7, 17 Error detection part 8 Saved data determination part 9 Correction memory 10 Correction part 11, 13 Demodulation part 12, 16 Error correction unit 14 First error correction unit 18 Second error correction unit 41 Trellis decoding units 42, 44, 45 RS decoding units 43, 46 Correction memory

Claims (8)

A first storage unit that receives the digital broadcast and stores data related to the data stream after decoding by the error correction inner code ;
A decoding unit that performs error correction decoding by an error correction outer code on the decoded data stream;
An estimation unit that estimates the type of decoded data output from the decoding unit;
A detection unit for detecting an error in the decoded data based on an estimation result of the estimation unit;
A correction unit for correcting an error in the decoded data based on a detection result of the detection unit;
A second accumulation unit that accumulates the corrected data corrected by the correction unit ,
The decoding unit
The relative first the corrected data stored in the second storage unit based on an error correction outer code contained in the data stream after the decoded stored in the storage unit, performs error correction again
A digital broadcast receiver characterized by that. The decoded data stream is a data stream related to a transport stream, and the data stored in the first storage unit is data related to a predetermined packet including an error correction outer code of the transport stream, The data stored in the second storage unit is packet data corresponding to the predetermined packet.
The digital broadcast receiver according to claim 1. The decoding unit
Based on the error correction outer code included in the bit string for one packet stored in the first storage unit, the number of times that error correction is performed on the packet data stored in the second storage unit is counted. Is output to the correction unit,
The correction unit determines whether or not packet data corrected based on the number of times or the estimation result regarding the type of packet data output from the estimation unit is stored again in the second storage unit.
The digital broadcast receiver according to claim 2. The detector is
An error is detected based on a predetermined syntax according to the type of packet estimated by the estimation unit and a bit value that can be estimated from the syntax ,
The digital broadcast receiver according to claim 3, wherein the correction unit corrects the error bit . A first accumulation step for receiving the digital broadcast and accumulating data relating to the data stream after decoding by the error correction inner code;
A decoding step of performing error correction decoding with an error correction outer code on the decoded data stream;
An estimation step for estimating the type of decoded data output from the decoding unit;
A detection step of detecting an error in the decoded data based on an estimation result of the estimation unit;
A correction step of correcting an error in the decoded data based on a detection result of the detection unit;
A second accumulation step for accumulating post-correction data corrected by the correction unit;
With
In the decoding step, an error is again generated with respect to the corrected data accumulated in the second accumulation step based on the error correction outer code included in the decoded data stream accumulated in the first accumulation step. Make corrections
A digital broadcast receiving method. The decoded data stream is a data stream related to a transport stream, and the data stored in the first storage step is data related to a predetermined packet including an error correction outer code of the transport stream, The data stored in the second storage step is packet data corresponding to the predetermined packet.
The digital broadcast receiving method according to claim 5. The decoding step includes
Based on the error correction outer code included in the bit string for one packet accumulated in the first accumulation step, the number of times of performing error correction on the packet data accumulated in the second accumulation step is counted,
In the correction step, it is determined whether or not packet data corrected based on the number of times or the estimation result relating to the type of packet data estimated in the estimation step is stored again in the second storage step.
The digital broadcast receiving method according to claim 6. The detecting step includes
An error is detected based on a syntax determined in advance according to the type of TS packet estimated in the estimation step and a bit value that can be estimated from the syntax,
The correcting step corrects the erroneous bit.
The digital broadcast receiving method according to claim 7.

Images