JP2005236363A - Packet separator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a packet separator capable of relaxing accesses to an external memory for storing extracted data and a log related to the extraction without the need for increasing the capacity and the number of buffers and without imposing excess load required for data analysis on a microcomputer. <P>SOLUTION: The packet separator (111) for extracting packets from a given transport stream and extracting section or non-section data included in the packets, includes: a section length calculation section (161) for calculating a section length as a characteristic of the section; a log size management section (171) for setting a size for a section log in accordance with the calculated section length; and a log generating section (123) for generating the section log with the set size. The log size management section (171) compresses the size of the section log when the calculated section length is relatively short. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a packet separator that separates section data and non-section data from a packet extracted from a transport stream, and more particularly to log generation in section data separation processing.

  In a conventional packet separation device, the result of filtering by a PID (packet identifier) filter or the result of filtering section data is always recorded as a log in a fixed format (for example, see Patent Document 1). In the following, starting with the description of the transport stream, section data extraction and log generation will be described.

  In recent years, broadcasting such as digital satellite broadcasting, digital cable broadcasting, and terrestrial digital broadcasting that transmits video, audio, and related data as digital data has been performed. Most of the time, MPEG (Moving Picture Experts Group) standard transport stream format is used, and digitized video and audio data and related data (program specification information, hereinafter referred to as “PSI”) are used. include.

  In digital broadcasting, video, audio, and various types of information are transmitted in a transport stream (hereinafter referred to as “TS”). FIG. 19 shows the format of TS. The TS is composed of a collection of transport stream packets (hereinafter referred to as “TSP”) having a fixed length of 188 bytes. The TSP is composed of three parts: a header, an adaptation field (hereinafter referred to as “AF”), and a payload (hereinafter referred to as “PL”). The TSP header has a fixed length of 4 bytes, and is composed of an 8-bit synchronization byte, a 13-bit PID, a 4-bit continuous counter (hereinafter referred to as “CC”) as a packet continuity index, and the like. AF is variable-length data, and the first byte is the data length, and represents the number of bytes of AF data that follows. PL is data having a length obtained by subtracting a 4-byte header and (AF data length + 1) bytes from a 188-byte TSP. Then, by collecting TSPs with the same PID, the original data divided every 188 bytes can be restored.

  The data included in the PL is large and can be divided into two types of formats: sections and PES (Packetized Elementary Stream) packets. The video and audio data are stored in the PL in the PES packet format. On the other hand, the PSI is stored in the PL in the section format. The section includes various identification data such as a table ID (hereinafter referred to as “TID”) and a section length.

  FIG. 20 shows the configuration of a digital broadcast receiver equipped with a conventional packet separator. A digital broadcast signal received by the antenna 140 passes through a tuner 141 in the digital broadcast receiver 201 and becomes a TS. The PID filter 120 in the packet separation device 211 inputs TS and extracts only necessary TSP. If the data is encrypted, the PID filter 120 descrambles the data and converts it to the original data. Each TSP is identified by PID, and this identification function is called PID filtering. The PID filter 120 performs the PID filtering.

  The section filter 121 further filters the data of the section included in the TSP and extracts necessary data. When the TSP includes data other than a section, the non-section processing unit 122 checks a header included in the data and outputs non-section data (for example, PES packet data). Further, the non-section processing unit 122 outputs the check result as non-SF error information. Whether or not the contents of the TSP are a section is determined by the section determination unit 125 and the determination result is output as a section flag. The section flag is used to control the section filter 121 and the non-section processing unit 122. Data extracted by the section filter 121 and the non-section processing unit 122 is written to the external memory 152 via the data transfer unit 224 and the external memory control unit 151.

  The video and audio data is simply buffered and then decoded by an audio video decoder (hereinafter referred to as “AV decoder”) 153. Section data such as PSI is read by the microcomputer 154 and the like, and becomes data such as an electronic program guide (EPG), a key for unscrambling, and billing information.

  FIG. 21 shows the internal configuration of the PID filter 120. The synchronization detection unit 301 detects synchronization for a TS that is an input of the PID filter 120, and identifies the head position of the TSP. The header analysis unit 302 extracts the TSP PID based on the head position. The header analysis unit 302 outputs PID error information when an error is detected by checking CC continuity and the like. The PID condition table 303 registers up to 64 PIDs (PID conditions) to be extracted. Further, the PID condition table 303 stores a valid flag indicating whether or not the corresponding PID condition is valid. The selector 304 selects and outputs the PID condition of the entry number having the same value as the counter value from the PID condition table 303 according to the counter value of the counter 305. The comparator 306 compares the selected PID condition with the extracted PID. If they match, the latch 307 latches the corresponding entry number and outputs it as a PID entry number. If the PID condition stored in the PID condition table 303 is not valid, the comparator 306 regards it as a mismatch. The latch 307 outputs a PID match signal indicating that the selected PID condition matches the extracted PID as a PID filtering result, and outputs the PID error information received from the header analysis unit 302. The TS switch 309 is turned on by receiving the PID match signal, and allows the TSP detected by the synchronization detection unit 301 to pass through. The PID entry number, the PID filtering result, and the PID error information are collectively referred to as “PID filter information”.

  FIG. 22 shows the internal configuration of the section filter 121. The section analysis unit 402 extracts a section header from the TSP that is an input of the section filter 121. The section analysis unit 402 outputs SF error information when an error is detected by examining a section length, a pointer field (hereinafter referred to as “PF”), a CRC, and the like. The SF condition table 403 registers a maximum of 64 sections (SF conditions) to be extracted. “X” in the SF condition represents don't care. The SF condition table 403 stores a valid flag indicating whether or not the corresponding SF condition is valid. The selector 404 selects and outputs the SF condition of the entry number having the same value as the counter value from the SF condition table 403 according to the counter value of the counter 405. The comparator 406 compares the selected SF condition with the extracted section header. If they match, the latch 407 latches the corresponding entry number and outputs it as the SF entry number. When the SF condition stored in the SF condition table 403 is not valid, the comparator 406 regards it as a mismatch. In addition, the latch 407 outputs an SF result such as an SF filtering length indicating how many bytes have been compared, a descending flag indicating that the filtering process has not been completed only from the data of the TSP, and the like. The SF error information received from the section analysis unit 402 is output. Further, the latch 407 outputs an SF match signal indicating that the selected SF condition matches the extracted section header. The TSP switch 409 is turned on by receiving the SF match signal, and passes the TSP input by the section filter 121. The SF entry number, the SF filtering length, the SF result, and the SF error information are collectively referred to as “SF information”. The SF information is important information for separating and selecting various information included in the TS.

  Returning to FIG. 20, the log generation unit 223 collects information such as the PID filter information from the PID filter 120, the SF information from the section filter 121, and the write address to the external memory 152, and converts the information into a predetermined format. Sort and generate log.

  FIGS. 23A, 23B, and 23C show the log formats for sections, PES packets, and errors. The lengths of the section log, PES log, and error log are 8 bytes, 6 bytes, and 4 bytes, respectively. The contents of the log will be described below with reference to FIGS. 23 (a) to 23 (c). In the figure, “*” indicates an empty bit.

  The “format” of 2-bit length represents the type of log. When the “format” value is “0”, it is a section log, “1” is a PES log, and “2” is an error log.

  The “PID entry number” having a 6-bit length represents which entry number from “0” to “63” matches the PID in the PID filtering process. The TS includes 13 bits, that is, 8192 types of PIDs, and it is possible to extract TSPs having 64 types of PIDs at the maximum.

  The “data length” of 12-bit length represents the length of section data or a PES packet. According to the MPEG standard, section and PES packet length information is embedded near the beginning of the data, and is almost the same as the “data length” recorded in the log. However, in actuality, the “data length” recorded in the log is the length recorded in the external memory 152. Therefore, when an error occurs during the filtering process and the data is interrupted, they match. Will not.

  “Error information” of 4-bit length in the section log and PES log is when the CRC value in the section does not match, the length information included in the data does not match the actual data length, or the section data crosses the TSP Includes the inconsistency between the start position of the section data pointed to by the PF and the section length, CC discontinuity information included in the header of the TSP, and the like.

  “Error information” of 8-bit length in the error log includes an error due to the bit of the transport error indicator, an error due to TS being out of synchronization, an AF length error, and the video and audio data as encrypted. It includes information such as an error to be output and an error due to CC discontinuity.

  The “pointer address” having a length of 24 bits represents the start address of the corresponding section and PES packet. In the case of an error log, there is no pointer address itself because there is no data extracted from the corresponding TSP.

  The “SF entry number” having a 6-bit length in the section log represents which entry number from “0” to “63” the section matches in the section filtering process.

  The “SF filtering length” of 5 bits in the section log indicates how many bytes from the head of the section match the section filter condition in the section filtering process. Normally, the value of “SF filtering length” matches the value set as the section condition, but when the section crosses the TSP in the section filtering process, the above-described descending flag is set and the section filter up to the number of bytes Indicates whether the condition is met.

  The “SF result” having a 2-bit length in the section log includes the above-described participation flag, information on whether or not the CRC calculation value is “0”, and the like.

  Returning to FIG. 20, the data transfer unit 224 inputs the section data output from the section filter 121, the log data output from the log generation unit 223, and the non-section data output from the non-section processing unit 122. Data is appropriately selected and output to the external memory control unit 151. The external memory control unit 151 inputs and outputs data with the external memory 152, the AV decoder 153, the microcomputer 154, and the data transfer unit 224.

FIG. 24 shows various data storage areas in the external memory 152. A data storage area is determined for each PID entry number, and data with the same PID is collected in the same area. In addition, logs corresponding to each data are collected in one place of the external memory 152. For example, the log LOG-A1 corresponds to the PES packet in the TSP with the entry number PID_A. Similarly, the logs LOG-B1, -B2, and -B3 correspond to the section data SECT-B1, -B2, and -B3 in the TSP of the entry number PID_B. Thus, one log is generated and stored for each PES packet and one section data. Note that in Patent Document 1, only information relating to the result of section filtering is referred to as a log, but in the present invention, information including a write address to the external memory 152 is referred to as a log. This is because when writing data to the external memory 152, information (pointer address) about which address the data is stored in is also required.
JP 2000-156705 A (pages 8 to 10, FIGS. 3 and 4)

  As described above, the conventional packet separator generates one fixed-length log per section. For this reason, as the number of sections increases, the number of logs increases accordingly. In particular, when a number of relatively short sections are included in one TSP, the log data volume may increase to the same extent as the section data volume. For example, if the section length is 8 bytes, a maximum of 23 (184 ÷ 8 = 23) sections can be included in one TSP. In this case, the log data amount is 184 bytes (8 × 23 = 184), and the log has the same size as the section data.

  As the TS input speed increases, the amount of data to be stored in the external memory 152 per unit time further increases. Further, when a so-called unified memory configuration is used in which a plurality of external memories 152, for example, the microcomputer 154 and the AV decoder 153 are shared, access to the external memory 152 becomes more congested and access waiting time increases. End up. In order to cope with this problem, in order to prevent data from overflowing, it is necessary to increase the buffer capacity for accessing the external memory 152, but this is preferable because it increases cost and circuit scale. Absent.

  On the other hand, the above problem can be avoided by not generating the log. However, in this case, the microcomputer 154 must perform data analysis according to the MPEG format based on the recorded section data, and the load on the microcomputer 154 increases. Using a high-performance microcomputer to cope with this is disadvantageous in terms of cost.

  In view of the above problems, the present invention provides a packet separation device that extracts extracted data and a log related to the extraction without increasing the capacity and quantity of the buffer and without applying an excessive load related to data analysis to the microcomputer. Provided is a log generation technique for relaxing access to an external memory to be stored.

  Means taken by the present invention to solve the above-mentioned problem is a packet separation device for extracting a packet from a given transport stream and extracting section or non-section data included in the packet, wherein the section Section characteristic extracting unit for extracting the characteristics of the log, a log size managing unit for setting the size of the section log according to the section characteristics extracted by the section characteristic extracting unit, and a size set by the log size managing unit And a log generation unit for generating the section log.

  According to this, the size of the section log is set according to the section characteristics. That is, the size of the section log is variable. Therefore, the size of the section log is optimized and access to the external memory where the log is stored is relaxed compared to the case where one section log for a fixed length is generated per section. .

  Specifically, the section characteristic extraction unit calculates a section length as the characteristic of the section. Further, the log size management unit compresses the size of the section log when determining that the section length calculated by the section characteristic extraction unit is relatively short.

  More specifically, the section characteristic extraction unit calculates the number of sections per predetermined number of packets as the characteristic of the section. Further, the log size management unit compresses the size of the section log when it is determined that the number of sections calculated by the section characteristic extraction unit is relatively large.

  Preferably, the log size management unit matches the identifiers of the consecutive packets in the transport stream, and the number of sections calculated by the section characteristic extraction unit for the preceding packet is relatively large. When it is determined that the size of the section log is compressed.

  Specifically, the section characteristic extraction unit calculates the data input speed of the section as the characteristic of the section. The log size management unit compresses the size of the section log when determining that the data input speed calculated by the section characteristic extraction unit is relatively fast.

  Further, the means taken by the present invention to solve the above-mentioned problems is a packet separation device that extracts a packet from a given transport stream and extracts section or non-section data from the extracted packet. A log size management unit configured to set a size of the section log based on information relating to output control of the section log, and a log generation unit configured to generate the section log having a size set by the log size management unit; Shall be provided.

  According to this, the size of the section log is set based on the information related to the section log output control. That is, the size of the section log is variable. Therefore, the size of the section log is optimized and access to the external memory where the log is stored is relaxed compared to the case where one section log for a fixed length is generated per section. .

  Specifically, the log size management unit obtains information relating to the availability of a buffer for temporarily storing the section log as information relating to the output control of the section log, and from the information, When it is determined that there is relatively little free space, the size of the section log is compressed.

  Specifically, the log size management unit obtains information related to the degree of access to the external memory that is the storage destination of the section log as information related to the output control of the section log, and the information When it is determined that access to the external memory is relatively congested, the size of the section log is compressed.

  Preferably, the packet separation device includes a short section number calculation unit that calculates the number of sections having a section length equal to or less than a predetermined value per predetermined number of packets. When the log size management unit determines that the number of short sections calculated by the short section number calculation unit is relatively large, the log size management unit compresses the size of the section log.

  As described above, according to the present invention, by changing the size of the section log according to the section characteristics, the external memory can be accessed without particularly increasing the capacity and quantity of the buffer for accessing the external memory that is the storage destination of the log. Access to is relaxed. Further, when it is not necessary to compress the log size, a normal size log is generated. Therefore, an excessive load is not applied to the microcomputer for analyzing the section data.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

(First embodiment)
FIG. 1 shows a configuration of a digital broadcast receiver 101 including a packet separation device 111 according to the first embodiment of the present invention. The packet separator 111 includes a PID filter 120, a section filter 121, a non-section processing unit 122, a log generation unit 123, a section determination unit 125, a section length calculation unit 161 as a section characteristic extraction unit, a log size management unit 171 and a data transfer. Part 224 is provided. Among these, the PID filter 120, the section filter 121, the non-section processing unit 122, the section determination unit 125, and the data transfer unit 224 are the same as those illustrated in FIG.

  Radio waves from the digital broadcasting satellite pass through the tuner 141 from the antenna 140 and become TS. The PID filter 120 inputs TS and outputs only necessary TSP. The PID filter 120 also outputs PID filter information such as a PID entry number.

  When the TSP includes a section, the section filter 121 further performs filtering processing on the TSP to extract section data. The extracted section data is sent to the data transfer means 224. When the TSP includes data other than a section, the non-section processing unit 122 checks the header of the TSP and sends the non-section data to the data transfer unit 124. Whether the TSP includes a section or a non-section is determined by the section determination unit 125, and the determination result is output as a section flag. The section filter 121 also outputs SF information such as an SF entry number. Further, the non-section processing unit 122 outputs non-SF error information.

  FIG. 2 shows an internal configuration of the section length calculation unit 161 and the log size management unit 171 shown in FIG. The section length calculation unit 161 receives the section data output from the section filter 121 and calculates the section length. As shown in the section structure of FIG. 19H, 12 bits of the second to third bytes from the head of the section data are the number of bytes of the subsequent data. That is, the section length is obtained by adding 3 to the 12-bit data in the second to third bytes of the TSP. Specifically, the section length is calculated as follows. That is, as shown in FIG. 2, the latch 611 latches bits 3 to 0 of the second byte of TSP and 7 to 0 of the third byte. Then, the adder 612 adds “3” to the latched bit to generate a section length.

  The log size management unit 171 determines the log data format, that is, the compression mode, according to the input section length. Specifically, the log size management unit 171 includes a first section determination unit 613 and a compression mode designation unit 631. The first section determination unit 613 determines whether the input section data is for the first section in the TSP. Further, the compression mode designating unit 631 (1) “TSP is non-section data” or “the input section data is related to the first section in the TSP” or “section length is 16 bytes or more. "In normal mode", (2) "Input section data relates to the second and subsequent sections in the TSP" and "Section length is 8 bytes or more and less than 16" If the compression mode is 1 ”and (3)“ the input section data relates to the second and subsequent sections in the TSP ”and“ the section length is less than 8 ”, the“ compression mode 2 ”is selected. , Respectively.

  Returning to FIG. 1, the log generation unit 123 collects information such as PID filter information from the PID filter 120, SF information from the section filter 121, and a write address to the external memory 152, and converts the information into a predetermined format. Sort and generate log. At this time, various logs are generated according to the log compression mode specified by the log size management unit 171. This log generation will be described later. Various logs generated by the log generation unit 123 are recorded in the external memory 152 via the data transfer unit 224 and the external memory control unit 151.

  FIG. 3 shows a format of the section log in the compression mode 1. The compression mode 1 is applied to the second and subsequent sections in the TSP. Therefore, as shown in FIG. 3, the log of the compression mode 1 is stored after the log of the normal mode (in FIG. 3, the upper one is the higher address). In the log format of the compression mode 1, a “compression mode” having a 2-bit length is added, the “data length” is shortened, and the “pointer address” and the “PID entry number” are omitted. “Format” represents the type of log as in the prior art, but when the value is “3”, it represents a compressed log. The “compression mode” having a length of 2 bits represents the type of the compression mode. When the value of “compression mode” is “1”, it indicates “compression mode 1”, and when it is “2”, it indicates “compression mode 2”. As described above, since each compression mode is applied when the section length is less than 16 bytes, the “data length” need only be 4 bits or more.

  In the compressed log, “pointer address” is omitted. However, since the logs are continuously arranged, the target pointer address can be obtained by adding the data length of the previous log to the pointer address of the previous log. In the compressed log, “PID entry number” is omitted. However, since the compressed log is applied to the second and subsequent sections of the TSP, the PID entry number in the compressed log is the PID entry number of the log of the first section of the TSP (the log is always in the normal mode). The same. By using the compressed log, instead of reducing the amount of log data, the microcomputer 154 needs to add the pointer address and the data length and investigate the first log of the TSP. That is, when the compressed log is used, the load on the microcomputer 154 increases, but important information such as the pointer address and the PID entry number is not lost. As described above, in the compression mode 1 log, by omitting a part of the normal mode log, the log size becomes half that of the normal mode.

  FIG. 4 shows the format of the log for the section in the compression mode 2. Similar to the case of the compression mode 1, the log of the compression mode 2 is stored after the log of the normal mode (the upper side in FIG. 4 is the higher address). In the compression mode 2 log, “SF filtering length” and “SF entry number” in the compression mode 1 log are omitted.

  The “SF filtering length” is data indicating how many bytes the section filtering process has been completed. Therefore, the data is useful when the participation flag (1 bit of the SF result data) indicating that the filtering has not been completed up to the data at the end of the TSP is set. However, when a lot of relatively short section data is included, since the filtering processing of most sections is completed, the data of the SF filtering length is hardly utilized. The “SF entry number” is data indicating the number of the SF entry number corresponding to the section in the section filtering process. When the SF entry number is necessary, the SF entry number can be obtained by performing section filtering processing by the program of the microcomputer 154. Therefore, even if the “SF filtering length” and “SF entry number” in the log of the compression mode 1 are discarded, it is considered that the influence on the performance of the packet separator 111 is small. As described above, in the compression mode 2 log, by omitting a part of the compression mode 1 log, the log size becomes a quarter of the normal mode.

  As described above, according to the present embodiment, when the section length is relatively short, the log corresponding to the section is generated in a compressed format. That is, the amount of log data per section data is suppressed. This increases the access to the external memory 152 without increasing the capacity and quantity of the buffers in the data transfer unit 224 and the external memory control unit 151, and without applying an excessive load related to data analysis to the microcomputer 154. Can be suppressed. When the section length is relatively long, a detailed log of the corresponding section is generated as in the conventional case. This facilitates analysis of the section data by the microcomputer 154.

  Further, the calculation of the section length can be performed very easily as described above. Therefore, the section length calculation unit 161 and the log size management unit 171 can be relatively easily configured, and the circuit scale of the packet separation device 111 according to the present embodiment is significantly increased as compared with the conventional one. There is nothing.

Note that the compression mode designation criteria by the compression mode designation unit 631 described in the present embodiment are merely examples, and the present invention is not limited to this.
(Second Embodiment)
FIG. 5 shows a configuration of a digital broadcast receiver 102 including a packet separation device 112 according to the second embodiment of the present invention. The packet separator 112 includes a PID filter 120, a section filter 121, a non-section processing unit 122, a log generation unit 123, a section determination unit 125, a section number calculation unit 162 as a section characteristic extraction unit, a log size management unit 172, and data transfer. Part 224 is provided. Among these, the PID filter 120, the section filter 121, the non-section processing unit 122, the section determination unit 125, and the data transfer unit 224 are the same as those illustrated in FIG.

  The point at which section data and the like are output from the section filter 121 and the point at which non-section data and the like are output from the non-section processing unit 122 are as described in the first embodiment.

  FIG. 6 shows an internal configuration of the section number calculation unit 162 and the log size management unit 172 shown in FIG. The section number calculation unit 162 receives the section data output from the section filter 121, and counts the number of sections up to the previous TSP, that is, two TSPs. Note that the present invention is not limited to this. There are various methods for counting the number of sections, for example, a method of counting only within one TSP and a method of counting across a plurality of TSPs, depending on which range of TSPs is targeted.

  In the section number calculation unit 162, a section counter 621 and a section counter 622 are counters that count the number of sections. These are counted for each TSP. That is, when the section counter 621 performs a counting operation, the section counter 622 holds the previous value. When the next TSP is input, the value of the section counter 622 is once reset, and the section counter 622 resumes counting from 0, while the section counter 621 holds the previous value.

  The toggle FF 623 inverts the output value for each TSP. The section counter 622 and the selector 625 receive the output of the toggle FF 623. Further, the section counter 621 and the selector 624 receive the output from the toggle FF 623 inverted by the inverter 626. The section counters 621 and 622 handle the output from the input toggle FF 623 as count enable and reset enable signals. The selectors 624 and 625 select the outputs from the section counters 621 and 622 based on the output of the toggle FF 623. The multiplier 627 performs weighting by multiplying the number of sections in the previous TSP by “0.3”. The multiplier 628 performs weighting by multiplying the number of sections in the current TSP by “0.9”. Then, the adder 629 adds the outputs of the multipliers 627 and 628 and outputs the addition result to the log size management unit 172 as a section number.

  The log size management unit 172 determines the compression mode according to the input number of sections. Specifically, the log size management unit 172 includes a first section determination unit 613 and a compression mode designation unit 632. The first section determination unit 613 is as described in the first embodiment. The compression mode designating unit 632 (1) “TSP is non-section data” or “the input section data is related to the first section in the TSP” or “the number of sections is less than 10” "Normal mode" for (2) "When the input section data is related to the second and subsequent sections in the TSP" and "The number of sections is 10 or more and less than 20" “Mode 1”, and (3) “Compression mode 2” when “input section data relates to second and subsequent sections in TSP” and “the number of sections is 20 or more” , Respectively.

  The generation of the compression mode log by the log generation unit 123 and the format of the compression log are as described in the first embodiment.

  As described above, according to the present embodiment, when the number of sections is relatively large, a log corresponding to the section is generated in a compressed format. That is, the amount of log data per section data is suppressed. This increases the access to the external memory 152 without increasing the capacity and quantity of the buffers in the data transfer unit 224 and the external memory control unit 151, and without applying an excessive load related to data analysis to the microcomputer 154. Can be suppressed. When the section length is relatively long, a detailed log of the corresponding section is generated as in the conventional case. This facilitates analysis of the section data by the microcomputer 154.

  Further, the calculation of the section length can be performed very easily as described above. Therefore, the section length calculation unit 161 and the log size management unit 171 can be relatively easily configured, and the circuit scale of the packet separation device 111 according to the present embodiment is significantly increased as compared with the conventional one. There is nothing. One log is generated for one section. Therefore, by counting the number of sections, the amount of log data generated can be controlled more accurately. For example, if there is only one relatively short section length section between relatively long section length sections, the number of sections in the TSP does not increase rapidly. Therefore, in such a case, a normal mode log, that is, a detailed log is generated.

Note that the compression mode designation criteria by the compression mode designation unit 632 described in the present embodiment are merely examples, and the present invention is not limited to this.
(Third embodiment)
FIG. 7 is a configuration diagram of the digital broadcast receiver 103 including the packet separation device 113 according to the third embodiment of the present invention. The packet separator 113 includes a PID filter 120, a section filter 121, a non-section processing unit 122, a log generation unit 123, a section determination unit 125, a section number calculation unit 163 as a section characteristic extraction unit, a log size management unit 173, a FIFO 180, Latches 181 and 182 and a data transfer unit 224 are provided. Among these, the PID filter 120, the section filter 121, the non-section processing unit 122, the section determination unit 125, and the data transfer unit 224 are the same as those illustrated in FIG.

  The TSP output from the PID filter 120 is input to the FIFO 180 and the section determination unit 125. The FIFO 180 delays the input TSP and outputs it. The section filter 121 and the non-section processing unit 122 receive the TSP output from the FIFO 180, and output section data and non-section data, respectively. The section determination unit 125 outputs a section flag indicating whether or not the input TSP includes a section. The latch 181 delays and outputs the section flag by the same time as the TSP delay by the FIFO 180. Similarly, the latch 181 delays and outputs the PID filter information output from the PID filter 120 by the same time as the TSP delay by the FIFO 180.

  FIG. 8 shows an internal configuration of the section number calculation unit 163 and the log size management unit 173 shown in FIG. The section number calculation unit 163 is considerably simplified than that according to the second embodiment. That is, the section number calculation unit 163 according to the present embodiment includes only a section counter 621 that is reset at the beginning of the TSP.

  The log size management unit 173 determines the compression mode according to the input number of sections. Specifically, the log size management unit 173 includes a first section determination unit 613 and a compression mode designation unit 633. The first section determination unit 613 is as described in the first embodiment. Unlike the one according to the second embodiment, the log size management unit 173 uses the PID match flag indicating that the PID matches and passes through the PID filter 120 in the PID filter information, and sets the compression mode. decide. The compression mode designating unit 633 selects (1) “the PID of the next TSP does not match” or “the next TSP is non-section data” or “the current TSP is non-section data” or “the input section. If the data is related to the first section in the current TSP or “the number of sections is less than 10”, “normal mode” is selected, and (2) “the PID of the next TSP matches” and “ When “the next TSP is section data” and “the input section data is related to the second and subsequent sections in the current TSP” and “the number of sections is 10 or more and less than 20” Compression mode 1 ”and (3)“ PID of next TSP matches ”and“ next TSP is section data ”and“ input section data ” Now those according to the second or subsequent sections in TSP "and" the number of sections is 20 or more, "a" compression mode 2 "in the case, respectively specify.

  The generation of the compression mode log by the log generation unit 123 and the format of the compression log are as described in the first embodiment.

  As described above, according to the present embodiment, the log size is controlled using the information of the next TSP by the PID filter information output from the PID filter 120 and the section flag output from the section determination unit 125. That is, even when the number of sections is relatively large, the next packet (TSP) is discarded by the PID filter 120, that is, when the PIDs do not match or when the TSP selected by the PID filter 120 is a section. If it is not data, a detailed log is generated for the current TSP section. Thereby, the load concerning the analysis of the section data by the microcomputer 154 can be reduced. Although the processing time of the section is temporarily extended by delaying the output of the TSP by the FIFO 180, when the TSP is discarded by the PID filter 120, the section number calculation unit 163 and the log size management unit 173 Since the next TSP section is not input, the process related to log generation is not delayed. Further, the access amount to the external memory 152 does not increase greatly when viewed in the time of two TSPs. In other words, it is only necessary to compress the log and suppress an increase in the amount of access to the external memory 152 only when TSPs including section data are continuous. Thus, a detailed log is created as compared with the second embodiment. More cases to do.

It should be noted that the compression mode designation criteria by the compression mode designation unit 633 described in the present embodiment are merely examples, and the present invention is not limited to this.
(Fourth embodiment)
FIG. 9 shows the configuration of a digital broadcast receiver 104 provided with a packet separator 114 according to the fourth embodiment of the present invention. The packet separator 114 includes a PID filter 120, a section filter 121, a non-section processing unit 122, a log generation unit 123, a section determination unit 125, a section input speed calculation unit 164 as a section characteristic extraction unit, a log size management unit 174, and data A transfer unit 224 is provided. Among these, the PID filter 120, the section filter 121, the non-section processing unit 122, the section determination unit 125, and the data transfer unit 224 are the same as those illustrated in FIG.

  The point at which section data and the like are output from the section filter 121 and the point at which non-section data and the like are output from the non-section processing unit 122 are as described in the first embodiment.

  The section input speed calculation unit 164 receives the section data output from the section filter 121 and calculates the input speed. Here, since one or less logs are generated for one section data, the section is set at a time interval shorter than the input time of the section data or at a time interval that is too long (for example, 1 second). Measuring input speed is meaningless. The normal data rate of digital broadcasting is several tens of Mbps, which is several tens of microseconds per 188 bytes of TSP. Therefore, it is appropriate to measure the section input speed at a time interval of several μ to 100 μsec.

  FIG. 10 shows an internal configuration diagram of the section input speed calculation unit 164 and the log size management unit 174 shown in FIG. The section input speed calculation unit 164 includes two types of section data byte number counters (hereinafter simply referred to as “counters”) 650 and 651. Among these, the counter 651 measures the section data input speed at a relatively short time interval (for example, about several microseconds). On the other hand, the counter 650 measures the section data input speed at a relatively long time interval (for example, about several tens of μ to 100 μs). The timer 652 stores the value of the counter 650 in the latch 653 at regular intervals and resets the counter 650. When the counter 650 is reset, it starts counting again. Similarly, the timer 654 stores the value of the counter 651 in the latch 655 at regular intervals and resets the counter 651. The counter 651 starts counting again when it is reset.

  The relatively short section data input speed output from the latch 655 and the relatively long section data input speed output from the latch 652 are weighted by the multipliers 656 and 657 and then added. The sum is obtained by the device 658. The summed value is output to the log size management unit 174 as the section input speed.

  The log size management unit 174 determines the compression mode according to the input section input speed. Specifically, the log size management unit 174 includes a first section determination unit 613 and a compression mode designation unit 634. The first section determination unit 613 is as described in the first embodiment. The compression mode designating unit 634 (1) “TSP is non-section data” or “input section data is related to the first section in TSP” or “section input speed is 4 MB / second” In the case of “normal mode”, (2) “the input section data relates to the second and subsequent sections in the TSP” and “the section input speed is 4 MB / second or more and less than 6 MB / second” “Compression mode 1”, (3) “input section data is related to the second and subsequent sections in the TSP” and “section input speed is 6 MB / second or more and less than 8 MB / second” "Compression mode 2" and (4) "the input section data relates to the second and subsequent sections in the TSP" and " Deployment input speed is in the case in which "8MB / sec or more is to specify each of the" compression mode 3 ".

  The generation of the compression mode log by the log generation unit 123 and the format of the compression log are as described in the first embodiment. Here, the log in the compression mode 3 will be described. FIG. 11 shows the format of the section log in the compression mode 3. The compression mode 3 log is a compression of N normal mode logs into one. The “number of sections” in the compression mode 3 log indicates how many of the normal mode logs are compressed.

  As described above, according to the present embodiment, when the section input speed is relatively high, the log corresponding to the section is generated in a compressed format. That is, the amount of log data per section data is suppressed. This increases the access to the external memory 152 without increasing the capacity and quantity of the buffers in the data transfer unit 224 and the external memory control unit 151, and without applying an excessive load related to data analysis to the microcomputer 154. Can be suppressed. When the section input speed is relatively low, a detailed log of the corresponding section is generated as in the conventional case. This facilitates analysis of the section data by the microcomputer 154.

  Even when the number of sections included in the TSP is relatively small, when the input speed of the TS, that is, the data rate is relatively fast, the section input speed is relatively fast. On the other hand, even when the number of sections included in the TSP is relatively large, when the input speed of the TS is relatively slow, the section input speed is relatively slow. Therefore, it is possible to generate a compressed log in consideration of the TS input speed.

  Further, by combining the section input speeds in a relatively short time and a long time, the increase in the access amount to the external memory 152 and the number of cases where a detailed log is generated are increased, that is, the load on the microcomputer 154 is reduced. Mitigation can be made better.

  It is not essential to combine the section input speeds for a relatively short time and a long time. However, for example, when the section input speed when viewed in a relatively long time is not so fast but temporarily increased in a short time, if only the measurement result in a relatively short time is used, the compression will be Many logs are generated. In such a case, it is expected that detailed logs will be generated more frequently by taking into account the section input speed in a relatively long time. On the other hand, when the section input speed is not so fast when viewed in a relatively long time, but it is continuously increased in a short time, using only the measurement results for a relatively long time will result in a detailed log. May be generated, the amount of access to the external memory 152 will increase excessively, and data may be lost. In such a case, an excessive access amount to the external memory 152 can be suppressed by taking into account the section input speed in a relatively short time.

Note that the compression mode designation criteria by the compression mode designation unit 634 described in the present embodiment are merely examples, and the present invention is not limited to this.
(Fifth embodiment)
FIG. 12 shows the configuration of a digital broadcast receiver 105 provided with a packet separator 115 according to the fifth embodiment of the present invention. The packet separation device 115 includes a PID filter 120, a section filter 121, a non-section processing unit 122, a log generation unit 123, a section determination unit 125, a short section number calculation unit 165, a log size management unit 175, and a data transfer unit 225. Yes. Among these, the PID filter 120, the section filter 121, the non-section processing unit 122, and the section determination unit 125 are the same as those shown in FIG.

  The point at which section data and the like are output from the section filter 121 and the point at which non-section data and the like are output from the non-section processing unit 122 are as described in the first embodiment.

  FIG. 13 shows an internal configuration of the data transfer unit 225 shown in FIG. The data transfer unit 225 receives the section data output from the section filter 120, the log data output from the log generation unit 123, and the non-section data output from the non-section processing unit 122, and buffers 501, 502, and 503. Temporarily store each data. The selector 504 selects the data stored in the buffers 501 to 503 based on information from the arbitration circuit 505 and the address management unit 506, and outputs the selected data and the storage destination address to the external memory control unit 151. . The arbitration circuit 505 receives buffer empty information from the external memory control unit 151, and controls data output from the selector based on the information. The address management unit 506 outputs a pointer address included in the log to the log generation unit 123 as address information. Further, the buffer 502 outputs buffer availability information to the log size management unit 175. In this embodiment, the number of empty bytes in the buffer 502 is used as the buffer availability information, but the present invention is not limited to this.

  FIG. 14 shows an internal configuration of the short section number calculation unit 165 and log size management unit 175 shown in FIG. The TSP head detector 660 in the short section number calculator 165 detects the head of the TSP. The latch 661 latches bits 3 to 0 of the second byte of TSP and bits 7 to 0 of the third byte. The adder 662 adds “3” to the latched bit to generate a section length. The short section counter 663 counts the number of sections whose section length is equal to or less than a predetermined value, and outputs the number as the number of short sections. The short section counter 663 resets the counter value when the TSP head detection unit 660 detects the head of the TSP. In the present embodiment, “8” is used as the predetermined value, but the present invention is not limited to this.

  The log size management unit 175 inputs the number of short sections from the short section number calculation unit 165 and the buffer availability information from the data transfer unit 225, and determines the log compression mode based on these information. Specifically, the log size management unit 175 includes a first section determination unit 613 and a compression mode designation unit 635. The first section determination unit 613 is as described in the first embodiment. The compression mode designating unit 635 (1) “TSP is non-section data” or “input section data relates to the first section in the TSP” or “the number of empty bytes is 64 or more” “2” indicates that “the input section data is related to the second and subsequent sections in the TSP” and “the number of empty bytes is 32 or more and less than 64”. 1 ”(3)“ The input section data relates to the second and subsequent sections in the TSP ”and“ the number of empty bytes is less than 32 ”and“ the number of short sections is less than 10 ” "Compression mode 2", and (4) "the input section data is related to the second and subsequent sections in the TSP" and "empty bytes" There is ", and less than 32 when" is the number of short sections 10 or more "respectively specify the" compressed mode 3 ".

  The generation of the compression mode log by the log generation unit 123 and the format of the compression log are as described in the fourth embodiment.

  As described above, according to the present embodiment, when the free space of the buffer 502 for temporarily storing the section log in the data transfer unit 225 is relatively small, the compressed log is generated, so that the data is less likely to overflow. On the other hand, when the free space is relatively large, a detailed log is generated as in the conventional case, so that the load on the microcomputer 154 can be reduced. Thus, by changing the size of the log to be generated according to the free space of the buffer, without increasing the capacity and quantity of the buffer, and without applying an excessive load related to data analysis to the microcomputer 154, An increase in access to the external memory 152 can be suppressed. Further, the capacity of the buffer can be reduced, and the circuit scale of the buffer can be reduced. Further, by taking into account the number of short sections that cause the log size to increase, finer control of the log size is possible.

  Note that the compression mode designation criterion by the compression mode designation unit 635 described in the present embodiment is merely an example, and the present invention is not limited to this.

In this embodiment, the compression mode is determined based on the number of short sections and the buffer availability information. However, the compression mode may be determined based only on the buffer availability information. In this case, the short section number calculation unit 165 may be omitted in particular.
(Sixth embodiment)
FIG. 15 shows a configuration of a digital broadcast receiver including a packet separation device according to the sixth embodiment of the present invention. The digital broadcast receiver 106 includes an external memory control unit 251 having a configuration different from that of the external memory control unit 151 according to the first to fifth embodiments. The packet separator 116 also includes a PID filter 120, a section filter 121, a non-section processing unit 122, a log generation unit 123, a section determination unit 125, a short section number calculation unit 165, a log size management unit 176, and a data transfer unit 224. I have. Among these, the PID filter 120, the section filter 121, the non-section processing unit 122, the section determination unit 125, and the data transfer unit 224 are the same as those illustrated in FIG. The short section number calculation unit 165 is the same as that shown in FIG.

  The point at which section data and the like are output from the section filter 121 and the point at which non-section data and the like are output from the non-section processing unit 122 are as described in the first embodiment.

  FIG. 16 shows an internal configuration of the external memory control unit 251 shown in FIG. The external memory control unit 251 inputs and outputs data to and from the microcomputer 154, the AV decoder 153, the data transfer unit 224, and the external memory 152. Write buffers 511, 512, and 513 temporarily store data input from microcomputer 154, AV decoder 153, and data transfer unit 224, and addresses corresponding thereto, respectively. The arbitration circuit 514 arbitrates access requests from the write buffers 511 to 513. In the case of data writing, the selector 515 selects any one of the write buffers 511 to 513 in accordance with an instruction from the arbitration circuit 514 and outputs the address and data stored in the selected write buffer to the external memory 152. In the case of data reading, the read buffers 520 and 521 read data from the external memory 152 and temporarily store it, and output them to the microcomputer 154 and the AV decoder 153.

  The external memory control unit 251 further includes an access congestion information management unit (hereinafter referred to as “access management unit”) 525 that measures the degree of access to the external memory 152. FIG. 17 shows an internal configuration of the access management unit 525. The OR gate 670 outputs the logical sum of the access request from the microcomputer 154, the access request from the packet separator 116, and the access request from the AV decoder 153. The counter 671 is a 7-bit counter that inputs the output from the OR gate 670 as a count enable signal. The counter 671 counts up in synchronization with the clock from the timer 680 when the output from the OR gate 670 is “1”. The timer 680 outputs a counter reset signal and a latch enable signal every predetermined time, for example, every 100 clocks.

  When the counter 671 receives the counter reset signal, the counter 671 outputs the counter value at that time and resets the counter value. The latch 672 latches the output of the counter 671. The output of latch 672 is shifted to latch 673. That is, the latch 673 latches the output of the counter 671 one clock before. Then, the output of the latch 673 is shifted to the latch 674. That is, the latch 674 latches the output of the counter 671 two clocks before. When receiving a latch enable signal from the timer 680, the latches 672 to 674 latch the input value.

  Multipliers 675, 676, and 677 weight the values of latches 672, 673, and 674, respectively. Specifically, the multiplier 675 multiplies the value of the latch 672 by “0.8”. The multiplier 676 multiplies the value of the latch 673 by “0.15”. Then, the multiplier 674 multiplies the value of the latch 674 by “0.05”. Adder 678 sums the output values of multipliers 675-677 and outputs 7-bit data. As a result, when the output value of the adder 678 is relatively small, it indicates that the access is not so crowded, and conversely, when the output value is relatively large, it indicates that the access is crowded.

  On the other hand, the multiplier 681 performs weighting on the empty information in the write buffer 512. In this embodiment, the number of empty bytes is used as the empty information of the write buffer 512, but this does not limit the present invention. Specifically, the multiplier 681 multiplies “0.8” by the number of empty bytes in the write buffer 512. The adder 679 subtracts the value of the multiplier 681 from the value of the adder 678 and outputs the calculation result as access congestion information. The reason why the value of the multiplier 681 is subtracted from the value of the adder 678 is that even if access is relatively crowded, it is considered that there is little possibility of data overflow when the number of empty bytes in the write buffer 512 is large. is there.

  FIG. 18 shows an internal configuration of the log size management unit shown in FIG. The log size management unit 176 inputs the number of short sections from the short section number calculation unit 165 and the access congestion information from the external memory control unit 251 and determines the log compression mode based on these information. Specifically, the log size management unit 176 includes a first section determination unit 613 and a compression mode designation unit 636. The first section determination unit 613 is as described in the first embodiment. The compression mode designating unit 636 (1) “TSP is non-section data” or “the input section data is related to the first section in the TSP” or “access congestion degree information is less than 70”. “Normal mode” is selected in the case (2) “Input section data relates to the second and subsequent sections in the TSP” and “Access congestion status information is 70 or more and less than 90” Compression mode 1 ”(3)“ The input section data is related to the second and subsequent sections in the TSP ”and“ the access congestion degree information is 90 or more ”and“ the number of short sections is less than 10. If “Yes”, select “Compression mode 2”, and (4) “Input section data is related to the second and subsequent sections in the TSP. There "and" access congestion information of 90 or more "and" number of short sections when in a "10 or more respectively specify the" compressed mode 3 ".

  The generation of the compression mode log by the log generation unit 123 and the format of the compression log are as described in the fourth embodiment.

  As described above, according to the present embodiment, when the access to the external memory 152 is relatively congested, a compressed log is generated, so that data is less likely to overflow. On the other hand, when access is relatively small, a detailed log is generated as in the conventional case, and the load on the microcomputer 154 can be reduced. In this way, by changing the size of the log to be generated in accordance with the degree of congestion of access to the external memory 152, an excessive load related to data analysis is added to the microcomputer 154 without increasing the capacity and quantity of the buffer. Without increasing the number of accesses to the external memory 152. Further, the capacity of the buffer can be reduced, and the circuit scale of the buffer can be reduced. Further, by taking into account the number of short sections that cause the log size to increase, finer control of the log size is possible.

  Note that the compression mode designation criteria by the compression mode designation unit 636 described in the present embodiment are merely examples, and the present invention is not limited to this.

  In this embodiment, the compression mode is determined based on the number of short sections and the access congestion information. However, the compression mode may be determined based only on the access congestion information. In this case, the short section number calculation unit 165 may be omitted in particular.

  As described above, the packet separation apparatus according to the present invention is suitable for use as a means for extracting various data such as video and audio data electronic program guides from a given transport stream, particularly in a digital broadcast receiver. .

1 is a configuration diagram of a digital broadcast receiver including a packet separation device according to a first embodiment of the present invention. FIG. 2 is an internal configuration diagram of a section length calculation unit and a log size management unit illustrated in FIG. 1. It is a figure which shows the format of the log for sections by compression mode 1. It is a figure which shows the format of the log for sections by compression mode 2. FIG. It is a block diagram of the digital broadcast receiver provided with the packet separation apparatus which concerns on the 2nd Embodiment of this invention. FIG. 6 is an internal configuration diagram of a section number calculation unit and a log size management unit illustrated in FIG. 5. It is a block diagram of the digital broadcast receiver provided with the packet separation apparatus which concerns on the 3rd Embodiment of this invention. FIG. 8 is an internal configuration diagram of a section number calculation unit and a log size management unit illustrated in FIG. 7. It is a block diagram of the digital broadcast receiver provided with the packet separation apparatus which concerns on the 4th Embodiment of this invention. FIG. 10 is an internal configuration diagram of a section input speed calculation unit and a log size management unit illustrated in FIG. 9. It is a figure which shows the format of the log for sections by compression mode 3. It is a block diagram of the digital broadcast receiver provided with the packet separation apparatus which concerns on the 5th Embodiment of this invention. It is an internal block diagram of the data transfer part shown in FIG. FIG. 13 is an internal configuration diagram of a short section number calculation unit and a log size management unit illustrated in FIG. 12. It is a block diagram of the digital broadcast receiver provided with the packet separation apparatus which concerns on the 6th Embodiment of this invention. FIG. 16 is an internal configuration diagram of an external memory control unit shown in FIG. 15. It is an internal block diagram of the access congestion condition information management part shown in FIG. FIG. 16 is an internal configuration diagram of a log size management unit shown in FIG. 15. It is a figure which shows the format of a transport stream. It is a block diagram of the digital broadcast receiver provided with the conventional packet separation apparatus. It is an internal block diagram of a PID filter. It is an internal block diagram of a section filter. It is a figure which shows the format of various logs. It is a figure which shows the storage area of various data in an external memory.

Explanation of symbols

111, 112, 113, 114, 115, 116 Packet separator 161 Section length calculation unit (section characteristic extraction unit)
162,163 Section number calculation section (section characteristic extraction section)
164 Section input speed calculation section (section characteristic extraction section)
171, 172, 173, 174, 175, 176 Log size management unit 123 Log generation unit 165 Short section number calculation unit

Claims (9)

A packet separation device for extracting a packet from a given transport stream and extracting section or non-section data included in the packet;
A section characteristic extraction unit for extracting characteristics of the section;
A log size management unit that sets the size of a section log according to the section characteristics extracted by the section characteristic extraction unit;
A packet separator comprising: a log generation unit configured to generate the section log having a size set by the log size management unit. The packet separator according to claim 1, wherein
The section characteristic extraction unit calculates a section length as a characteristic of the section,
The log size management unit compresses the size of the section log when determining that the section length calculated by the section characteristic extraction unit is relatively short. The packet separator according to claim 1, wherein
The section characteristic extraction unit calculates the number of sections per predetermined number of packets as the characteristic of the section,
The packet size separation unit compresses the size of the section log when the log size management unit determines that the number of sections calculated by the section characteristic extraction unit is relatively large. The packet separator according to claim 3, wherein
The log size management unit determines that the identifiers of the consecutive packets in the transport stream match each other, and the number of sections calculated by the section characteristic extraction unit for the preceding packet is relatively large And a packet separator for compressing the size of the section log. The packet separator according to claim 1, wherein
The section characteristic extraction unit calculates a data input speed of the section as the characteristic of the section.
The log size management unit compresses the size of the section log when it is determined that the data input speed calculated by the section characteristic extraction unit is relatively high. A packet separation device that extracts packets from a given transport stream and extracts section or non-section data from the extracted packets.
A log size management unit that sets the size of the section log based on information related to the output control of the section log;
A packet separator comprising: a log generation unit configured to generate the section log having a size set by the log size management unit. The packet separator according to claim 6, wherein
The log size management unit obtains information on the availability of a buffer for temporarily storing the section log as information relating to output control of the section log, and the buffer availability is relatively determined from the information. When it is determined that there is little, the size of the section log is compressed. The packet separator according to claim 6, wherein
The log size management unit obtains information related to the degree of access to the external memory that is the storage destination of the section log as information related to the output control of the section log, and accesses the external memory from the information The packet separating apparatus compresses the size of the section log when it is determined that the section log is relatively congested. The packet separation device according to claim 7 or 8,
A short section number calculation unit that calculates the number of sections whose section length is a predetermined value or less per predetermined number of packets,
The packet size management unit compresses the size of the section log when determining that the number of short sections calculated by the short section number calculation unit is relatively large. .

Images