WO2004084062A1 - パターン検出装置 - Google Patents
パターン検出装置 Download PDFInfo
- Publication number
- WO2004084062A1 WO2004084062A1 PCT/JP2004/001894 JP2004001894W WO2004084062A1 WO 2004084062 A1 WO2004084062 A1 WO 2004084062A1 JP 2004001894 W JP2004001894 W JP 2004001894W WO 2004084062 A1 WO2004084062 A1 WO 2004084062A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- pattern
- detection
- signal
- entry
- data
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F7/00—Methods or arrangements for processing data by operating upon the order or content of the data handled
- G06F7/02—Comparing digital values
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L7/00—Arrangements for synchronising receiver with transmitter
- H04L7/04—Speed or phase control by synchronisation signals
- H04L7/08—Speed or phase control by synchronisation signals the synchronisation signals recurring cyclically
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2207/00—Indexing scheme relating to methods or arrangements for processing data by operating upon the order or content of the data handled
- G06F2207/02—Indexing scheme relating to groups G06F7/02 - G06F7/026
- G06F2207/025—String search, i.e. pattern matching, e.g. find identical word or best match in a string
Definitions
- the present invention relates to a pattern detection device.
- the present invention relates to a pattern detection device (pattern detection device) that detects whether data having a predetermined pattern exists in data stored in a memory or data input from the outside. Evening inspection device).
- JP-A-5-76056, JP-A-7-37332, and JP-A-9-32 ⁇ 726 disclose predetermined data to be searched.
- a pattern detection circuit pattern search device
- searched a pattern to be detected
- a matching method of performing pattern matching for determining whether or not the data stored in the memory space matches the search target pattern using C is disclosed.
- data is sequentially read while moving a read address to a memory space in which data to be detected is stored, and it is compared whether or not the data matches the entry pattern.
- the read address at the time of the match is stored in the storage means as the detection position (or match position).
- This match position indicates in which position in the memory space data having the same arrangement as the entry tree exists.
- the desired data processing is performed using the match position as a mark.
- a typical pattern detection circuit includes a comparison circuit and a detection position registration circuit.
- system layer multiple types of system identification codes (stream IDs) are specified, and as the video layer, multiple start codes (start codes) are specified.
- pattern detection is performed using the stream ID output from the apple as an entry pattern.
- pattern detection is performed using a start code to be detected as an entry pattern.
- the bitstream is sequentially read from the memory, and pattern detection is performed.
- pattern detection is performed using a one-byte code, for example, a stream ID of a system layer or a start code of a video layer as an entry pattern. This requires a lot of registry space or memory to store match positions, and storing the entry pattern itself requires a great deal of hardware resources.
- the conventional pattern detection circuit by performing sequential processing each time a hit is made, hardware resources for holding a match position can be reduced, but pattern detection and conditional branching after a hit can be performed. To reduce this overhead, it is necessary to detect patterns in a certain unit. However, in order to achieve this, it is necessary to have a mechanism to perform pattern detection by integrating the memory space to be detected in the pattern detection circuit to a certain extent.However, since this is not usually provided, the overhead of pattern detection is reduced. There is a disadvantage that cannot be realized. . In the comparison circuit, the stored data is sequentially read from the memory space to be searched, and it is compared whether the read data matches the entry pattern.
- a detection match signal (hit signal) is output from the comparison circuit.
- It consists of a search position storage circuit, a register, a memory, and the like.
- a search position that is, a read address at that time or a count value from a predetermined reference position is obtained.
- the search result matches the entry pattern (search target pattern) and the hit signal is output. Since the detected position must be stored in the memory, it is necessary to prepare a register or a large-capacity memory with a sufficient visit width.
- a register having a bit width equal to the number of hit signals or a memory having a capacity equal to the number of hit signals is required.
- the number of registered matching positions when the m pieces, total 2 n xm bits wide registers or total 2 n xm address component having a memory is required.
- n and m increase, so does the amount of hardware resources such as registers or memory required to store match positions.
- MPEG which is one of the standards for compressing and encoding moving images and audio signals
- pattern detection in a compressed and encoded data sequence (hereinafter, referred to as a bit stream or a data stream) is performed by a system layer and a system layer. It is performed in each video layer.
- the bit stream of the MPEG includes, for example, as shown in FIG. 3A and FIGS. 4 to 7, a 3-byte identification code [0x00001], followed by a 1-byte data.
- a code is provided for each system layer or video layer.
- the present invention has been made in view of the above circumstances, and an object of the present invention is to detect a pattern having a desired pattern from a memory space to be searched by pattern matching.
- An object of the present invention is to provide a power supply circuit that can save resources for storing a position and reduce the amount of processing overhead after detection of the power supply.
- a pattern detection circuit comprises: a detection pattern storage means for storing an entry pattern to be detected; and a flag storage means for storing a flag signal corresponding to each of the entry patterns. And comparing means for comparing the input data with the entry pattern stored in the detection pattern storage means. When the input data and the entry pattern match as a result of the comparison by the comparison means, And a detection position storage means for storing the detection position according to a flag signal corresponding to the entry pattern stored in the flag storage means.
- the flag signal corresponding to the entry pattern stores a detection position.
- the pattern detection window signal is set to the active state and the flag signal corresponding to the entry pattern indicates that the detection position is not stored, the pattern detection window signal is set to the non-active state. It further has a position storage control means.
- the detection position storage means stores a detection position of the entry pattern when the pattern detection window signal is in an active state, and sets the pattern detection window signal to an inactive state. Do not memorize the detection position of the entry pattern
- the input data is image data and audio data compressed and encoded according to the MPEG standard
- the entry pattern is It is set in accordance with the identification data indicating the beginning of the packet included in the compression-encoded data.
- the flag signal is set for each entry pattern to be searched.
- the input data is compared with the entry pattern by the comparing means, and as a result of the comparison, if the input data matches the entry pattern, whether the flag signal corresponding to the entry pattern is valid,
- the flag signal indicates the storage of the detection position
- the entry pattern and the detection position are stored when indicating the storage of the detection position, and conversely, the detection is performed when the flag signal indicates that the detection position is not stored. No position is stored. For this reason, it is possible to specify in advance the detection positions that need to be registered, and it is possible to efficiently use the resources of the storage unit that stores the detection positions.
- FIG. 1 is a configuration diagram showing one embodiment of a pattern detection circuit according to the present invention.
- FIG. 2 is a configuration diagram showing a configuration of a pattern detection non-detection control unit in the pattern detection circuit of the present embodiment.
- FIGS. 3A to 3C are diagrams illustrating an example of a data stream that is compression-encoded according to the MPEG standard.
- FIG. 4 is a diagram showing a stream identification number of a system layer in the MPEG standard.
- FIG. 5 is a diagram showing an example of a start code of a video layer according to the MPEG standard.
- FIG. 6 is a diagram showing an example of an entry pattern for detecting a start code of each layer.
- FIG. 7 is a diagram showing an example of each entry pattern and a corresponding flag signal.
- FIG. 1 is a configuration diagram showing one embodiment of a pattern detection device according to the present invention.
- the pattern detection device 100 of the present embodiment includes a plurality of comparison units ⁇ 20 -11, 120—N (N> 0, N is an integer), a pattern detection / non-detection control unit 130 And a detection position registration unit 140.
- a memory / bit stream interface (MEM / BS-I / F: B / S I / F; hereinafter, memory / BS-I / F) 110 is connected to a pattern detection device 100.
- the input data IND and the read pointer RDP read from the storage means are provided.
- the input data IND is, for example, a video signal encoded by MPEG illustrated in FIG. 3A, and the readout point RDP is an address of the video signal.
- FIG. 1 representatively shows only the internal configuration of the first comparison unit 120-1.
- 1st comparison part 120- ;! Has a comparator 122, an entry pattern holding unit 14 that holds a search target pattern or an entry pattern ETP1, and a flag holding unit 126 that holds a flag FLG1.
- the number of comparison sections 1 20— :! to 1 20—N matches the entry patterns ETP 1 to ETPN to be searched.
- the entry pattern ETP1-ETPN and the flags FLG1 to ETPN have a one-to-one correspondence, for example, as in the entry pattern ETP1 and the flag FLG1.
- the entry patterns ETP] to ETPN are patterns to be searched (detected) among the data included in the input data IND.
- a detailed example of the entry patterns ETP 1 to ETPN will be described with reference to FIG. 3A.
- Each of the flags FLG1 to FLGN corresponds to the entry patterns ETP1 to ETPN, for example, as described in detail with reference to FIG.
- the reset control flip-flop 133 in the detection control unit 130 is composed of a set signal SET for a reset signal and a reset signal (or CLR) for a reset signal.
- the comparator 122 includes a pattern detection device; input data I ND input from a memory / BS-I / F 110 external to the I 00, and a second input from the entry line storage unit # 24.
- One-entry pattern ETP1 is compared, and a first hit signal HIT1 is output according to the result of the comparison.
- the comparator 122 outputs, for example, a high-level (1) hit signal HIT] if the input data IND matches the entry pattern ETP1, and outputs a low-level hit signal if they do not match. Outputs the hit signal HIT1 of (0).
- the entry pattern holding unit 124 stores the first entry pattern ETP1 input from outside the pattern detection device 100, reads out the stored first entry pattern ETP1, and outputs it to the comparator 122. I do.
- the flag holding unit 126 holds the second flag signal FLG 1 input from the outside of the pattern detecting device 00.
- the control signal CTL1 like the flag FLG1, consists of a set signal and a clear signal.
- the control signal CTL;! Generates a pattern detection window signal P DW that controls registration of the detected match position according to CTLN.
- FIG. 38 shows an example of the power detection window signal 0 $.
- each of the plurality of comparison units 120-20 to 120 -N includes the comparator 122 and the entry pattern holding unit, as described above. 14 and a flag holding unit 1 26.
- Comparison with P1 to ETPN is performed, and hit signals HIT1 to HITN and control signals CTL1 to CTLN are output based on the comparison result. That is, in the pattern apple display device of the present embodiment] 0 0, a plurality of comparison units; It is possible to detect multiple (N) entry patterns ETP1 to ETPN by using ⁇ 120-N.
- the plurality of flags FLG1 to FLGN specify control information for generating the pattern detection window signal PDW for the corresponding entry patterns ETP1 to ETPN, respectively.
- the power detection / non-detection control unit 130 is configured to control the holding rice according to the hit signals HIT lH ITN and the control signals CTL] to CTLN from the plurality of comparison units 12 0— ⁇ to 120—N. Generates a pull signal HEB, and outputs it to the detection position registration unit 140 that stores (registers) the read pointer RDP output from the memory /: BS-I / F110.
- FIG. 2 shows an example of the configuration of the pattern detection // non-detection control section 130.
- the pattern detection / non-detection control section 130 is composed of OR gates 13 1, 13 22-1, 13 22-2. Reset-set (RS) flip-flop 133 and AND gate 134 have.
- the OR gate 1331 calculates the logical sum (OR) of the plurality of hit signals HIT 1 to HI TN output from the plurality of comparison units 1 20— :! to 120—N, and as a result, If at least one of the plurality of hit signals HIT 1 to HITN has a high-level hit signal, the high-level (1) hit signal HIT is output to the AND gate 134. On the other hand, the OR gate 1331 outputs a low-level (0) hit signal HIT when all of the plurality of hit signals HIT1 to HITN are at low level.
- OR gate; I32-2 is a logical sum of clear signals CLR1 to CLRN included in a plurality of control signals CTL1 to CTLN output from a plurality of comparators 1 20-1 to I20-N If there is a high-level clear signal in at least one of the clear signals CLR1 to CLRN, for example, the high-level (1) is used as the reset signal RST and the reset signal of the RS flip-flop 133 Input to input terminal R. On the other hand, the OR gate I32-2 outputs a low-level (0) reset signal RST when all of the clear signals CLR1 to CLRN are low.
- the RS flip-flop 133 outputs a pattern detection window signal PDW in response to the set signal SET input from the OR gate 132-1 and the reset signal RST input from the OR gate 132-2. That is, when the high-level (1) set signal SET is applied to the set signal input terminal S, the RS flip-flop 133 outputs the high-level (1) pattern detection window signal PDW and outputs the high-level (1 When the reset signal RST is applied to the reset signal input terminal R, the pattern detection window signal PDW is set to low level (0).
- the AND gate 34 calculates the logical product of the hit signal HIT from the OR gate 3) and the pattern detection window signal PDW from the RS flip-flop 133, and both the hit signal HIT and the pattern detection window signal PDW are at high level. Only when, the high-level (1) holding enable signal HEB is output to the detection position registration unit 140.
- a plurality of comparison sections 1 20- ⁇ to 1 2 When a high-level hit signal is output from any of 0 to N, a high-level hit signal HIT is output by the OR gate 13 1.
- an RS flip-flop 133 outputs an active state, for example, a high-level pattern detection window signal PDW. While the pattern detection window signal PDW is held at the high level, the hit signal HI output from the OR gate 13] is output as the hold enable signal HEB via the AND gate 13 4. You.
- the reset signal RST is output by the OR gate 132-2.
- the RS flip-flop 133 outputs an inactive state, for example, a D-level power detection window signal P DW. While the pattern detection window signal PDW is held at a low level, the output of the AND gate 134 is also held at a low level. At this time, the hit signal HI output from the OR gate 13 1 is output. Ding is ignored (disabled).
- the detection position registration unit 140 has a plurality of registers 1 to a register M.
- the detection position registration unit 140 stores the read pointer RDP output from the memory / BS-IZF 110 in accordance with the hold enable signal HEB from the pattern detection / non-detection control unit ⁇ 30, as an entry ⁇ -Register as a detection position where a pattern matching the pattern exists. More specifically, in the detection position registration unit 140, the held enable signal HE B from the pattern detection non-detection control unit 130 is monitored, and the held enable signal H EB is set to the high level in the active state. At one time, the read pointer RDP output from the memory / BS-I / F] 0 is sequentially stored in the register 1 to the register M.
- the plurality of comparators are 120-1 to 120— Multiple hit signals HIT 1 to HI TN from N and multiple control signals CTL 1 to CTLN
- the pattern detection window signal PDW generated in response to the set signals SET 1 to SETN included in the control signals CTL 1 to CTLN and the clear signals CLR 1 to CLRN is at a high level, and the plurality of comparators 120 When any one of the signals HIT1 to HITN is at a high level, it is possible to output a high-level holding enable signal HEB to the inspection position registration unit 140.
- the detection position registration unit 140 stores the read pointer RDP input from the memory / BS-IZF 110 into the register 1 to the register M corresponding to the read pointer RDP. Memorize things. As a result, the registration of the detection position (match position) can be performed to the register 1 to the register M only for the necessary one of the detected entry patterns. In other words, since it is only necessary to prepare the register 1 to the register M for the pattern to be detected, the “dual lithography” of the pattern detection apparatus 100 can be used effectively. o
- the input data IND input from the memory / BS-I / F 10 to the pattern detection apparatus of the present embodiment] 00 is a moving image or audio data compressed and encoded according to a predetermined coding standard, for example, the MPEG standard. It is.
- the compression-encoded data is read out from the memory or other recording medium, and is converted into a bit stream, that is, a continuous data sequence via the memory / BS-; I / F 1] 0.
- the pattern is supplied to the pattern detection apparatus 100 of the present embodiment.
- a read pointer RDP indicating the detection position is also output.
- the read pointer RDP is a pointer (for example, a read address) for reading the data I / D from the memory, or a location at each read starting from a predetermined time. It is a count value that increases in fixed units.
- a visit stream compressed and encoded according to the MPEG standard is divided into packets as shown in Fig. 3A.
- the figure shows an example of moving image and audio data according to the MPEG standard.
- a PES packet (PES—pa) is followed by a packet header (pae k_header) or a system header yst em—header. cket) is located
- the packet header contains a packet y indicating the packet header, a packet code, and attribute information indicating the attribute of the packet header.
- the system header contains attribute information, etc., following the system header code indicating the start of the system header.
- a PES packet includes video data, that is, compression-encoded moving image data, and audio data, that is, compression-encoded audio data.
- video data that is, compression-encoded moving image data
- audio data that is, compression-encoded audio data.
- a video stream start code (Vide 0 str eam st ar t c o d e) or an audio stream start code (au dio str eam st ar tcode) indicating the start of the packet is arranged. ing.
- FIGS. 4 and 5 show an example of the stream identification number (stream ID) of the system layer.
- Figure 5 shows an example of a video layer start code. Note that the stream ID or the start code shown in FIGS. 4 and 5 is arranged following the 3-bit code CO X 00 0001].
- FIG. 6 is a diagram showing an entry pattern for detecting a start code of each layer.
- the entry pattern for detecting the packet start code is 32 bits (4 bytes) of data [0x000001 ba], and is used to detect the system header code.
- the entry stream for detecting the audio stream start code is 4 bits of data [0 x 00 0 0 0 1 e-] or [0 x 0 0 0 0 1 d-].
- the entry pattern for detecting the video stream start code is 4 bytes (000000001e).
- the hyphen "1" is any one of hexadecimal numbers [0 to 9, a to f (a corresponds to 10 in decimal number, f corresponds to 15 in 10 decimal number)] Means 4-bit day and night.
- one entry pattern EPT1 to EPTN to be detected in advance and flag signals FLG1 to FLGN is set.
- the entry pattern is held in the entry pattern holding unit 124, and the flag signal is held in the flag holding unit 126.
- the comparator 122 compares the input data I ND provided from the memory / IZF 110 with the entry pattern. As a result, the input data I ND and the entry pattern are compared.
- the hit signal HIT 1 is output when the values match, and at that time, the flag FLG 1 held by the flag holding unit 126 is output as a control signal 0
- FIG. 7 shows an example of the entry patterns ETP] to ETPN to be detected and the flag signals FLG1 to FLGN corresponding to the respective entry patterns in the pattern detection apparatus 100 of the present embodiment.
- the flag signals FLG1 to FLGN consist of, for example, two bits each, and these two bits control the RS flip-flop 133 as shown in FIG. , Set signal SET and clear signal (or reset signal) CLR.
- the flag signal is held in the flag holding unit 126 shown in FIG. As described above, the flag holding unit 26 is controlled by the control signal CTL according to the flag signals FLG1 to FLGN. 1 to CTLN, that is, generate set signals SET 1 to SETN and clear signals CLR 1 to CLRN.
- the flag signals are (1, 1), (], 0), (0, 1), and (
- the set signal is kept active, for example, at a high level, and the clear signal is kept inactive, that is, kept at a low level.
- the packet start code that is, the flag signal corresponding to the 32-bit entry pattern CO 0000001 aO is set to (1, 0).
- the video stream start code that is, the flag signal corresponding to the entry pattern [0X000001e-] is set to (0, 1).
- the audio stream start code that is, the flag signal corresponding to the entry pattern [0x00001c-] or [0x000001d-] is set to (0,)).
- the flag signal corresponding to the entry pattern [0x000 0 0 1 b—] is set to (0, 0).
- the pattern detection apparatus 100 of the present embodiment detects a data sequence that matches the entry pattern from the input data stream. Further, each detection position, that is, a match position, is registered in a corresponding registry of the detection position registration unit 40 in accordance with a flag signal corresponding to each entry pattern.
- the pattern detection window signal PDW is maintained in the active state, for example, at a high level in the pattern detection / non-detection control unit 130, and accordingly, the pattern detection Z non-detection control unit 130 As a result, the active breakdown holding enable signal HEB is output. For this reason, the match position corresponding to the packet start code is registered in the detection position registration unit 140 at the corresponding registry evening.
- the PES packet is input following the packet header with the first packet in the bucket.
- a video stream start code is placed at the top of the PES packet. That is, the video stream start code at the head of the EPS bucket is detected by the entry stream [0x00001e-].
- the hold enable signal HEW is generated according to the hit signal HIT. B is output. In response, the match position is registered in the detection position registration unit 140. Also, as shown in FIG.
- the pattern detection / non-detection control unit 1 since the flag signal corresponding to the entry pattern [0x00001e-] is set to (0, 1), the pattern detection / non-detection control unit 1 At 30, the pattern detection window signal PDW is switched to the inactive state, for example, to the low level as shown in FIG. For this reason, the duration of the PES packet During the entry period, for example, the hit signal HIT detected according to the user data start code [0 x 00 0 0 0 1 2 Group start code [0 x 00 0 0 0 1 b 8] etc. becomes invalid Therefore, the registration of the multi-position corresponding to these entry patterns is not performed.
- the pattern detection device of the present embodiment 00 detects an entry pattern [0 0 0 0 0 0 1 1 b a ] corresponding to a packet since in accordance with the flag signal corresponding to the pattern (l s 0) Pas evening Ichin detection window signal PDW is set to Akutipu state, the path evening one emission detection / non-detection control unit 1 3 0, the hit signal hIT Accordingly, the holding enable signal HE B is output. For this reason, the match position corresponding to the entry pattern is registered in the detection position registration unit 140 at the registration evening. Then, when the video stream start code at the head of the PES packet is detected, the flag signal corresponding to the entry pattern [0X00001e-] is set to (0, 1).
- the match position is not registered, so it is necessary. Only the correct match positions are registered, and the hardware resources of the detection position registration unit 140, for example, the efficiency of the registration can be improved.
- the pattern detection apparatus 100 of the present embodiment is used, as described above, not only when the PES packet of the video stream is input following the packet code, but also the packet Even when a system header and a PES packet of the audio stream are input following the toque, the pattern detection window signal PDW is correctly controlled by the pattern detection Z non-detection control unit 130, and the necessary hit A match position can be registered for only the signal.
- the packet stream code in the front face of the packet header is an entry pattern [0x0000 0 lba ] Is detected by Then, the detected match position is stored in the detection position registration unit.
- the pattern detection Z non-detection control unit 130 sets the power ring window signal PDW to the active state.
- the system header code at the top of the system header is detected by the entry pattern [0; s 0 0 0 0 0; lt3—].
- the flag signal corresponding to the entry pattern [0 3 ⁇ 4 0 0 0 0 1 b—] is (0, 0).
- the state of the detection window signal PDW does not change and is maintained in the active state.
- the audio stream start code at the beginning of the PES packet is the entry pattern “0x0 0 0 0 0 1c—] or [0x0 0 0 00 1 d—]
- the signal is (0, 1) Therefore, upon detection of the audio stream start code, the pattern detection window signal PDW of the pattern detection / non-detection control unit 130 is switched to the inactive state.
- the match position is registered without outputting the holding enable signal HEB.
- Such L Such L ,.
- the entry pattern to be detected and the corresponding entry pattern are stored in the respective comparison sections by the entry pattern holding section 124 and the flag holding section 126.
- the flag signal is stored, the input data I ND is compared with the entry pattern held by the entry pattern holding unit 124 by the comparator 122, and a hit signal is output when the input pattern matches and the entry pattern is matched.
- Flag holding unit 1 2 6 Signal, and the pattern detection Z non-detection control unit] 30 outputs the holding enable signal HEB only when a required entry pattern is detected in accordance with the hit signal and control signal.
- the entry pattern and the detection position are registered in the detection position registration unit 140.
- the detection position can be registered only for the necessary apples, and the hardware resources such as the register in the detection position registration unit 140 for registering the detection position can be efficiently used.
- various alternative techniques performed by those skilled in the art are not limited to the above-described examples.
- the plurality of registers 1 to M in the detection position registration unit 140 can be replaced with a semiconductor memory device such as a RAM.
- the case where the pattern ETP and the flag FLG are input and held in these entry one pattern holding unit ⁇ 24 and the flag holding unit 126 is illustrated, but if the entry pattern ETP and the flag FLG are fixed, Data can be stored in advance.
- the entry-to-pattern holding section 24 and the flag holding section 126 can be constituted by a semiconductor memory or the like.
- the flag holding unit 126 has a function of outputting the held flag FLG when the first hit signal HIT1 is output from the comparator 122, that is, a gate function. Just do it.
- each of the comparison units 1 2 0—] to ⁇ 2 0— The entry pattern holding unit 124 and the flag holding unit 126 can be realized by one semiconductor memory. Therefore, the configuration example in FIG. 1 is conceptual, and in fact, a plurality of entry patterns stored in a plurality of comparison sections 12 0—1 to 12 0—N.
- the holding section 124 and the plurality of flag holding sections 126 can be constituted by one semiconductor memory or the like.
- pattern detection device of the present invention illustrated in FIG. 1 can be configured as one semiconductor integrated device.
- the pattern detection apparatus of the present invention is applicable only to a video signal or the like. Not done.
- the read pointer RDP described with reference to FIG. 1 corresponds to the read address of the memory.
- the pattern detection device of the present invention only the necessary match positions are registered in accordance with the flag signal set corresponding to each entry pattern (search target pattern). Since the match position can be discarded, the resources of the circuit for registering the match position (detection position), that is, the register or memory for storing the entry pattern and the corresponding match position, must be used efficiently. Can be.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- General Engineering & Computer Science (AREA)
- Computational Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Synchronisation In Digital Transmission Systems (AREA)
- Storage Device Security (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/542,606 US20060050779A1 (en) | 2003-02-21 | 2004-02-19 | Pattern detection device |
EP04712708A EP1596277A4 (en) | 2003-02-21 | 2004-02-19 | SHAPE DETECTOR |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2003044407A JP3807379B2 (ja) | 2003-02-21 | 2003-02-21 | パターン検出回路 |
JP2003-044407 | 2003-02-21 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2004084062A1 true WO2004084062A1 (ja) | 2004-09-30 |
Family
ID=33027113
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2004/001894 WO2004084062A1 (ja) | 2003-02-21 | 2004-02-19 | パターン検出装置 |
Country Status (6)
Country | Link |
---|---|
US (1) | US20060050779A1 (ja) |
EP (1) | EP1596277A4 (ja) |
JP (1) | JP3807379B2 (ja) |
KR (1) | KR20050108343A (ja) |
CN (1) | CN100530075C (ja) |
WO (1) | WO2004084062A1 (ja) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2005093562A1 (ja) * | 2004-03-29 | 2005-10-06 | Kyoto University | データ処理装置、データ処理プログラム、およびデータ処理プログラムを記録した記録媒体 |
CN104516820B (zh) * | 2015-01-16 | 2017-10-27 | 浪潮(北京)电子信息产业有限公司 | 一种独热码检测方法和独热码检测器 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002099407A (ja) * | 2000-09-25 | 2002-04-05 | Toshiba Corp | スタートコード検索回路 |
JP2002142192A (ja) * | 2000-11-01 | 2002-05-17 | Sony Corp | 信号処理装置および方法、ならびに、記録装置および方法 |
JP2002335162A (ja) * | 2001-05-09 | 2002-11-22 | Matsushita Electric Ind Co Ltd | 復号装置及び復号方法 |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR0160668B1 (ko) * | 1994-12-30 | 1999-01-15 | 김광호 | 영상압축 비트스트림의 스타트코드 검출장치 |
JP3450553B2 (ja) * | 1995-10-31 | 2003-09-29 | 東芝マイクロエレクトロニクス株式会社 | 可変長符号復号装置 |
JPH10255485A (ja) * | 1997-03-10 | 1998-09-25 | Kawasaki Steel Corp | 連想メモリおよびネットワークフレーム中継器 |
US6094443A (en) * | 1997-10-30 | 2000-07-25 | Advanced Micro Devices, Inc. | Apparatus and method for detecting a prescribed pattern in a data stream by selectively skipping groups of nonrelevant data bytes |
DE69838672T2 (de) * | 1997-12-26 | 2008-03-06 | Matsushita Electric Industrial Co., Ltd., Kadoma | Vorrichtung zur reproduktion eines kodierten signals |
JP3569800B2 (ja) * | 1998-12-24 | 2004-09-29 | カシオ計算機株式会社 | 画像処理装置及び画像処理方法 |
-
2003
- 2003-02-21 JP JP2003044407A patent/JP3807379B2/ja not_active Expired - Fee Related
-
2004
- 2004-02-19 WO PCT/JP2004/001894 patent/WO2004084062A1/ja active Application Filing
- 2004-02-19 KR KR1020057013374A patent/KR20050108343A/ko not_active Application Discontinuation
- 2004-02-19 US US10/542,606 patent/US20060050779A1/en not_active Abandoned
- 2004-02-19 EP EP04712708A patent/EP1596277A4/en not_active Withdrawn
- 2004-02-19 CN CNB2004800043228A patent/CN100530075C/zh not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002099407A (ja) * | 2000-09-25 | 2002-04-05 | Toshiba Corp | スタートコード検索回路 |
JP2002142192A (ja) * | 2000-11-01 | 2002-05-17 | Sony Corp | 信号処理装置および方法、ならびに、記録装置および方法 |
JP2002335162A (ja) * | 2001-05-09 | 2002-11-22 | Matsushita Electric Ind Co Ltd | 復号装置及び復号方法 |
Non-Patent Citations (1)
Title |
---|
See also references of EP1596277A4 * |
Also Published As
Publication number | Publication date |
---|---|
EP1596277A4 (en) | 2008-07-02 |
CN100530075C (zh) | 2009-08-19 |
JP3807379B2 (ja) | 2006-08-09 |
CN1751289A (zh) | 2006-03-22 |
EP1596277A1 (en) | 2005-11-16 |
KR20050108343A (ko) | 2005-11-16 |
JP2004266370A (ja) | 2004-09-24 |
US20060050779A1 (en) | 2006-03-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR0152916B1 (ko) | 데이타 동기화장치 및 방법 | |
US6996070B2 (en) | TCP/IP offload device with reduced sequential processing | |
KR20190085555A (ko) | 연산 캐시 | |
JP2004507858A (ja) | 圧縮アルゴリズムのハードウエアにおける実現 | |
US20070294472A1 (en) | Integrated circuit with associated memory function | |
WO2004084062A1 (ja) | パターン検出装置 | |
CN102207848A (zh) | 指令取回装置、处理器和程序计数器加法控制方法 | |
US6661792B1 (en) | Apparatus for processing data packet of ethernet switch system and method thereof | |
US20060101484A1 (en) | Decoding apparatus and encoding apparatus | |
US10637498B1 (en) | Accelerated compression method and accelerated compression apparatus | |
US20200244281A1 (en) | Accelerated compression method and accelerated compression apparatus | |
US7191295B2 (en) | Sensing word groups in a memory | |
US6792006B1 (en) | Data multiplexing device multiplexing transport stream generated from encoded data | |
US6414726B1 (en) | Device for identifying packets of digital data and a receiver for digital television signals equipped with such a device | |
JP2005512192A5 (ja) | ||
US6259659B1 (en) | Method and apparatus for configuring a sector map for an optical storage device | |
US20020025003A1 (en) | Apparatus for processing a bit stream | |
JP2001036491A (ja) | データストリームをフィルタリングする装置及び方法 | |
JP2008182550A (ja) | パケットバッファfifoメモリ装置 | |
US6429794B1 (en) | Format converter | |
JP2000059226A (ja) | 最小マッチ長が3のプリマッチストリングマッチアレイ | |
JPH09180468A (ja) | 連想記憶装置 | |
US7324421B1 (en) | Method and apparatus for data bit align | |
KR100206937B1 (ko) | 데이타 동기화 장치 및 방법 | |
JP2001053709A (ja) | データ一致検出装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
ENP | Entry into the national phase |
Ref document number: 2006050779 Country of ref document: US Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10542606 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020057013374 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2004712708 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 20048043228 Country of ref document: CN |
|
WWP | Wipo information: published in national office |
Ref document number: 2004712708 Country of ref document: EP Ref document number: 1020057013374 Country of ref document: KR |
|
WWP | Wipo information: published in national office |
Ref document number: 10542606 Country of ref document: US |