CN101547353B - Decoding accelerator for variable length codes - Google Patents

Decoding accelerator for variable length codes Download PDF

Info

Publication number
CN101547353B
CN101547353B CN 200810089812 CN200810089812A CN101547353B CN 101547353 B CN101547353 B CN 101547353B CN 200810089812 CN200810089812 CN 200810089812 CN 200810089812 A CN200810089812 A CN 200810089812A CN 101547353 B CN101547353 B CN 101547353B
Authority
CN
Grant status
Grant
Patent type
Prior art keywords
code
codeword
decoding
decoded
input
Prior art date
Application number
CN 200810089812
Other languages
Chinese (zh)
Other versions
CN101547353A (en )
Inventor
白向晖
萩谷太郎
Original Assignee
富士通株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date

Links

Abstract

The invention provides a decoding accelerator for variable length codes. The decoding accelerator for the variable length codes comprises a code word distinguishing unit which is configured to judge whether a code word to be decoded in an input code segment is a short code word, and a short code word decoding accelerating unit which is configured to calculate code segment request information for requesting a next input code segment after the code word distinguishing unit determines that the code word to be decoded in the input code segment is the short code word so that the code segment request information is acquired to request the next input code segment before the decoding of the code word to be decoded is finished. The decoding accelerator for the variable length codes can be applied to a decoding device of the variable length codes to increase the decoding speed of the whole decoding device of the variable length codes by accelerating the decoding short code words with higher frequency of occurrences.

Description

可变长码解码加速装置 Variable length code decoding means accelerator

技术领域 FIELD

[0001] 本发明涉及可变长码的解码,更具体而言,涉及一种可变长码解码加速装置、利用该可变长码解码加速装置的可变长码解码装置及方法。 [0001] The present invention relates to a variable-length code decoding, and more particularly, to a variable length code decoding apparatus and method for decoding variable length code acceleration device by using the variable length code decoding accelerator.

背景技术 Background technique

[0002] 可变长码解码器是视频解码器的重要组成部分。 [0002] The variable length code decoder is an important part of the video decoder. 编码器根据符号在码流中出现的频率采用不同长度的码字进行编码,出现频率高的符号采用较短的码字进行编码,出现频率低的符号采用较长的码字进行编码。 The encoder uses the frequency of occurrence of the symbol in the code stream of codewords of different length encoding, a high occurrence frequency of a shorter codeword symbols encoded, the occurrence of low frequency symbol using longer codewords encoded. 由于码字的长度不是固定的,整个解码过程具有很高的相互依赖性,因此很难通过并行处理的方法来加快解码的速度。 Since the length of the codeword is not fixed, the entire decoding process with high interdependence, it is difficult to accelerate the decoding speed through parallel processing method. 目前最常用的可变长码解码方法是查表法。 The most commonly used method is the variable-length code decoding lookup table. 这种方法通过从码流中读取具有固定比特数的码段作为索引在码表中进行检索。 This method code segment having a fixed number of bits by reading from the code stream in the code as an index to search the table. 另外,为了减小码表的大小,加快码表检索速度,提出了诸如多级查表之类的一些优化方法。 Further, in order to reduce the size of the code table, the code table to accelerate the retrieval speed, as proposed several optimization methods multistage lookup table or the like. 这些优化方法例如1993年10月12日出版的名称为“Variable length decoding using lookup tables”的美国专利No. 5,253,053中所公开的采用两级或更多级查找表的可变长码解码方法,以及2000年1月4日出版的名称为“Dual-speed variable length decodingarchitecture for MPEG-2 video data” 的美国专利No. 6,011,498 中所公开的单独使用高速查找表来确定MPEG-2视频流中的主要数据:DCT系数的用于解码MPEG 视频数据的方法。 US Patent No. 5,253,053 of these optimization methods such as October 12, 1993 issue of the name "Variable length decoding using lookup tables" in the disclosed use of two or more levels of variable-length code lookup table use a separate decoding method, and the name of January 4, 2000, published as "Dual-speed variable length decodingarchitecture for MPEG-2 video data" of US Patent No. 6,011,498 disclosed a high-speed lookup table to determine MPEG -2 main video data stream: a method for decoding MPEG video data for the DCT coefficients. 这些相关专利的全部内容通过弓I用被结合于此。 The entire contents of these patents is by using bow I herein.

[0003] 如上所述,在常用的利用查找表的可变长码解码方法中,通常包括三个主要步骤或阶段:1.请求码段:解码器在这个阶段会输出用于请求下一个输入码段的码段请求信息,例如针对MPEG-2/4/VC-1标准,解码器将桶形移位器移位位数输出给桶形移位器,或者针对AVS标准,解码器将指数哥伦布码阶数输出给指数哥伦布码解码器以请求下一个输入码段;2.获得码段:在这个阶段,例如桶形移位器/指数哥伦布码解码器等用于产生下一个输入码段的单元将会根据移位位数/指数哥伦布码阶数等产生下一个输入码段给解码器; 3.解码码字:在这个阶段,输入码段将被作为索引,以在码表中检索得到被编码的符号以及解码下一个码字(即,请求下一输入码段)所需要的信息(例如桶形移位器移位位数或指数哥伦布码阶数)。 [0003] As described above, in a conventional lookup table using the variable length code decoding method, typically involves three main steps or stages: 1 Request code segment: the decoder output for this stage will request the next input code segment code segment request information, for example, MPEG-2/4 / VC-1 standard, the decoder outputs the shift amount of the barrel shifter to the barrel shifter, or for the AVS standard, the index decoder Golomb code of order exponential Golomb code is output to the decoder to request the next input code section; code segments 2 obtained: at this stage, for example, the barrel shifter / decoder exponential Golomb code and the like for generating an input code the next segment the unit will enter a code segment to a decoder according to the number of shifts produced / number of exponential Golomb code oRDER; the decoding codewords: at this stage, the input code segments will be used as an index to retrieve a code table the resulting encoded symbols and decode the next codeword (i.e., requests the next input code segment) information (e.g., the number of shifts or the order of the exponential Golomb code the barrel shifter) required. 以上三个步骤顺序执行,如图5所示,请求码段和获得码段阶段通常分别占用一个时钟周期,而由于视频标准中的码表都很大,通常被存放在静态存储器中,查找码表即解码码字阶段通常要占用两个以上的时钟周期,因而访问存储器的时间大大超过了请求码段和获得码段所需要的时间。 Sequentially performing the above three steps, shown in Figure 5, the request code segment and access code segment phases are usually contained in one clock cycle, and since the video standard code tables are large, usually stored in static memory, the lookup code tABLE i.e. decoded codeword stage typically takes up more than two clock cycles, so the memory access time is much longer than the period request code and code segments to obtain desired. 因此可变长码解码的速度主要取决于查找码表的时间。 Thus the speed of decoding variable length code depends on the time code lookup table. 对于较长的码字,由于解码了足够多的比特,所以访问存储器所需的较大的时间延迟是可以接受的。 For longer code word, since enough decoded bits, the time required to access a larger memory latency is acceptable. 但是对于短码字而言,过慢的解码速度会降低整个解码器的性能。 But for short codeword, the decoding speed is too slow will reduce the overall performance of the decoder.

[0004] 针对以上短码字解码速度影响整个解码器性能的问题,本发明提出了将长码字和短码字的解码区别对待的解码方法,从而提高整个可变长码解码器的解码速度。 [0004] For the above problem short code word decoding speed affect the overall performance of the decoder, the present invention proposes a decoding method for decoding the difference between the short and the long codeword codewords treated, thereby increasing the decoding speed of the entire variable length code decoder .

发明内容 SUMMARY

[0005] 根据本发明的一个方面,提供了一种可变长码解码加速装置,该装置包括:码字区分单元,被配置为判断输入码段中的待解码码字是否是短码字;以及短码字解码加速单元, 被配置为在码字区分单元判断出待解码码字为短码字之后,立即计算用于请求下一输入码段的码段请求信息,以使得在待解码码字的解码完成之前得到码段请求信息以请求下一输入码段。 [0005] In accordance with one aspect of the present invention, there is provided a variable length code decoding acceleration device, the apparatus comprising: a codeword distinguishing unit is configured to determine the codeword to be decoded in the input code is a short period if the codeword; and a short code word decoding acceleration unit configured to distinguishing unit determines that the code word to be decoded code word after a short codeword, code segments immediately next input code segments for requesting calculation request information, so that the code to be decoded to obtain a code segment request information requests the next input code word has been decoded before segments.

[0006] 根据本发明,优选地,短码字为长度小于或等于N比特的码字,这里N是根据在不对短码字进行优化的情况下解码码字所需要的平均时间和所期望的解码速度而预先设定的正整数。 [0006] According to the present invention, preferably, short code word length is less than or equal to a codeword of N bits, where N is the average time to decode the codeword case optimized at right short code words required and the desired decoding speed is preset positive integer.

[0007] 根据本发明,优选地,待解码码字是根据MPEG-2、MPEG_4或VC-I标准编码的码字。 [0007] According to the present invention, preferably, the code word to be decoded codeword is MPEG-2, MPEG_4 or VC-I standard based encoding. 在这种情况下,码段请求信息是用于获得下一输入码段的移位位数。 In this case, the code segments request information for obtaining the next input code segments of the shift amount. 码字区分单元根据可变长码编码规则,通过检查输入码段中的高位比特来判断输入码段中的待解码码字是否是短码字。 The codeword discrimination unit variable length code encoding rule, judged by the high-order bits in the code segment to be decoded check the input codeword if the input code is a short code word segment. 短码字解码加速单元被配置为根据可变长码编码规则,输出与被判断为短码字的待解码码字相对应的移位位数。 Short code word decoding unit is configured to accelerate the shift amount according to a variable length code encoding rules, outputs a short code word is determined to be corresponding to the decoded codeword.

[0008] 根据本发明,优选地,待解码码字是根据AVS标准编码的码字。 [0008] According to the present invention, preferably, the code word to be decoded is a code word encoded in accordance with the AVS standard. 在这种情况下,码段请求信息是用于获得下一输入码段的指数哥伦布码阶数。 In this case, the code segments request information for obtaining the index of the next input code section order Golomb code. 码字区分单元将输入码段的数值与预定阈值进行比较,并且当输入码段的数值小于或等于预定阈值时,码字区分单元判断出输入码段中的待解码码字是短码字。 Codeword distinction unit of the input value with a predetermined threshold value of the code segment are compared, and when the value of the input code segments are less than or equal to a predetermined threshold, the codeword discrimination unit determines to be decoded codeword of the input code segments are short code words. 此外,优选地,所述短码字是长度小于或等于N比特的码字,并且所述预定阈值是N比特指数哥伦布码所能表示的最大数值。 In addition, preferably, the short code word length is less than or equal to a codeword of N bits, and the predetermined threshold value is the maximum value N of bits that can be represented by the exponential Golomb code. 并且,短码字解码加速单元被配置为在待解码码字为短码字时,根据可变长码编码规则获得与输入码段的数值相对应的指数哥伦布码阶数。 And, a short code word decoding unit is configured to accelerate at a short codeword to be decoded codeword, exponent Golomb order to obtain the input value of the code segment corresponding to variable length code according to the encoding rules.

[0009] 因此,根据本发明的可变长码解码加速装置例如可以被应用在可变长码解码装置中,以加速出现频率较高的短码字的解码,从而大大提高可变长码解码器的解码速度。 [0009] Thus, the variable length code decoding apparatus according to the present invention, the accelerator may be applied, for example, variable length code decoding means, to accelerate the higher frequency shorter codeword decoding occurs, thereby greatly improving the variable-length code decoding 's decoding speed.

附图说明 BRIEF DESCRIPTION

[0010] 为了更完整的理解本发明,现参考以下结合附图的描述,在所有附图中将使用相同的标号来指示相同或类似的部件,其中: [0010] For a more complete understanding of the invention, is now described with reference to the following drawings, the same reference numerals in the drawings will refer to the same or similar components, wherein:

[0011] 图1示出了根据本发明的可变长码解码过程; [0011] FIG. 1 shows a variable length code decoding process according to the present invention;

[0012] 图2示出了根据本发明的一个实施例的可用于解码MPEG-2/4/VC-1标准的码流的可变长码解码装置; [0012] FIG. 2 shows a variable length code decoding means for decoding MPEG-2/4 / VC-1 standard code stream according to one embodiment of the present invention may be used;

[0013] 图3示出了根据本发明的另一实施例的可用于解码AVS标准的码流的可变长码解码装置; [0013] FIG. 3 shows a variable length code decoding means for decoding code may be the AVS another embodiment of the present invention according to flow;

[0014] 图4示出了根据本发明的又一实施例的可用于解码AVS标准的码流的可变长码解码装置; [0014] FIG. 4 shows a variable length code decoding means for decoding may be the AVS stream according to still another embodiment of the present invention;

[0015] 图5示出了现有技术中的可变长码解码过程。 [0015] FIG. 5 illustrates a variable length code decoding process of the prior art. 具体实施方式 detailed description

[0016] 下面将参考图1到图4描述根据本发明的可变长码解码装置的具体实施例。 [0016] below with reference to FIGS. 1 to 4 according to the specific embodiments described in the variable-length code decoding device according to the present invention. 在以下的描述中,分别针对视频压缩编解码中常用的MPEG-2/4/VC-1标准和AVS标准对本发明进行描述,但是本领域技术人员应当理解这种描述只是为了说明的目的,而适用于本发明的压缩编解码标准并不局限于此,例如还可以是JPEG标准、CCITT标准或者任何其它可变长编解码标准。 In the following description, respectively, for video compression codecs commonly used MPEG-2/4 / VC-1 standard and the AVS standard of the present invention will be described, those skilled in the art will appreciate that such description is for illustrative purposes, and compression codec standards applicable to the present invention is not limited to this, for example, the JPEG standard may also be, the CCITT standard or any other variable length coding standards.

[0017] 图1示出了根据本发明的可变长码解码过程,其中上半部分示出了输入码段中的待解码码字为短码字的解码过程,下半部分示出了输入码段中的待解码码字为长码字的解码过程。 [0017] FIG. 1 shows a variable length code decoding process according to the present invention, in which the upper half shows the decoding process decoded code word to be input to the code segment of short code words, the lower half shows the input codeword to be decoded in the decoding process for the code segment length codeword.

[0018] 在描述图1之前,首先对本发明中出现的输入码段和待解码码字这两个术语进行简单的说明。 [0018] Before describing FIG 1, first, the input code segments of the present invention appear in the code word to be decoded and the two terms will be briefly described. 通常情况下,可变长码流可被分成一段一段的固定长度的码段,然后这些码段被连续地输入到解码器中以进行解码。 Typically, the variable-length code stream may be divided into a fixed-length code segments of a section, then the code segments are successively input to the decoder for decoding. 每个固定长度的输入码段中可能包含一个或多个按照可变长编码规则编码的变长编码码字。 In each input fixed-length code segment may comprise one or more variable-length coding rule according to the variable length encoded codeword. 在解码器对一个输入码段进行解码时,一般只解码从该输入码段的最高位开始的一个编码码字,而输入码段中其余的未被解码的比特将在解码器对后续输入码段进行解码时被解码。 When a decoder decodes the input code section, a generally decoding encoded code words from the most significant bit of the input code segment, while the remaining segment of the input code is not decoded bits in the input code subsequent decoder when decoding the decoded segment. 因此,本发明中将这些固定长度的码段称为输入码段,而将实际被解码的从输入码段的最高位开始的一个具有可变长度的编码码字称为该输入码段中的待解码码字。 Thus, the fixed length code segment in the present invention is referred to as input code segments, but will actually be decoded from the most significant bit of the input code segment begins with a variable-length encoded codeword in the input code is called a segment the codeword to be decoded.

[0019] 如图1所示,为了加速可变长码的解码,在计算用于请求下一输入码段的码段请求信息时,针对待解码码字为短码字的输入码段和待解码码字为长码字的输入码段采取了不同的计算方式。 [0019] 1, in order to accelerate the variable-length code decoding, in the calculation code segment for the next input code segments request requesting information, for an input code segment to be decoded code word and a short code word to be decoded code word length codeword input code segments take a different calculation. 对于待解码码字为短码字的情况而言,在获得当前输入码段之后,在对输入码段中的待解码码字进行解码的同时,可由专用电路来计算用于请求下一输入码段的码段请求信息,而无需通过访问码表存储器来获得该码段请求信息,从而可以在待解码码字被解码完成之前得到码段请求信息以请求下一码段。 For the codeword to be decoded is the case of short code words, after obtaining the current input code segments, while the decoded code word to be input code decoding section, a dedicated circuit may be used to calculate the next input code request code segment segment request information without obtaining the requested information by accessing the code section of the code table memory to obtain a code segment can request information to request the next segment code to be decoded before the codeword is decoded. 因此,请求下一码段的阶段不是在对待解码码字的解码完成之后才开始的,而是与解码码字阶段并行地进行,从而加速了短码字的解码速度。 Thus, the next code segment request phase in treating not only started after the completion of decoding the decoded code word, but in parallel with the decoding codeword stage, thereby accelerating the decoding speed short codewords. 而对于待解码码字为长码字的情况而言,则仍然在解码阶段得到用于请求下一码段的码段请求信息,在解码完成之后开始请求下一码段。 For the case of a long codeword to be decoded codeword, the decoding stage still obtain a next code segment code segment request information request, after completion of the decoding start request to the next code segment. 根据本发明,由于出现频率较高的短码字的解码速度大大提高,所以整个可变长码解码过程被加速。 According to the present invention, due to the high frequency short code word decoding speed greatly improved the appearance, so the entire variable length code decoding process is accelerated. 这里,需要注意的是,在图1中,只是示意性地示出了通常情况下解码过程中的各个操作阶段在执行时间上的相对关系,该图示不应当被理解为对本发明的各个操作阶段的执行时间的限定,而只要解码过程符合本发明的总体思想即可。 Here, it should be noted that, in FIG. 1 only schematically shows the relative relationship between the various operating stages of the decoding process in the execution time Normally, the illustration should not be construed as each operation of the present invention. stage defined execution time, but the general idea as long as the decoding process consistent with the present invention.

[0020] 下面结合图2描述根据本发明的一个实施例的针对MPEG-2/4/VC-1标准的码流的可变长码解码装置及其解码过程。 [0020] FIG 2 is described below the variable-length code decoding device and a decoding process for MPEG-2/4 / VC-1 standard, according to one embodiment of the present invention in conjunction stream. 图2示出了根据本实施例的用于解码MPEG-2/4/VC-1视频流的可变长码解码装置201。 FIG 2 shows / variable length code VC-1 decoding means for decoding the video stream according to the present embodiment MPEG-2/4 201. 该可变长码解码装置201的输入是桶形移位器212所输出的具有固定长度的输入码段(InCOcKeg)202。 The input variable length code decoding means 201 is an input code segment (InCOcKeg) 202 having a fixed length of the barrel shifter 212 output. 输入码段202中的待解码码字在可变长码解码装置201中被解码以生成解码码字(DecC0de)213。 Input code word to be decoded code section 202 is decoded to generate a decoded codeword (DecC0de) 213 variable-length code decoding device 201. 图2所示的可变长码解码装置采用如图1所示的解码过程。 Variable length code decoding device shown in FIG 2 using a decoding process shown in FIG. 为了实现该解码过程,对于待解码码字为短码字的情况而言,用于请求后一个码段的码段请求信息将通过短码字解码加速单元207来计算得到,而不是通过访问解码单元208中的码表存储器来得到。 To achieve the decoding process for a codeword to be decoded is the case of short code words, the code segment is used to request a code segment to request information obtained short code word decoding unit 207 is calculated by the acceleration, instead of accessing the decoding unit 208 to obtain a code table memory. 下面具体描述图2中的各个功能单元在解码过程中的具体操作。 The following detailed description of the specific operation of each functional unit 2 in the decoding process of FIG.

[0021 ] 在可变长码解码装置201中,在从桶形移位器212获得输入码段202之后,输入码段202被输入到解码单元208进行解码,同时输入码段202还被输入到码字区分单元203。 [0021] In the variable-length code decoding device 201, the input code after obtaining section 212 from the barrel shifter 202, the input code section 202 is input to the decoding unit 208 for decoding, while the input code section 202 is also input to codeword discrimination unit 203. 在码字区分单元203中,判断输入码段202中的待解码码字是长码字(码字长度大于N)还是短码字(码字长度小于或等于N)。 Codeword discrimination unit 203, the decoded codeword is determined to be input code section 202 is a long code word (code word length greater than N) or a short code word (code word length is less than or equal to N). 如果输入码段202中的待解码码字被判断为短码字, 则码字区分单元203将输入码段202经由短码字信号(SCode) 205输入到短码字解码加速 If the input code word to be decoded code section 202 is determined to be a short codeword, the codeword distinguishing unit 203 is input via the input section 202 code short code word signal (SCode) 205 to a short code word decoding acceleration

5单元207。 5 unit 207.

[0022] 这里,N值可以根据正常情况下(即,未对短码字解码进行加速的情况下)解码一个码字所需要的平均时间(时钟周期数)和所要求的解码速度(比特/时钟周期)来预先设定。 (In other words, no accelerating the short code word decoding) [0022] Here, N values ​​can normally mean time (number of clock cycles) decoding one codeword needed and desired decoding rate (bits / clock cycles) set in advance. 例如,在要求解码速度至少为1比特/时钟周期,而解码一个码字平均需要5个时钟周期的情况下,如果不对短码字的解码进行优化,则长度小于5比特的码字的解码速度将会低于1比特/时钟周期,因此我们可将码字长度小于5的码字(即N = 4)设定为短码字以进行优化。 For example, in the case where the required decoding speed of at least one bit / clock cycle, decode one code word average of five clock cycles, if the decoding does not short codeword to optimize the length of less than 5 bits decoding speed codewords will be less than 1 bit / clock cycle, so we can codeword length codeword is less than 5 (i.e. N = 4) is set shorter codeword to optimize. 注意,此处关于N值的设定说明在其它实施例中也是适用的。 Note that the setting of the value of N in the other embodiments described herein are also suitable examples.

[0023] 根据MPEG-2/4/VC-1标准中所规定的编码规则,码字区分单元203可以利用如下的匹配机制来判断输入码段中的待解码码字是否为短码字。 [0023] The encoding rules standard MPEG-2/4 / VC-1 as predetermined code words using the following distinguishing unit 203 may be determined to be matching mechanism decoded code word segment whether the input code short code words. 例如N = 4,则从输入码段的最高位开始将输入码段与根据编码规则所规定的长度小于或等于4比特的码字进行比较, 以判断输入码段中的待解码码字是否是短码字。 For example, N = 4, the most significant bit from the input start code segment compares the input code segment length is less than or equal to 4 bits of the codeword according to a predetermined encoding rule to determine whether the code word to be decoded input code segment is short code word. 例如在MPEG-2标准中,短码字的判决规则可以根据MPEG-2标准中的设定被规定如下。 In the MPEG-2 standard, for example, the short codeword decision rule may be specified as follows according to the MPEG-2 standard is set. 在解码码表为DCT系数表0的情况下,短码字的判决规则如表1-1所示,而在解码码表为DCT系数表1的情况下,短码字的判决规则如表1-2所示。 In the case where the decoded code is 0 Table DCT coefficients, decision rule short code words as shown in Table 1-1, in the decoding of the DCT coefficient code table of Table 1, the short codeword decision rule shown in Table 1 FIG -2. 注意,在本文中所出现的表都只是为了表示映射机制,而非实际存储的映射表。 Note that the table in this article appearing just to express mapping mechanism, rather than the actual storage mapping table.

[0024] 表1-1 [0024] Table 1-1

[0025] [0025]

Figure CN101547353BD00061

[0026] 表1-2 [0026] Table 1-2

[0027]输入码段高四位 待解码码字的长度 待解码码字是否为短码字IOxx 2比特 是OlOx 3比特 是IlOx 3比特 是0111 4比特 是0110 4比特 是其他 > 4比特 否 H [0027] para four input symbols to be decoded codeword length of the codeword to be decoded short code word whether the bit is OlOx 3 IOxx 2 IlOx 3 bits are bits 4 bits 0111 is 01,104 bits other> 4 bits No

[0028] 其中,“X”代表该位为'0'或'1'。 [0028] wherein, "X" representing the bit is '0' or '1'.

[0029] 在解码码表为DCT系数表0的情况下,如表1-1所示,如果输入码段的最高两位为10或11,则根据MPEG-2标准,待解码码字即为10或11,其码字长度为2比特,因而该待解码码字可被判定为短码字;如果输入码段的最高三位为011,则待解码码字即011,其码字长度为3比特,因而该待解码码字可被判定为短码字;如果输入码段的最高四位为0100或0101,则待解码码字即0100或0101,其码字长度为4比特,因而该待解码码字可被判定为短码字;如果输入码段是其他情况,则根据MPEG-2标准,待解码码字的长度大于4比特,因此待解码码字可被判定为长码字。 [0029] In the case where the decoded code table of the DCT coefficient table 0, as shown in Table 1-1, if the input code segments of the two most significant bits is 10 or 11, according to the MPEG-2 standard, namely the code word to be decoded 10 or 11, which codeword length is 2 bits, so the codeword to be decoded may be determined that the short codeword; three most significant bits if the input code segments 011, i.e. the codeword to be decoded 011, which is the codeword length 3 bits, and thus the decoded code word can be determined to be a short code word; if the input code segments of the top four bits is 0100 or 0101, that is to be decoded codeword 0100 or 0101, which codeword length is 4 bits, and thus the to be decoded codeword can be determined that the short codeword; if the input code segments other case, in accordance with MPEG-2 standard, the length to be decoded codeword is greater than 4 bits, to be decoded codeword can be determined as long code word. 类似地,在当前解码码表为DCT系数表1的情况下,如表1-2所示,如果输入码段的最高两位为10,则根据MPEG-2标准,待解码码字即10,其码字长度为2比特,从而待解码码字可被判定为短码字;如果输入码段的最高三位为010或110, 则待解码码字即010或110,其码字长度为3比特,从而待解码码字可被判定为短码字;如果输入码段的最高四位为0111或0110,则待解码码字即0111或0110,其码字长度为4比特,从而待解码码字可被判定为短码字;如果输入码段是其他情况,则根据MPEG-2标准,当前实际被解码码字的长度大于4比特,因此这些输入码段中的待解码码字可被判定为长码字。 Similarly, in the code table for the currently decoded DCT coefficients in the case of Table 1, as shown in Table 1-2, if the top two bits of the input code segments is 10, according to the MPEG-2 standard, i.e., 10 code words to be decoded, which codeword length of 2 bits so that the codeword to be decoded may be determined that the short codeword; three most significant bits if the input code segments 010 or 110, i.e., the decoded codeword to be 010 or 110, which is the codeword length 3 bits so that the codeword to be decoded may be determined that the short codeword; if the input code segments of the top four bits is 0111 or 0110, the code word to be decoded i.e. 0111 or 0110, which codeword length is 4 bits, so as to be decoded code words can be determined that the short codeword; if the input code segments other case, in accordance with MPEG-2 standard, the current actual decoded length of the codeword is greater than 4 bits, to be decoded codeword of these input code segment may be determined as long code word.

[0030] 如上所述,码字区分单元203可以根据可变长码编码规则来判断输入码段中的待解码码字是否短码字,具体而言是根据当前解码码表的类型和输入码段202从最高位开始的至多N个比特来判断输入码段202中的待解码码字是否是长度小于或等于N比特的短码字。 [0030] As described above, the codeword distinguishing unit 203 may be determined to be decoded in the codeword if the input code short code word segment, specifically this type of decoder input code and the code table according to the variable-length code in accordance with coding rule section 202 starting from the highest bit up to N bits to determine whether the input code word to be decoded code section 202 is equal to or less than the length of the short code words of N bits. 例如,本领域技术人员可以在码字区分单元203中设计逻辑电路,来从输入码段的最高位开始将输入码段与当前解码码表中所规定的长度小于或等于N比特的码字进行比较,而一旦发现与当前解码码表中的某个短码字相匹配,则该判断过程就可以立即结束。 For example, those skilled in the art can codeword discrimination unit 203 designed logic circuit to the input code sections with the current code word N bits of a length less than or equal to the decoded code specified in the table from the most significant bit input code segment comparison, and if it is found to the current decoded code table matches a short code words, the determination process will be ended immediately. 因此,这里实际被比较的输入码段中的比特数最多为从最高位开始的4个比特。 Thus, where the actual input number of bits compared in the period up to the start code of 4 bits from the most significant bit.

[0031 ] 注意,在表1-1和1-2中所示的判决规则只是在编解码标准为MPEG-2标准的情况下的示例性判决规则,本领域技术人员应当很容易根据诸如MPEG-4/VC-1等的其它标准中所规定的编码规则来类似地设定在这些编解码标准中所适用的短码字判决规则。 [0031] Note that, in the exemplary case of Table decision rules and decision rules illustrated only 1-2 codec standard MPEG-2 standard 1-1, those skilled in the art should readily according as MPEG- other standard 4 / VC-1 or the like as a predetermined encoding rule is similarly set in a short codeword decision rule these codecs applicable standards.

[0032] 根据以上判决规则,码字区分单元203的输出有两个:模式信号(Mode) 204、短码字信号(SCode)205。 [0032] According to the above decision rule, output codeword has two distinguishing unit 203: mode signals (Mode) 204, a short code word signal (SCode) 205. 模式信号204指示输入码段202中的待解码码字是短码字(例如,模式信号=“0”)还是长码字(例如,模式信号=“1”)。 To be decoded codeword mode signal 204 indicative of the input code section 202 are short code words (e.g., the mode signal = "0") or long codeword (e.g., the mode signal = "1"). 如果模式信号指示输入码段202中的待解码码字是短码字,则码字区分单元203将输入码段202经由短码字信号205输出到短码字解码加速单元207。 If the code word to be decoded signal indicates that the input mode code segment 202 is a short codeword, the codeword discrimination unit 203 outputs the input section 202 to the code short code word decoding unit 207 via the acceleration signal 205 short code words.

[0033] 如上所述,对于待解码码字为长码字的输入码段和待解码码字为短码字的输入码段而言,计算用于请求下一输入码段的码段请求信息是在不同的单元中实现的。 [0033] As described above, the code word to be decoded is input codeword length and code segment to be decoded codeword period is shorter in terms of the input code codeword, calculation code segment for requesting the next input code segment request information It is implemented in different units. 在本实施例中,该码段请求信息是桶形移位器212的移位位数。 In the present embodiment, the code segment is the shift request information bits of the barrel shifter 212. 具体而言,针对待解码码字被判断为短码字的输入码段202,由短码字解码加速单元207根据短码字信号205计算并输出短码字移位位数(SCodeSBitNum) 209,而针对待解码码字被判断为长码字的输入码段202,由解码单元208根据输入码段202来访问码表存储器以得到并输出长码字移位位数(LCodeSBitNum)210。 Specifically, for the code word to be decoded it is determined that the input code period is shorter codeword 202, a short code word decoding unit 207 calculates and outputs the acceleration short code word shifted bits (SCodeSBitNum) 209 according to signal 205 by a short code word, and for the code word to be decoded is determined as the code length codeword input section 202, decoding section 208 to obtain a codeword length and outputs the shift amount (LCodeSBitNum) 210 input code segments 202 to access the memory in accordance with the code table.

[0034] 下面描述短码字解码加速单元207计算短码字移位位数的过程。 [0034] short code word decoding shift amount calculation short code word acceleration unit 207 will be described below. 移位位数就是当前输入码段中的待解码码字的长度。 Shifted bits is the length of the currently decoded codeword code segment to be input. 在本实施例中,可以利用一种映射机制来计算输入码段中的待解码码字的长度,即根据当前解码码表类型和输入码段中从最高位开始的至多N 个比特来计算输入码段中的待解码码字的长度。 In the present embodiment, one of the mapping mechanism may be utilized to calculate the length of the decoded code word to be input code segment, i.e., the input is calculated according to the most N bits and currently decoded code table type input code segment starting from the most significant bit the length of the decoded code word of the code segment to be. 例如在MPEG-2标准中,在当前解码码表为DCT系数表0的情况下,计算短码字移位位数的映射机制如表2-1所示,而在当前解码码表为DCT系数表1的情况下,计算短码字移位位数的映射机制如表2-2所示。 For example, the MPEG-2 standard, in the case where the code table currently decoded DCT coefficient table 0, the shift amount calculation map mechanisms short code words as shown in Table 2-1, and in the code table for the currently decoded DCT coefficients the case of table 1, the shift amount calculating short code word mapping mechanism as shown in table 2-2. 另外,本领域技术人员应当清楚,在其它编解码标准的情况下,可以根据这些标准中所规定的编码规则设定类似的映射机制,来计算短码字移位位数209。 Further, it should be apparent to those skilled, in other cases codec standard may be set similar to the mapping mechanism according to the coding rules defined in these standards, short code words to calculate the shift amount 209.

[0035] 表2-1 [0035] Table 2-1

[0036] [0036]

Figure CN101547353BD00081

[0037] 表2-2 [0037] Table 2-2

[0038] [0038]

Figure CN101547353BD00091

[0039] 注意,如上所述,虽然码字区分单元203和短码字解码加速单元207的操作都与当前解码码表的类型有关,但是本领域技术人员应当清楚对于MPEG-2/4标准而言,当前解码码表的类型是由视频解码器中的其它模块决定的,对于可变长码解码器来说并不是经常变化的,而相当于是一个固定的参数,因此在实际的实现方式中,码字区分单元203和短码字解码加速单元207只需要接收来自外部的当前解码码表类型信息(图中未示出),根据该信息按以上描述的方式选择使用适当的映射关系即可。 [0039] Note that, as described above, the operation distinguishing unit 203 although the code word and a short code word decoding acceleration unit 207 are related to the current type of decoder code table, those skilled in the art will appreciate that for MPEG-2/4 standard words, this type of decoder code table is determined by the video decoder of the other modules, for the variable length code decoder is not always changing, while the equivalent of a fixed parameter, so in practical implementations codeword distinguishing unit 203 and the short code word decoding acceleration unit 207 needs to receive the current decoding type information code table (not shown), based on the information from the outside to select the manner described above using an appropriate mapping relation to .

[0040] 以下描述根据本实施例的解码单元208中所进行解码过程。 [0040] The following describes the decoding process for the decoding unit 208 according to the embodiment. 在解码码字阶段,解码单元208使用输入码段202作为索引,以查找码表存储器中的可变长码码表来得到解码码字(DecCode)(游程/系数值)213和用于请求下一输入码段的长码字移位位数210。 In stage decoded codeword, the decoding unit 208 using the input section 202 as an index code, variable length code to find the code table in the code table memory to obtain a decoded codeword (DecCode) (run / coefficient values) 213 for the next request and a shift amount input long code word of the code segment 210. 解码单元208可以使用现有技术中通用的解码单元,因此,根据本发明的可变长码解码加速装置可以很容易地应用于现有的可变长码解码装置。 The decoding unit 208 may use common in the prior art decoding unit, therefore, the acceleration apparatus according to the variable length code decoder according to the present invention can be readily applied to a conventional variable length code decoding means.

[0041] 例如,根据MPEG-2,针对DCT系数表0,存储在码表存储器中的用于解码并生成移位位数的映射表如表3所示。 [0041] For example, according to MPEG-2, 0 for DCT coefficient table stored in the code table for decoding memory and generates a shift count mapping table shown in Table 3.

[0042] 表3 [0042] TABLE 3

[0043] [0043]

Figure CN101547353BD00092

[0044]OOlOlxxx 0 3 5OOlllxxx 3 1 5OOllOxxx 4 1 5OOOllOxx 1 2 6OOOlllxx 5 1 6OOOlOlxx 6 1 6 [0044] OOlOlxxx 0 3 5OOlllxxx 3 1 5OOllOxxx 4 1 5OOOllOxx 1 2 6OOOlllxx 5 1 6OOOlOlxx 6 1 6

[0045] 其中,“X”代表该位可以为'0'或'1'。 [0045] wherein, "X" representing the bit can be '0' or '1'.

[0046] 需要注意的是,虽然由解码单元208输出的用于请求下一输入码段的信息被简单地称为长码字移位位数210,但是由于根据本实施例的解码单元208是使用现有技术中的通用解码单元来实现的,所以不管输入码段中的待解码码字是长码字还是短码字,解码单元208都会在解码完成后输出移位位数,因此解码单元208所输出的长码字移位位数210 中实际也包括短码字移位位数。 [0046] It is noted that, although the information of the next input code decoding section 208 outputs a request by a unit is referred to simply as a long code word shift bits 210, but since the decoding unit embodiment 208 of the present embodiment is the prior art using common decoding unit is achieved, so regardless of the input code word to be decoded is a code segment length codeword or short codeword, the decoding unit 208 will output the shift amount after completion of decoding, the decoding unit length codeword output from the shift amount 208 210 also include short codeword actual shift amount. 但是本领域技术人员应当可以理解由于选通器214的存在,由解码单元208所输出的短码字移位位数将不会被输出到桶形移位器211,也就是说, 该短码字移位位数只是中间输出信息,对实际操作是不起作用的。 Those skilled in the art should be appreciated that due to the presence of gate 214, short code words shifted bits output by the decoding unit 208 will not be output to the barrel shifter 211, i.e., the short code word intermediate output only the number of shifts, the actual operation does not work.

[0047] 此外,由码字区分单元203输出的模式信号204被输入到选通器214,以控制选通器214将短码字移位位数209或长码字移位位数210输出到桶形移位器212。 [0047] In addition, the mode signal output unit 203 by the distinguishable codewords inputted to the gate 204 is 214, 214 to control the gate 210 outputs the short code word shifted bits or 209 bits length codeword to the shift barrel shifter 212. 例如,如果模式信号=“0”,则短码字移位位数209经由选通器214被输出作为移位位数(SbitNum) 211 以请求下一输入码段;否则,如果模式信号=“1”,则长码字移位位数210经由选通器214 被输出作为移位位数211以请求下一输入码段。 For example, if the mode signal = "0", the short code word bits 209 214 is shifted via the gate output as a shift bit number (SbitNum) 211 to request the next input code segment; or, if the mode signal = " 1 ", the shift length codeword bits 210 via the gate 214 is outputted as the shift bits 211 to request the next input code segments.

[0048] 下面结合图3描述根据本发明另一实施例的针对AVS标准的码流的可变长码解码装置301及其解码过程。 [0048] FIG. 3 is described below the variable length code decoding apparatus according to another embodiment for the AVS embodiment of the present invention, code stream 301 and the combined decoding process. 图3是用于解码AVS标准的码流的可变长码解码装置301。 FIG 3 is a variable length code decoding means 301 for decoding the code stream AVS standard. 该可变长码解码装置301的输入是指数哥伦布码解码器311所输出的具有固定比特数的输入码段(InCOcKeg)302。 The input variable length code decoding means 301 is an input code segment (InCOcKeg) 302 having a fixed number of bits in Exp-Golomb decoder 311 output. 输入码段302中的待解码码字在可变长解码装置301中被解码以生成解码码字(DecC0de)312。 Input code word to be decoded code section 302 is decoded to generate a decoded codeword (DecC0de) 312 variable-length decoding apparatus 301. 可变长码解码装置301也是采用如图1所示的解码过程。 Variable length code decoding means 301 also uses the decoding process shown in FIG. 1. 为了实现该解码过程,用于请求后一个码段的码段请求信息(在本实施例中,该码段请求信息为指数哥伦布码阶数)将通过短码字解码加速单元306来计算得到,而不是在解码完成之后才得到。 To achieve this decoding process, the code segment is used to request a code segment request information (in the present embodiment, the code segment information request order Exp-Golomb) unit 306 calculates the acceleration obtained by a short code word decoding, but not before we get after decoding is completed.

[0049] 在可变长码解码装置301中,在从指数哥伦布码解码器311获得输入码段302之后,输入码段302被输入到解码单元307以对其中的待解码码字进行解码,同时输入码段302还被输入到码字区分单元303。 [0049] In the variable length code decoding means 301, 311 after the input code section 302 is obtained from the exponential Golomb code decoder, the input code section 302 is input to the decoding unit 307 to be decoded for decoding a codeword in which, at the same time enter the code segment 302 is also input to the code word discrimination unit 303. 在码字区分单元303中,判断输入码段302中的待解码码字是长码字(码字长度大于N比特)还是短码字(码字长度小于或等于N比特)。 Codeword discrimination unit 303, the decoded codeword is determined to be input code section 302 is a long code word (code word length greater than N bits) or a short code word (code word length equal to or less than N bits). 然后, 如果码字区分单元303判断出输入码段302中的待解码码字为短码字,则码字区分单元303 将输入码段302经由短码字信号(SCode) 305输入到短码字解码加速单元306。 Then, if the codeword discrimination unit 303 determines to be decoded code of an input code segment 302 a short codeword, the codeword distinguishing unit 303 inputs the code segment 302 is input via a short codeword signal (SCode) 305 to a short code word acceleration decoding unit 306.

[0050] 码字区分单元303可以通过将输入码段的数值与预定的阈值比较来判断输入码段中的待解码码字是否是短码字。 [0050] codeword distinguishing unit 303 by the input value with a predetermined threshold value comparing code segment determines whether the decoded code word to be input code segments are short code words. 这里,输入码段是由指数哥伦布码解码器输出的固定长度的二进制比特流,其数值就是该比特流所表示的十进制数,因此输入码段的数值是很容易确定的。 Here, the input section is a binary code bit fixed-length output by an exponential Golomb code stream decoder, which is the decimal value indicated by the bit stream, the input value of the code segment is readily determined. 如果输入码段的数值小于或等于预定阈值,则判定输入码段中的待解码码字是短码字;相反,如果输入码段的数值大于该预定阈值,则判定该输入码段中的待解码码字是长码字。 If the value of the input code segments are less than or equal to a predetermined threshold value, it is determined to be decoded code word input code segments are short code words; Conversely, if the input value of the code segment is greater than the predetermined threshold value, in the input code segments to be determined decoded codeword is the codeword length. 例如N = 4,由于4比特指数哥伦布码所能表示的最大数是7,所以可将该阈值设定为7,因此判决规则如表4所示。 For example, N = 4, since the 4-bit exponential Golomb code can be represented by the maximum number is 7, it is possible to set the threshold value 7, so decision rule shown in Table 4. 如果输入码段的数值小于或等于7,则输入码段中的待解码码字被判定为短码字;而如果输入码段的数值大于7,则输入码段中的待解码码字被判定为长码字。 If the value of the input code segments are less than or equal to 7, the input to be decoded codeword in a code segment is determined to be shorter codeword; if the input value of the code segment is greater than 7, the input to be decoded codeword code segment is determined as long code word.

[0051] 表4 [0051] TABLE 4

[0052] [0052]

Figure CN101547353BD00111

[0053] 注意,根据指数哥伦布码的编码规则,实际上当输入码段的数值<=7时,输入码段中的待解码码字也有可能是长度大于4比特的码字。 [0053] Note that, according to the coding rule exponential Golomb code, in fact, when the value of the input code segments of <= 7, the codeword to be decoded in the input code segments may also be possible codeword length greater than 4 bits. 但是,这种情况比较少,而且即使将某些长度大于4比特的码字判断为短码字而对其进行优化,也并不影响本实施例的效果, 因为本发明只要保证长度小于或等于4比特的码字的解码被优化即可。 However, this is relatively small, and determines a codeword length is greater than 4, even if some of the bits of the short code word to be optimized, it does not affect the effect of the present embodiment, since the present invention as long as the length is less than or equal to 4 bits of decoded codeword can be optimized.

[0054] 码字区分单元303的输出有两个:模式信号(Mode) 304和短码字信号(SCode)305。 [0054] The distinguishing unit 303 outputs a codeword has two: a mode signal (Mode) 304 and a short code word signal (SCode) 305. 模式信号304指示当前输入码段302中的待解码码字是短码字(例如,模式信号=“0”)还是长码字(例如,模式信号=“1”)。 Mode signal 304 indicates that the current code word to be decoded input code section 302 is a short code word (e.g., the mode signal = "0") or long codeword (e.g., the mode signal = "1"). 如果模式信号指示输入码段302中的待解码码字是短码字,则码字区分单元303经由短码字信号305将输入码段302输出到短码字解码加速单元306。 If the code word to be decoded signal indicates that the input mode code segment 302 is a short codeword, the codeword discrimination unit 303 via a short code word of the input signal 305 is output to the short code segment 302 the code word decoding acceleration unit 306.

[0055] 短码字解码加速单元306根据经由短码字信号305输入的输入码段302的数值来计算短码字指数哥伦布码阶数(SCodeExpOrder) 308。 [0055] short code word decoding unit 306 calculates the acceleration based on the value of the input code segment input signal 305 via a short code word 302 is shorter codeword order Exp-Golomb (SCodeExpOrder) 308.

[0056] 在AVS标准中,指数哥伦布码阶数由解码所用的码表决定。 [0056] In AVS standard, the order of the exponential Golomb code is determined by the code table used for decoding. 在描述短码字解码加速单元306的具体操作过程之前,下面先简要说明根据AVS标准,确定短码字指数哥伦布码阶数的原理。 Before describing the specific short code word decoding operation acceleration unit 306, the following briefly described with reference to the AVS standard, the principle of determining the number of short code words Golomb code index order.

[0057] 根据AVS标准中所规定的编码规则,可以得出如表5-1至表5-3所示的对应关系, 其中根据码段数值和当前码表,可以得知下一码表。 [0057] The AVS encoding rules defined in the standard, the correspondence relationship can be obtained as shown in Table 5-1 to Table 5-3, such as, according to which the current value and the code segment code table, the table can be that the next code.

[0058] 表5-1 [0058] Table 5-1

[0059] [0059]

Figure CN101547353BD00121

[0060] 表5-2 [0060] Table 5-2

[0061] [0061]

[0062] [0062]

Figure CN101547353BD00131

[0063] [0063]

表5-3[0064] Table 5-3 [0064]

T \当一\前码 vlcO 一chroma vlcl—chroma vlc2 一chroma vlc3—chroma vlc4_chroma0 vie 1—chroma vlcl 一chroma vlc2_chroma vlc3—chroma vlc4 chroma1 vie 1 一chroma vlcl 一chroma vlc2_chroma vlc3_chroma vlc4 chroma2 vlcl chroma vlcl chroma vlc2 chroma vie 3—chroma vlc4 chroma3 vie 1 一chroma vlcl_chroma vlc2—chroma vlc3_chroma vlc4—chroma4 vlcl—chroma vlcl—chroma vlc2_chroma vie 3—chroma vlc4 chroma5 vlcl 一chroma vlc2—chroma vlc2—chroma vie 3—chroma vlc4 chroma6 vlcl 一chroma vlc2 一chroma vlc2 一chroma vie 3—chroma vlc4 chroma7 vlcl chroma vlcl chroma vlc3_chroma vlc3_chroma vlc4—chroma T \ When a \ front yard vlcO a chroma vlcl-chroma vlc2 a chroma vlc3-chroma vlc4_chroma0 vie 1-chroma vlcl a chroma vlc2_chroma vlc3-chroma vlc4 chroma1 vie 1 a chroma vlcl a chroma vlc2_chroma vlc3_chroma vlc4 chroma2 vlcl chroma vlcl chroma vlc2 chroma vie 3-chroma vlc4 chroma3 vie 1 a chroma vlcl_chroma vlc2-chroma vlc3_chroma vlc4-chroma4 vlcl-chroma vlcl-chroma vlc2_chroma vie 3-chroma vlc4 chroma5 vlcl a chroma vlc2-chroma vlc2-chroma vie 3-chroma vlc4 chroma6 vlcl a chroma vlc2 a chroma vlc2 a chroma vie 3-chroma vlc4 chroma7 vlcl chroma vlcl chroma vlc3_chroma vlc3_chroma vlc4-chroma

[0065] 根据AVS标准中的编码规则,指数哥伦布码阶数与所用的码表相对应。 [0065] The AVS standard encoding rules, the order of the exponential Golomb code with the code corresponding to the appearances. 因此,可以得出如表6-1至6-3所示的映射关系,其中根据码段数值和当前码表可以确定出指数哥伦布码阶数。 Thus, the mapping relationship can be obtained as shown in Table 6-1 to 6-3, which may be determined according to the exponential Golomb code order value and the current code segment code table.

[0066] 表6-1 [0066] Table 6-1

[0067] [0067]

Figure CN101547353BD00141

[0068] [0068]

Figure CN101547353BD00142

[0069] [0069]

[0070] [0070]

表6-2 Table 6-2

Figure CN101547353BD00151

[0071] 表6-3 [0071] Table 6-3

Figure CN101547353BD00152

[0074] 利用AVS标准的上述特性,根据本实施例的短码字解码加速单元306可以根据当前码表和码段数值得出待解码短码字的指数哥伦布码阶数。 [0074] With the above-described characteristics of AVS, accelerated short code word decoding according to the present embodiment unit 306 and the code table according to the current value of the code segment to be decoded short code word derived exponential Golomb code order. 下面详细描述根据本实施例在短码字解码加速单元306中所进行的操作。 According to the present embodiment, the operation performed in the short code word decoding acceleration unit 306 is described in detail below.

[0075] 当码字区分单元303确定输入码段302中的待解码码字为短码字时,输入码段302 经由短码字信号305被输入到短码字解码加速单元306。 [0075] When the discrimination unit 303 determines a codeword to be decoded codeword input code segment 302 a short code word, the code input section 302 via the input 305 is a short codeword to signal short code word decoding acceleration unit 306. 然后,短码字解码加速单元306可以根据如表6-1至6-3所示的映射关系来计算用于请求下一输入码段的信息,即短码字指数哥伦布码阶数308。 Then, accelerated short code word decoding unit 306 may be information of the next input code segments for requesting calculated according to the mapping shown in Table 6-1 to 6-3, i.e., the short codeword order Exp-Golomb 308.

[0076] 此外,短码字解码加速单元306还根据如表5-1至5_3所示的映射关系来确定下一码表类型,并保存所确定的下一码表类型,以便用于在下一待解码码字是短码字时的操作。 [0076] In addition, a short code word decoding unit 306 further acceleration is determined according to the mapping relation as shown in Tables 5-1 to 5_3 next code type table, the table and save the determined type of the next code, for use in the next codewords are to be decoded at the time of operation of short code words. 也就是说,在前一待解码码字是短码字的情况下,短码字解码加速单元306使用所存储的上次确定出的码表类型来确定当前待解码码字的指数哥伦布码阶数以及下一码表类型。 That is, a first codeword to be decoded in the case of short code words, a short code word decoding acceleration unit 306 using the stored last determined the type of code table to determine the current codeword to be decoded in order Exp-Golomb number and type of the next code table. 另外,在前一待解码码字是长码字的情况下,短码字解码加速单元306使用经由码表类型标识信号(CodTabIdf) 315从解码单元307中的码表选择/指数哥伦布码阶数生成单元313 接收的码表类型,来确定当前待解码码字的指数哥伦布码阶数以及下一码表类型。 Further, the previous codeword to be decoded is a case where the length codeword, a short code word decoding unit 306 uses the number of acceleration signal via a code table type identifier (CodTabIdf) 315 from the code table selection unit 307 in the decoding / Order Exp-Golomb type code table generating unit 313 received, to determine a current codeword to be decoded order Exp-Golomb code and a next table type. 本领域技术人员可以容易地理解,例如可以通过模式信号304来控制短码字解码加速单元306选择使用适当的码表类型标识。 Those skilled in the art can readily appreciate, for example, may be controlled to accelerate the short code word decoding unit 306 selects the signal 304 by the mode using the appropriate code table type identifier.

[0077] 上述的表5-1至5-3以及表6-1至6_3中所示的映射关系可以存储或者编程在短码字解码加速单元306中,也可以存储在其他存储设备中。 [0077] The above-described Table 5-1 and Table 6-1 to 5-3 to the mapping relationship shown in 6_3 may be stored or programmed in accelerating short code word decoding unit 306, may be stored in other storage devices. 例如,当待解码码字为短码字时,用于确定下一码表类型的模块也可以在短码字解码加速单元306之外单独提供,然后该模块所确定的码表类型再被输入到短码字解码加速单元306以用于其在下一待解码码字是短码字时的操作。 For example, when a codeword to be decoded short code word, the next code table for determining the type of module may also be provided separately from the acceleration short code word decoding unit 306, the module then determines the type of the code table, are input 306 for its codeword to be decoded next is the operation when a short code words to a short code word decoding acceleration unit. 此外,短码字解码加速单元306或单独设置的用于确定下一码表类型的模块所计算出的表示下一码表类型的值可以存储在设置在短码字解码加速单元306 中的寄存器中,或者其他存储设备中,例如,可以使用设置在解码单元307中的存储设备。 Furthermore, the short codeword decoding unit 306 or the acceleration module separately provided for determining the type of the next code table represent the calculated value for the next type of code tables may be stored in a register provided in accelerating short code word decoding unit 306 , or in other storage devices, for example, may be used provided the storage device 307 in the decoding unit.

[0078] 以下描述根据本实施例的解码单元307中的解码过程。 [0078] The following describes a decoding process decoding unit 307 according to the embodiment. 用于解码输入码段中的待解码码字的可变长码码表被存放在解码单元307中的码表存储器中。 Code table for variable length code to be decoded codeword decoding input code segment is stored in the decoding unit 307 in the code table memory. 解码单元307使用输入码段302作为索引,查找码表存储器中的可变长码码表来得到解码码字(游程/系数值)312。 A decoding unit 307 using the input section 302 as an index code, variable length code lookup code table in the code table memory to obtain a decoded codeword (run / coefficient value) 312. 例如,在AVS标准中,用于解码码字的可变长码码表vlcOjntra如表7所示。 For example, in the AVS standard, a code table for variable length code decoding codewords vlcOjntra As shown in Table 7.

[0079] 表7 [0079] TABLE 7

[0080] [0080]

Figure CN101547353BD00161

[0081] 此外,在得到解码码字之后,解码码字(游程/系数值)312将会被输入到码表选择/指数哥伦布码阶数生成单元313中以计算并输出长码字指数哥伦布码阶数(LCodeExpOrder) 309。 [0081] Further, after obtaining the decoded code words, decoded code words (run / coefficient value) 312 will be inputted to the code table selector / exponential Golomb code order generation unit 313 to calculate and output the length codeword Exp-Golomb order (LCodeExpOrder) 309. 与针对短码字的指数哥伦布码阶数计算方法类似,码表选择/指数哥伦布码阶数生成单元313通过当前解码所得到的系数值来决定解码下一个码字要使用的码表,并进而决定用于请求下一个码字的指数哥伦布码阶数。 Similarly for the short codeword index order of calculation Golomb code, the code table selection / exponential Golomb code order generation unit 313 determines the code table decode the next codeword to be used by the coefficient values ​​decoded from the current, and thus decides to request the next codeword for the exponential Golomb code order. 根据以上描述,本领域技术人员应当理解,码表选择/指数哥伦布码阶数生成单元313中可以包括用于实现上述功能的逻辑电路或逻辑单元。 According to the above description, one skilled in the art will appreciate, the code table selection / order exponential Golomb code generation unit 313 may include a logic circuit, or logic unit to achieve the above functions.

[0082] 需要注意的是,在本实施例中,虽然由解码单元307输出的用于请求下一输入码段的信息被简单地称为长码字指数哥伦布码阶数309,但是由于在本实施例中,不管输入码段中所包含的待解码码字是长码字还是短码字,码表选择/指数哥伦布码阶数生成单元313都会在解码完成后根据解码码字计算并输出指数哥伦布码阶数,因此码表选择/指数哥伦布码阶数生成单元313所输出的长码字指数哥伦布码阶数中实际也包括短码字指数哥伦布码阶数。 [0082] Note that, in this embodiment, although used by the decoding unit 307 outputs the request information of the next input code segment is simply referred to as exponential-Golomb code length codeword number order of 309, but since the present embodiment, regardless of the input to be decoded code word in the code segment included in a long code word or short code word, the code table selection / exponential Golomb code order generation unit 313 will calculate and output index based on the decoded codeword after decoding Golomb code order, so the code table selection / generation order exponential Golomb code number 313 MLCs Golomb code length codeword index output actually also include short codeword order exponential Golomb code. 但是本领域技术人员应当可以理解由于选通器314的存在,由码表选择/ 指数哥伦布码阶数生成单元313所输出的短码字指数哥伦布码阶数将不会被输出到指数哥伦布码解码器311,也就是说,该短码字指数哥伦布码阶数只是中间输出信息,对实际操作是不起作用的。 Those skilled in the art should be appreciated that due to the presence of gate 314, the code table selector generates order Exp-Golomb / short codeword index output unit 313 Golomb code order will not be output to the exponential Golomb code decoder 311, i.e., the exponential-Golomb code short code words only the intermediate output order information for the actual operation does not work.

[0083] 注意,虽然在图3中,码表选择/指数哥伦布码阶数生成单元313被提供在解码单元307的内部,但是实际上该码表选择/指数哥伦布码阶数生成单元也可以被提供在解码单元307的外部。 [0083] Note that although in FIG. 3, the code table selection / order exponential Golomb code generation unit 313 is provided in the interior of the decoding unit 307, but in fact the code table selector / order exponential Golomb code generation unit may be providing the outer decoding unit 307.

[0084] 此外,由码字区分单元303输出的模式信号304被输入到选通器314,以控制选通器314输出到指数哥伦布码解码器311的指数哥伦布码阶数(ExpOrder) 310。 [0084] In addition, the mode signal output from the unit 303 to distinguish between code word 304 is input to the gate 314 to control the output of the gate 314 to the exponential Golomb code decoder 311 exponential Golomb code of order (ExpOrder) 310. 如果模式信号=“0”,则短码字指数哥伦布码阶数308将经由选通器314被输出作为指数哥伦布码阶数310以请求下一输入码段;否则,如果模式信号=“1”,则长码字指数哥伦布码阶数309将经由选通器314被输出作为指数哥伦布码阶数310以请求下一输入码段。 If the mode signal = "0", the short codeword order Exp-Golomb gate 308 is the output 314 via the Golomb code as an index order of 310 to request the next input code segment; or, if the mode signal = "1" , the exponential-Golomb code length codewords order of 309 to gate 314 is output via the Golomb code as an index order of 310 to request the next input code segments.

[0085] 根据本实施例的可变长码解码装置与现有技术中的可变长码解码装置相比,仅仅增加可变长码解码加速模块,因此很容易基于现有的可变长码解码装置来实现。 [0085] Compared with the prior art variable length code decoding means variable length code decoding apparatus according to the present embodiment, only the variable length code decoder to increase acceleration module, it is easy to variable-length code based on the existing decoding means. 但是,由于在码表选择/指数哥伦布码阶数生成单元313中,不管待解码码字是长码字还是短码字,都会计算用于获得后一输入码段的指数哥伦布码阶数,而由码表选择/指数哥伦布码阶数生成单元313所输出的短码字指数哥伦布码阶数实际上是不起作用的。 However, since a code table selection / order exponential Golomb code generation unit 313, whether the code word to be decoded is a code word or short length code words, the calculation will be used to obtain an input exponential Golomb code segment order, and a code table selector exponential Golomb code numbers generated short order Exp-Golomb codewords output unit 313 order / is actually ineffective. 因此,在本发明的另一实施例中,对图3所示的可变长码解码装置进行了改进。 Thus, in another embodiment of the present invention, variable-length code decoding device shown in FIG. 3 has been improved.

[0086] 下面结合图4描述根据本发明的另一实施例的针对AVS标准的码流的可变长码解码装置401及其解码过程。 [0086] below with reference variable length code decoding apparatus stream 4 depicts another embodiment of the present invention for AVS-401 and its decoding process. 图4示出了根据本实施例的用于解码AVS标准的码流的可变长码解码装置401,其中使用相同的标号表示与图3所示的可变长码解码装置301中相同的单元和信号,并且在下面的描述中,省略了关于与图3中相同的模块的描述,而仅描述图4与图3所示的可变长码解码装置的不同之处。 FIG. 4 shows a variable length code decoding means 401 for decoding a code according to the present embodiment of AVS stream, wherein like reference numerals denote like the variable length code decoding apparatus 301 shown in FIG units and a signal, and in the following description, the same description is omitted on the module in FIG. 3, and only the differences from FIG. 4 and the variable length code decoding device shown in FIG. 3.

[0087] 根据该实施例的可变长码解码装置401与图3所示的可变长码解码装置301的区别仅在于通过开关415控制解码码字是否被输入到码表选择/指数哥伦布码阶数生成单元313。 [0087] The variable length code decoding means 401 of the embodiment in FIG 3 the difference variable length code decoding apparatus 301 illustrated only in that whether the input to the code table selection / exponential Golomb code by the switch 415 controls the decoding codeword order number generation unit 313. 开关415的断开/闭合是由码字区分单元303输出的模式信号来控制的。 Switch 415 opening / closing mode by the code word output from the signal distinguishing unit 303 controlled. 例如,如果模式信号=“0”,则开关415断开,如果模式信号=“1”,则开关415闭合。 For example, if the mode signal = "0", the switch 415 is turned off, if the mode signal = "1", the switch 415 is closed. 当待解码码字为短码字时,码字区分单元303输出模式信号=“0”,从而使开关415断开,此时码表选择/指数哥伦布码阶数生成单元313不工作,而当待解码码字为长码字时,码字区分单元303输出模式信号=“1”,从而使开关415闭合,此时解码码字(游程/系数值)312就会被输入到码表选择/指数哥伦布码阶数生成单元313,因此码表选择/指数哥伦布码阶数生成单元313 将计算并输出长码字指数哥伦布码阶数(LCOdeEXp0rder)309。 When a codeword to be decoded short code word, the code word distinguishing unit 303 outputs the mode signal = "0", so that the switch 415 is turned off, this time code table selector / order exponential Golomb code generation unit 313 does not operate, and when is to be decoded codeword length of a codeword, codeword distinguishing unit 303 outputs the mode signal = "1", so that the switch 415 is closed, then decoded codeword (run / coefficient value) 312 will be inputted to the code table selector / order exponential Golomb code generation unit 313, so the code table selection / order exponential Golomb code generation unit 313 calculates and outputs the exponential-Golomb code length codeword order (LCOdeEXp0rder) 309. 注意,与图3类似,虽然在图4中码表选择/指数哥伦布码阶数生成单元313也被提供在解码单元407的内部,但是实际上该码表选择/指数哥伦布码阶数生成单元也可以被提供在解码单元407的外部。 Note, similar to FIG. 3, although it is provided inside the decoding unit 407 in FIG. 4 the code table selector / exponential Golomb code of order generation unit 313 but in practice the code table selector / exponential Golomb code generation unit can also order It may be provided in the outer decoding unit 407.

[0088] 本发明并不限于上述实施例,并且在不脱离本发明精神的范围内可以通过修改部分配置来实施本发明。 [0088] The present invention is not limited to the above embodiments, and without departing from the present invention may be implemented by modifying portion disposed within the spirit of the present invention. 例如,本发明的可变长码解码装置和方法可以被用于除以上所描述的MPEG-2/4/VC-1标准和AVS标准以外的各种可变长编解码标准,而且本领域技术人员应当清楚针对不同的标准如何识别短码字以及计算用于请求输入码段的码段请求信息。 For example, the variable length code decoding apparatus and method of the present invention may be used in various other than MPEG-2/4 / VC-1 standard and the AVS standard described above variable length coding standard, and those skilled in the It should be clear how the person identification short code words for different standards and code segments for calculating code segment request input request information.

[0089] 此外,以上附图和相关实施例的说明中所提到的各个模块,例如码字区分单元、短码字解码加速单元等中可能包括为了使得该模块实现在本发明实施例中的相应功能而设置的存储单元、控制单元、时钟提供单元等。 [0089] Further, the above description and the figures related modules in the embodiment mentioned, for example, a codeword differentiating unit, acceleration short code word decoding unit or the like may be included in order to make the module is implemented in an embodiment of the present invention the storage unit provided corresponding function, the control unit, the clock supply unit and the like. 这些单元所能实现的功能都是本领域技术人员所公知的,因此在本发明的描述中未进行具体说明,但是本领域技术人员应当清楚如何根据本发明所描述的功能来适当地设计提供这些功能单元。 These units can realize the function are known to those skilled in the art, and therefore not specifically described in the description of the present invention, those skilled in the art would know how to design these appropriately depending upon the functionality described herein functional unit. 此外,本发明中所公开的各个模块以及这些模块中的提供相应功能的功能单元可以用软件、硬件或者其结合来实现,并且各个模块之间的信号可以以任何可能的方式被传递。 In addition, the functional units provide the respective functions of the various modules disclosed in the present invention, and these modules may be software, hardware, or a combination thereof to achieve, between the various modules and the signal may be transmitted in any way possible.

[0090] 另外,需要注意的是,在说明书中所提到的各个表只是为了表明映射关系,而并不一定是实际使用的映射表,并且这些表中的每一个所表示的映射关系可以用单独的映射表来实现,也可以与其它表合并在一个映射表中来实现。 [0090] Further, it is noted that, each table in the specification referred to only to show a mapping relation, and not necessarily the actual use of the mapping table, and each of the mappings shown in these tables can be used separate mapping table to achieve, can be combined with other tables in a mapping table to achieve. 而且,本发明实施例中所提到的各个模块可以单独实现,也可以适当地与其它模块集成在一起来实现。 Further, the embodiment of the present invention, various modules mentioned embodiments may be implemented separately, it can be appropriately integrated with other modules to achieve together.

[0091] 综上,对本发明的以上实施例的说明只是用于帮助理解本发明的方法及其核心思想,而不希望限制本发明的范围;同时,对于本领域的普通技术人员,依据本发明的思想,在具体实施方式及应用范围上还会有各种变化和改型,这些变化和改型均落入所附权利要求所限定的本发明的范围内。 [0091] In summary, the above described embodiments of the present invention is only used to help understand the method and core idea of ​​the present invention and are not intended to limit the scope of the present invention; while those of ordinary skill in the art, according to the present invention I thought, in specific embodiments and applications will have various changes and modifications, such changes and modifications fall within the scope of the invention as defined by the appended claims.

Claims (11)

  1. 1. 一种可变长码解码加速装置,其特征在于包括:码字区分单元,被配置为判断输入码段中的待解码码字是否是短码字;以及短码字解码加速单元,被配置为在所述码字区分单元判断出所述待解码码字为短码字之后,立即计算用于请求下一输入码段的码段请求信息,以使得在所述待解码码字的解码完成之前得到所述码段请求信息以请求下一输入码段。 A variable length code decoding acceleration device, comprising: a codeword distinguishing unit is configured to determine whether the code word to be decoded in the input code is a short period code word; and a short code word decoding acceleration unit after the codeword is configured to distinguish means determines that the code word is shorter codeword to be decoded, the next code segment is calculated immediately input code segment request information for requesting, such that decoding the codeword to be decoded the resulting code segment request information to request the next input code segments before completion.
  2. 2.如权利要求1所述的可变长码解码加速装置,其中,所述短码字为长度小于或等于N 比特的码字,所述N是根据解码码字所需要的平均时间和所期望的解码速度而预先设定的正整数。 2. The variable length code decoder as claimed in claim 1, said accelerator, wherein said short code word length is less than or equal to a codeword of N bits, N is the decoded codeword based on the average time required and the decoding the desired speed predetermined positive integer.
  3. 3.如权利要求1所述的可变长码解码加速装置,其中,所述待解码码字是根据MPEG-2、 MPEG-4或VC-I标准编码的码字。 The variable length code decoder as claimed in claim 1, said accelerating means, wherein the codeword is to be decoded according to MPEG-2, MPEG-4 or VC-I standard coded codewords.
  4. 4.如权利要求3所述的可变长码解码加速装置,其中,所述码段请求信息是用于获得所述下一输入码段的移位位数。 The variable length code decoder as claimed in claim 3, said accelerating device, wherein the code segment request information for obtaining the shift amount for the next input code segments.
  5. 5.如权利要求3所述的可变长码解码加速装置,其中所述码字区分单元根据可变长码编码规则,通过检查所述输入码段中高位比特来判断所述输入码段中的待解码码字是否是短码字。 5. The variable length code decoder as claimed in claim 3, said accelerating device, wherein the codeword distinguishing unit according to a variable length code encoding rule, the upper bits of the input code segment determination by checking the input code segments whether the code word to be decoded is shorter codeword.
  6. 6.如权利要求4所述的可变长码解码加速装置,其中,所述短码字解码加速单元被配置为根据可变长码编码规则,输出与被判断为短码字的所述待解码码字相对应的移位位数。 The variable length code decoder as claimed in claim 4, said accelerating means, wherein said short code word decoding unit is configured to accelerate the variable length code encoding rules, outputs a short code word is determined to be the decoded code word corresponding to the shift amount.
  7. 7.如权利要求1所述的可变长码解码加速装置,其中,所述待解码码字是根据AVS标准编码的码字。 7. The variable length code decoder as claimed in claim 1, said accelerating device, wherein the code word to be decoded is a code word encoded in accordance with the AVS standard.
  8. 8.如权利要求7所述的可变长码解码加速装置,其中,所述码段请求信息是用于获得所述下一输入码段的指数哥伦布码阶数。 8. The variable length code decoder as claimed in claim 7, said accelerating device, wherein the code segment request information for obtaining the index of the next input code section order Golomb code.
  9. 9.如权利要求7所述的可变长码解码加速装置,其中,所述码字区分单元将所述输入码段的数值与预定阈值进行比较,并且当所述输入码段的数值小于或等于所述预定阈值时,所述码字区分单元判断出所述输入码段中的待解码码字是短码字。 9. The variable length code decoder as claimed in claim 7 of the accelerating device, wherein the distinguishing means codeword of the input code segment value with a predetermined threshold value, and when the input value is less than or code segment is equal to the predetermined threshold value, the discrimination unit determines that the code word to be decoded code word in the input code word is a short code segment.
  10. 10.如权利要求9所述的可变长码解码加速装置,其中,所述短码字是长度小于或等于N比特的码字,并且所述预定阈值是N比特指数哥伦布码所能表示的最大数值。 The variable length code decoder as claimed in claim 9, said accelerating means, wherein said short code word length is less than or equal to a codeword of N bits, and the predetermined threshold is a N-bit exponential Golomb code can be represented by maximum value.
  11. 11.如权利要求8所述的可变长码解码加速装置,其中,所述短码字解码加速单元被配置为在所述待解码码字为短码字时,根据可变长码编码规则获得与所述输入码段的数值相对应的指数哥伦布码阶数。 11. The variable length code decoder as claimed in claim 8, the accelerating means, wherein said short code word decoding unit is configured to accelerate the codeword to be decoded short code word, according to the variable-length code encoding rules obtaining said input numerical order Exp-Golomb code segment corresponds.
CN 200810089812 2008-03-28 2008-03-28 Decoding accelerator for variable length codes CN101547353B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200810089812 CN101547353B (en) 2008-03-28 2008-03-28 Decoding accelerator for variable length codes

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN 200810089812 CN101547353B (en) 2008-03-28 2008-03-28 Decoding accelerator for variable length codes
JP2009080360A JP5201052B2 (en) 2008-03-28 2009-03-27 Variable length code decoding optimization device

Publications (2)

Publication Number Publication Date
CN101547353A true CN101547353A (en) 2009-09-30
CN101547353B true CN101547353B (en) 2011-09-07

Family

ID=41194182

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200810089812 CN101547353B (en) 2008-03-28 2008-03-28 Decoding accelerator for variable length codes

Country Status (2)

Country Link
JP (1) JP5201052B2 (en)
CN (1) CN101547353B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102547267B (en) * 2010-12-29 2014-04-16 福州瑞芯微电子有限公司 Video data decoding method and system
US8593309B2 (en) * 2011-11-30 2013-11-26 Intel Mobile Communications GmbH Method for efficient decoding of variable length codes

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1703090A (en) 2005-06-09 2005-11-30 清华大学 High-speed changeable long code parallel decoder
CN1839550A (en) 2003-07-15 2006-09-27 英特尔公司 Method of decoding variable length prefix codes
CN1913638A (en) 2005-08-09 2007-02-14 C&S技术有限公司 H.264 decoding method for fast CAVLC
CN101116342A (en) 2005-03-30 2008-01-30 英特尔公司 Multistandard variable length decoder with hardware accelerator

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6215424B1 (en) * 1998-12-16 2001-04-10 Thomson Licensing S.A. System for variable length codeword processing suitable for video and other applications
JP2001007706A (en) * 1999-06-24 2001-01-12 Toshiba Corp Decoding device for variable length code
JP2003209474A (en) * 2002-01-15 2003-07-25 Matsushita Electric Ind Co Ltd Variable length coding decoding device
JP2006060723A (en) * 2004-08-24 2006-03-02 Victor Co Of Japan Ltd Variable length code decoder
JP4079965B2 (en) * 2005-10-03 2008-04-23 株式会社アクセル Decoding system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1839550A (en) 2003-07-15 2006-09-27 英特尔公司 Method of decoding variable length prefix codes
CN101116342A (en) 2005-03-30 2008-01-30 英特尔公司 Multistandard variable length decoder with hardware accelerator
CN1703090A (en) 2005-06-09 2005-11-30 清华大学 High-speed changeable long code parallel decoder
CN1913638A (en) 2005-08-09 2007-02-14 C&S技术有限公司 H.264 decoding method for fast CAVLC

Also Published As

Publication number Publication date Type
JP5201052B2 (en) 2013-06-05 grant
CN101547353A (en) 2009-09-30 application
JP2009246973A (en) 2009-10-22 application

Similar Documents

Publication Publication Date Title
US6195024B1 (en) Content independent data compression method and system
US5587710A (en) Syntax based arithmetic coder and decoder
US7714747B2 (en) Data compression systems and methods
US7289047B2 (en) Decoding variable length codes while using optimal resources
US20090168868A1 (en) Systems and apparatuses for performing CABAC parallel encoding and decoding
US5710562A (en) Method and apparatus for compressing arbitrary data
US6842124B2 (en) Variable length decoder
US5675332A (en) Plural-step chunk-at-a-time decoder for variable-length codes of Huffman type
US20030085822A1 (en) High performance memory efficient variable-length coding decoder
US20040240559A1 (en) Context adaptive binary arithmetic code decoding engine
US7443318B2 (en) High speed context memory implementation for H.264
US20040059770A1 (en) Method and apparatus for arithmetic coding, including probability estimation state table creation
US7262722B1 (en) Hardware-based CABAC decoder with parallel binary arithmetic decoding
US20060294174A1 (en) Hardware-based CABAC decoder
US20020199083A1 (en) High code-density microcontroller architecture with changeable instruction formats
US20090089549A1 (en) H.264 Video Decoder CABAC Core Optimization Techniques
US5663726A (en) High speed variable-length decoder arrangement with reduced memory requirements for tag stream buffering
WO1995019662A1 (en) Data compression apparatus and method
US20090019071A1 (en) Blocking for combinatorial coding/decoding for electrical computers and digital data processing systems
US20020154042A1 (en) Variable-length decoding apparatus and method
US20020163965A1 (en) Video frame compression/decompression hardware system
US5617089A (en) Huffman code decoding circuit
US6069575A (en) Variable length code decoder
US5812074A (en) High speed data syntax parsing apparatus
US6865668B1 (en) Variable-length, high-speed asynchronous decoder circuit

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C14 Granted