WO2012136065A1 - 一种支持多个数据单元任意置换的方法和装置 - Google Patents

一种支持多个数据单元任意置换的方法和装置 Download PDF

Info

Publication number
WO2012136065A1
WO2012136065A1 PCT/CN2011/082944 CN2011082944W WO2012136065A1 WO 2012136065 A1 WO2012136065 A1 WO 2012136065A1 CN 2011082944 W CN2011082944 W CN 2011082944W WO 2012136065 A1 WO2012136065 A1 WO 2012136065A1
Authority
WO
WIPO (PCT)
Prior art keywords
data unit
replacement
thirty
data
data units
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2011/082944
Other languages
English (en)
French (fr)
Inventor
林文琼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
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
Application filed by ZTE Corp filed Critical ZTE Corp
Priority to EP11863214.0A priority Critical patent/EP2696294B1/en
Priority to US14/008,562 priority patent/US9332495B2/en
Publication of WO2012136065A1 publication Critical patent/WO2012136065A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. Transmission Power Control [TPC] or power classes
    • H04W52/02—Power saving arrangements
    • H04W52/0209—Power saving arrangements in terminal devices
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/76—Architectures of general purpose stored program computers
    • G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
    • G06F15/7807—System on chip, i.e. computer system on a single chip; System in package, i.e. computer system on one or more chips in a single package
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06F7/76—Arrangements for rearranging, permuting or selecting data according to predetermined rules, independently of the content of the data
    • G06F7/762—Arrangements for rearranging, permuting or selecting data according to predetermined rules, independently of the content of the data having at least two separately controlled rearrangement levels, e.g. multistage interconnection networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/0003—Software-defined radio [SDR] systems, i.e. systems wherein components typically implemented in hardware, e.g. filters or modulators/demodulators, are implented using software, e.g. by involving an AD or DA conversion stage such that at least part of the signal processing is performed in the digital domain
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the present invention relates to the field of vector processors in chip design, and more particularly to a method and apparatus for supporting arbitrary replacement of multiple data units. Background technique
  • GSM Global System for Mobile Communications
  • UMTS Universal Mobile Telecommunications System
  • WLAN Wireless Local Area Network
  • TD-SCDMA Time Division Synchronous Code Division Multiple Access
  • LTE Long Term Evolution
  • the vector processor In order to support the processing of multi-mode baseband, the vector processor is required to be able to perform operations for G times per second.
  • the vector processor processes a large number of data units each time, typically eight, sixteen or thirty-two.
  • a data unit a data unit of 32 bits, represents a pair of iq floating point data. Then, in order to handle these vector data flexibly, any permutation of the position between the data units in the vector processor is an indispensable operation.
  • the prior art has proposed a scheme for arbitrarily replacing the positions between data units.
  • FIG. 1 in order to realize the replacement of the position of thirty-two data units, thirty-two thirty-two-selectors are required, FIG.
  • the inO, inl, ..., in31 represent thirty-two data units to be replaced, outO, outl, ..., out31 represent the position replacement results of thirty-two data units.
  • Thirty-two data units need to be input into the thirty-two thirty-two-selectors, and each thirty-two-selector is selected from One of the thirty-two data units input is selected, and finally thirty-two data units after position replacement are obtained.
  • the method can achieve arbitrary replacement of the position of the data unit, it can be seen that the number of selectors utilized in the implementation process is large, resulting in a large area occupied by the data unit replacement in the chip design, so that the power consumption is also large.
  • the present invention provides a method for supporting arbitrary replacement of a plurality of data units, and arranging a pattern of arbitrarily replacing N data unit positions; the method further includes:
  • the replacement operation is performed on the N data units according to the configuration of the pattern
  • N represents the number of data units, N is equal to 2 to the power of m, and m is a positive integer.
  • the method is:
  • the pattern of the configuration includes: a configuration of a control line corresponding to the cross selector, and a replacement operation for controlling two data units in the replacement process.
  • the performing four data unit replacement operations on the thirty-two data units is: dividing thirty-two data units into eight groups, each group including four data units, respectively for four data units in each group Both perform the permutation operation.
  • the performing eight data unit replacement operations on the thirty-two data units is: combining eight sets of data units that have performed four data unit replacement operations, forming four sets of eight data unit sequences, and pressing
  • the configured pattern performs eight data unit permutation operations on four sets of data unit sequences, respectively.
  • the performing sixteen data unit replacement operations on the thirty-two data units is: combining four sets of data units that have performed eight data unit replacement operations, forming two sets of sixteen data unit sequences, And sixteen data unit replacement operations are performed on the two sets of data unit sequences according to the configured pattern.
  • the performing thirty-two data unit replacement operations on the thirty-two data units is: combining two sets of data units that have performed sixteen data unit replacement operations to form a set of thirty-two data units Sequence, and performing thirty-two data unit permutation operations on the set of data unit sequences in accordance with the configured pattern.
  • the present invention also provides an apparatus for supporting arbitrary replacement of a plurality of data units, the apparatus comprising: a configuration module and a replacement processing module;
  • the configuration module is configured to configure a pattern of arbitrarily replacing N data unit positions; and the replacement processing module is configured to perform a replacement operation on the N data units according to the pattern configured by the configuration module during the data unit replacement process;
  • N represents the number of data units, N is equal to 2 to the power of m, and m is a positive integer.
  • N 32,
  • the configuration module is configured to configure a pattern of arbitrarily replacing thirty-two data unit positions; the replacement processing module is configured to sequentially perform thirty-two data units according to a pattern configured by the configuration module during the data unit replacement process Four data unit permutations, eight data unit permutations, sixteen data unit permutations, and thirty-two data unit permutations are performed.
  • the replacement processing module further includes: a four data unit replacement module, an eight data unit replacement module, a sixteen data unit replacement module, and a thirty-two data unit replacement module;
  • the four data unit replacement module is configured to perform four data unit replacement operations on thirty-two data units according to the pattern configured by the configuration module, and send the replacement result to the eight data unit replacement module;
  • the eight data unit replacement module is configured to perform eight data unit replacement operations on thirty-two data units sent by the four data unit replacement module according to the pattern configured by the configuration module, and send the replacement result to the sixteen data unit.
  • the sixteen data unit replacement module is configured to perform sixteen data unit replacement operations on thirty-two data units sent by the eight data unit replacement module according to the pattern configured by the configuration module, and send the replacement result to the thirty Two data unit replacement module;
  • the thirty-two data unit replacement module is configured to perform thirty-two data unit replacement operations on the thirty-two data units sent by the sixteen data unit replacement module according to the pattern configured by the configuration module.
  • the method and apparatus for supporting arbitrary replacement of a plurality of data units configure a pattern in which N data units are arbitrarily replaced; in the data unit replacement process, a replacement operation is performed on N data units according to the configuration of the pattern. If any replacement operation is performed on thirty-two data units by using the method of the present invention, a 15th-order switching operation is required, and a total of 382 two-choice selectors are required. Compared with the prior art, the present invention requires two alternatives. The number of devices is much smaller than the existing 992. Therefore, the present invention can reduce the area occupied by the data unit replacement circuit in the chip, and the reduction in the circuit area also reduces the power consumption.
  • 1 is a schematic diagram of a method for implementing an arbitrary replacement of thirty-two data units
  • FIG. 2 is a schematic diagram of an implementation process of a method for supporting arbitrary replacement of thirty-two data units according to an embodiment of the present invention
  • 3 is a schematic flowchart of sorting four data units according to a pattern according to an embodiment of the present invention
  • 4 is a flowchart of four data unit position replacement according to an embodiment of the present invention
  • FIG. 5 is a circuit diagram of a second-selector selector corresponding to each cross-connection in the data unit position replacement according to the present invention
  • FIG. 6 is a flowchart of eight data unit position replacement according to an embodiment of the present invention.
  • FIG. 8 is a flowchart of thirty-two data unit position replacement according to an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of an apparatus for supporting arbitrary replacement of thirty-two data units according to an embodiment of the present invention. detailed description
  • the basic idea of the present invention is to: configure a pattern in which N data unit positions are arbitrarily replaced; in the data unit replacement process, perform a permutation operation on N data units according to the configuration of the pattern.
  • N represents the number of data units
  • N is equal to 2 to the power of m
  • m is a positive integer.
  • the pattern is: a parameter corresponding to the position of the result obtained after the replacement of the position of the data unit of the input vector processor. For example: Set the data units e0, el, e2, and e3 from left to right. The order of the positions to be output is e3, e0, el, and e2. Then the values of the parameters p0, pl, p2, and p3 of the configuration pattern are respectively 1 , 2, 3, 0.
  • the pattern also includes the cross selector required for data exchange, that is, the configuration of the control line corresponding to the selector.
  • the present invention is described by taking thirty-two data units as an example, but the solution of the present invention is not limited to any permutation of thirty-two data unit positions, which can also be analogized to sixty-four or more data units. Any replacement. Of course, the present invention must support any permutation of four, eight, and sixteen data units. The specific implementation method is described later.
  • FIG. 2 is a schematic flowchart of a method for supporting arbitrary replacement of thirty-two data units according to an embodiment of the present invention. As shown in FIG. 2, the implementation process of the method is as follows:
  • Step 201 configuring a pattern of arbitrarily replacing thirty-two data unit positions
  • the pattern of arbitrarily replacing thirty-two data units is configured, it is performed according to actual needs, that is, according to the result obtained by replacing the required position of thirty-two data units.
  • the left side of Fig. 3 represents the replacement pattern, and the right side is the data unit position replacement flow actually input to the vector processor, wherein the horizontal arrows indicate the control lines.
  • the configuration method of the pattern of arbitrarily replacing thirty-two data unit positions is the same as the configuration method of the pattern of arbitrarily replacing the four data unit positions, mainly for configuring the control lines of each cross selector for controlling two data in the replacement process.
  • the thirty-two data unit position replacement is completed in four steps, and each step corresponds to a corresponding pattern.
  • the specific form of the pattern is not described in detail herein, and can be compared with the flowchart of the subsequent step data unit replacement.
  • Step 202 Perform four data unit replacement operations on thirty-two data units according to the configuration of the pattern;
  • the thirty-two data units are first divided into eight groups, each group including four data units, and the replacement operations are performed on each of the four data units in each group.
  • thirty-two data units are: a0,al,a2,...,a31, each four data units are grouped in order, the first group is a0,al,a2,a3,
  • the two data positions are interchanged according to the control of the control line of the cross selector in the pattern, and the replacement operations for a0, a, a2 and a3 are realized, and the other seven groups perform the replacement operation in the same way.
  • FIG. 4 is a flow chart of four data unit position replacement according to an embodiment of the present invention.
  • the replacement of four data units requires a third-order exchange operation, and the exchange operations between the two data units are performed in the corresponding patterns.
  • the control line of the cross selector is controlled by the control line of the two selectors. It can also be seen from Fig. 4 that the replacement of four data units requires 5x2 cross selectors, that is, 10 two-select selectors, and then eight groups, that is, thirty-two data units, require a total of 80 two-selection. A selector.
  • FIG. 5 is a circuit diagram of a second-selector selector corresponding to each cross-connection in the data unit position replacement according to the present invention, wherein the inO and inl are two data units before the position exchange, and the two positions after the outO and outl positions are exchanged.
  • the data unit, ctl is the control line, and the two-selector determines whether to swap the positions of the two data units under the control of the control line.
  • each of the intersecting lines formed corresponds to two alternative selectors, thereby obtaining the number of alternative selectors to be used in each step.
  • Step 203 Perform eight data unit replacement operations on thirty-two data units according to the configuration of the pattern
  • the eight data units in the two sets of four data unit replacement operations that have been completed in step 202 are replaced by a sequence of eight data units, that is, the eight sets of data units in the replacement operation in step 202 are combined. , forming four sets of eight data unit sequences, according to the configuration of the pattern Eight data unit replacement operations are performed on the four sets of data units, respectively.
  • FIG. 6 is a flowchart of eight data unit position replacement according to an embodiment of the present invention.
  • the data units b0, bl, b2, b3, b4, b5, b6, and b7 are replaced, where b0 ⁇ b31 are used.
  • the purpose of the thirty-two data units is to distinguish between thirty-two data units that have not been subjected to any positional replacement in step 202, and perform four operations on thirty-two data units of a0, a1, a2, ..., a31.
  • the result of the data unit replacement operation is represented by b0, bl, b2, ..., b31.
  • the replacement of the eight data units also needs to perform the third-order switching operation. It can also be seen that the replacement of the eight data units requires a total of 9x2 cross selectors, that is, 18 two-select selectors, then four. The group, that is, thirty-two data units, requires a total of 72 two-selectors.
  • Step 204 Perform sixteen data unit replacement operations on thirty-two data units according to the configuration of the pattern
  • the sixteen data units in the two sets of eight data unit replacement operations that have been completed in step 203 are replaced by a pattern into sixteen data unit sequences, that is, four sets of data units that are subjected to the replacement operation in step 203.
  • the two combine to form two sets of sixteen data unit sequences, and perform sixteen data unit replacement operations according to the configuration of the pattern.
  • FIG. 7 is a flowchart of sixteen data unit position replacement according to an embodiment of the present invention.
  • the data unit C 0,cl, c 2, . . . , cl5 is replaced, where c0 ⁇ c31 is used to indicate the The purpose of the thirty-two data units is to distinguish from thirty-two data units that perform eight data unit position permutations in step 203, and perform eight operations on b0, bl, b2, ..., b31 thirty-two data units.
  • the result of the data unit replacement operation is represented by c0, cl, c2, ..., c31.
  • Fig. 7 It can be seen from Fig. 7 that the replacement of sixteen data units also needs to perform the fourth-order exchange operation. It can also be seen that the replacement of sixteen data units requires a total of 25x2 cross selectors, that is, 50 two-select selectors. Then, two groups, that is, thirty-two data units, require a total of 100 two-selectors.
  • Step 205 Perform thirty-two data unit replacement operations on thirty-two data units according to the configuration of the pattern; Specifically, the thirty-two data units of the two sets of sixteen data unit replacement operations in step 204 are replaced by a pattern into thirty-two data unit sequences, that is, two groups of the replacement operations in step 204. The data units are combined to form a set of thirty-two data unit sequences, and thirty-two data unit replacement operations are performed in accordance with the configuration of the pattern.
  • FIG. 8 is a flowchart of the positional replacement of thirty-two data units according to an embodiment of the present invention.
  • the data units d0, dl, d2, . . . , d31 are replaced, where the three are represented by d0 ⁇ d31.
  • the purpose of the twelve data units is to distinguish thirty-two data units that perform sixteen data unit position permutations in step 204, and perform thirty-two data units on c0, cl, c2, ..., c31.
  • the results of the six data unit replacement operations are represented by d0, dl, d2, ..., d31.
  • the delay of the two-selection selector is about 0.2 ns. If it needs to work at 500 Mhz, then any replacement operation of thirty-two data units takes two clock cycles to complete, and the speed is very high. fast.
  • the clock cycle is 2 ns, so the 3 ns execution time is divided into two cycles, each cycle being 1.5 ns, so the present invention can be completed in two cycles at a 500 MHz clock.
  • the present invention also provides an apparatus for supporting arbitrary replacement of a plurality of data units, including: a configuration module and a replacement processing module;
  • the configuration module is configured to configure a pattern in which the N data unit positions are arbitrarily replaced.
  • the replacement processing module is configured to perform a replacement operation on the N data units according to the pattern configured by the configuration module during the data unit replacement process.
  • the device includes: a configuration module and a replacement processing module;
  • the configuration module is configured to configure a pattern of arbitrarily replacing thirty-two data unit positions; the replacement processing module is configured to sequentially perform thirty-two data units according to a pattern configured by the configuration module during the data unit replacement process Four data unit permutations, eight data unit permutations, sixteen data unit permutations, and thirty-two data unit permutations are performed.
  • the configuration module may be implemented by software or by hardware, and is preferably implemented by software, which can reduce the area occupied by the device in the chip.
  • the replacement processing module further includes: a four data unit replacement module, an eight data unit replacement module, a sixteen data unit replacement module, and a thirty-two data unit replacement module;
  • the four data unit replacement module is configured to perform four data unit replacement operations on thirty-two data units according to the pattern configured by the configuration module, and send the replacement result to the eight data unit replacement module;
  • the eight data unit replacement module is configured to perform eight data unit replacement operations on thirty-two data units sent by the four data unit replacement module according to the pattern configured by the configuration module, and send the replacement result to the sixteen data unit.
  • the sixteen data unit replacement module is configured to perform sixteen data unit replacement operations on thirty-two data units sent by the eight data unit replacement module according to the pattern configured by the configuration module, And transmitting the replacement result to the thirty-two data unit replacement module;
  • the thirty-two data unit replacement module is configured to perform thirty-two data unit replacement operations on the thirty-two data units sent by the sixteen data unit replacement module according to the pattern configured by the configuration module.
  • the inventive arrangement can be widely used for arbitrarily changing the position of vector data units in a vector processor, or arbitrarily replacing the positions of a plurality of data units in parallel in other applications.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Design And Manufacture Of Integrated Circuits (AREA)
  • Logic Circuits (AREA)

Abstract

本发明公开了一种支持多个数据单元任意置换的方法，包括：配置N个数据单元任意置换的图样；在数据单元置换过程中，根据图样的配置对N个数据单元执行置换操作。本发明还同时公开了一种支持多个数据单元任意置换的装置，运用该方法和装置可减小数据单元置换电路在芯片中占用的面积，同时减小功耗。

Description

一种支持多个数据单元任意置换的方法和装置 技术领域
本发明涉及芯片设计中的矢量处理器领域， 尤其涉及一种支持多个数 据单元任意置换的方法和装置。 背景技术
目前， 通信协议包括全球移动通讯系统 （GSM )、 通用移动通信系统 ( UMTS )、 无线局域网络（ WLAN )、 时分同步码分多址（ TD-SCDMA ) 和长期演进（LTE ) 等等多种模式。 如果采用原有的特定应用集成电路 ( ASIC, Application Specific Integrated Circuit )方法来实现多模兼容的移动 终端的芯片， 必然面临芯片面积大、 功耗大且缺乏灵活性等缺点。 当前提 出的软件无线电（SDR, Software-Defined Radio )技术正是解决多模通信移 动终端芯片设计问题极具潜力的技术， 可编程矢量处理器为 SDR的核心架 构。
为了支持多模基带的处理， 要求矢量处理器每秒必须能执行数 G次的 操作， 矢量处理器每次处理的数据单元的数目很大， 一般为八个、 十六个 或三十二个数据单元， 一个数据单元为 32bits, 代表一对 iq浮点数据。 那 么， 为了灵活处理这些矢量数据， 矢量处理器内的数据单元间位置的任意 置换是必不可少的操作。
现有技术已提出数据单元间的位置任意置换的方案， 如图 1 所示， 为 了实现对三十二个数据单元位置的置换， 需要三十二个三十二选一的选择 器， 图 1 中 inO, inl,...,in31表示位置待置换的三十二个数据单元， outO, outl,...,out31表示三十二个数据单元的位置置换结果。三十二个数据单元需 分别输入到三十二个三十二选一的选择器内， 每个三十二选一的选择器从 输入的三十二个数据单元中选出一个数据单元， 最终得到位置置换后的三 十二个数据单元。 如果将所述三十二选一的选择器折合成二选一选择器， 则该方法共需要 32x31=992个二选一选择器。 虽然该方法能实现数据单元 位置的任意置换， 但是可以看出其实现过程所利用的选择器的数目较大， 导致在芯片设计中数据单元置换电 占用很大的面积， 使得功耗也较大。 发明内容
有鉴于此， 本发明的主要目的在于提供一种支持多个数据单元任意置 换的方法和装置， 可减小数据单元置换电路在芯片中占用的面积， 同时减 小功耗。
为达到上述目的， 本发明的技术方案是这样实现的：
本发明提供了一种支持多个数据单元任意置换的方法， 配置 N个数据 单元位置任意置换的图样； 该方法还包括：
在数据单元置换过程中， 根据图样的配置对 N个数据单元执行置换操 作；
其中， 所述 N表示数据单元的个数， N等于 2的 m次幂， m为正整数。 其中， 所述 N=32时， 该方法为：
配置三十二个数据单元位置任意置换的图样；
在数据单元置换过程中， 根据图样的配置对三十二个数据单元依次执 行四个数据单元置换、 八个数据单元置换、 十六个数据单元置换以及三十 二个数据单元置换的操作。
其中， 所述配置的图样包括： 交叉选择器对应的控制线的配置， 用于 置换过程中控制两个数据单元的置换操作。
其中， 所述对三十二个数据单元执行四个数据单元置换操作， 为： 将三十二个数据单元分成八组， 每组包括四个数据单元， 分别对各组 中的四个数据单元均执行置换操作。 其中， 所述对三十二个数据单元执行八个数据单元置换操作， 为： 将已执行四个数据单元置换操作的八组数据单元两两结合， 形成四组 八个数据单元序列， 并按已配置的图样对四组数据单元序列分别执行八个 数据单元置换操作。
其中， 所述对三十二个数据单元执行十六个数据单元置换操作， 为： 将已执行八个数据单元置换操作的四组数据单元两两结合， 形成两组 十六个数据单元序列， 并按已配置的图样对两组数据单元序列分别执行十 六个数据单元置换操作。
其中， 所述对三十二个数据单元执行三十二个数据单元置换操作， 为： 将已执行十六个数据单元置换操作的两组数据单元相结合， 形成一组 三十二个数据单元序列， 并按已配置的图样对该组数据单元序列执行三十 二个数据单元置换操作。
本发明还提供了一种支持多个数据单元任意置换的装置， 该装置包括: 配置模块和置换处理模块； 其中，
所述配置模块， 用于配置 N个数据单元位置任意置换的图样； 所述置换处理模块， 用于在数据单元置换过程中， 根据配置模块所配 置的图样对 N个数据单元执行置换操作；
其中， 所述 N表示数据单元的个数， N等于 2的 m次幂， m为正整数。 其中， 所述 N=32时，
所述配置模块， 用于配置三十二个数据单元位置任意置换的图样； 所述置换处理模块， 用于在数据单元置换过程中， 根据配置模块所配 置的图样对三十二个数据单元依次执行四个数据单元置换、 八个数据单元 置换、 十六个数据单元置换以及三十二个数据单元置换的操作。
其中， 所述置换处理模块还包括： 四数据单元置换模块、 八数据单元 置换模块、 十六数据单元置换模块和三十二数据单元置换模块； 其中， 所述四数据单元置换模块， 用于根据配置模块所配置的图样对三十二 个数据单元执行四个数据单元置换操作， 并将置换结果发送给八数据单元 置换模块；
所述八数据单元置换模块， 用于根据配置模块所配置的图样对四数据 单元置换模块所发的三十二个数据单元执行八个数据单元置换操作， 并将 置换结果发送给十六数据单元置换模块；
所述十六数据单元置换模块， 用于根据配置模块所配置的图样对八数 据单元置换模块所发的三十二个数据单元执行十六个数据单元置换操作， 并将置换结果发送给三十二数据单元置换模块；
所述三十二数据单元置换模块， 用于根据配置模块所配置的图样对十 六数据单元置换模块所发的三十二个数据单元执行三十二个数据单元置换 操作。
本发明提供的支持多个数据单元任意置换的方法和装置， 配置 N个数 据单元任意置换的图样； 在数据单元置换过程中， 根据图样的配置对 N个 数据单元执行置换操作。 如果运用本发明方法对三十二个数据单元执行任 意置换操作， 则需执行 15阶交换操作， 共需 382个二选一选择器， 与现有 技术相比， 本发明所需二选一选择器的数目远小于现有的 992个， 因此， 本发明可减小数据单元置换电路在芯片中占用的面积， 电路面积的减小也 相应减小了功率的消耗。
此外， 本发明的置换操作仅需两个时钟周期， 速度较快。 附图说明
图 1为现有三十二个数据单元任意置换的方法实现示意图；
图 2 为本发明实施例支持三十二个数据单元任意置换的方法实现流程 示意图；
图 3为本发明实施例四个数据单元依据图样排序的流程示意图； 图 4为本发明实施例四个数据单元位置置换的流程图；
图 5 为本发明数据单元位置置换时每个交叉连线对应的二选一选择器 电路图；
图 6为本发明实施例八个数据单元位置置换的流程图；
图 7为本发明实施例十六个数据单元位置置换的流程图；
图 8为本发明实施例三十二个数据单元位置置换的流程图；
图 9为本发明实施例支持三十二个数据单元任意置换的装置结构示意 图。 具体实施方式
本发明的基本思想是： 配置 N个数据单元位置任意置换的图样； 在数 据单元置换过程中， 根据图样的配置对 N个数据单元执行置换操作。
其中， 所述 N表示数据单元的个数， N等于 2的 m次幂， m为正整数。 以三十二个数据单元为例， 那么， 在数据单元置换过程中， 将根据图样的 配置对三十二个数据单元依次执行四个数据单元置换、 八个数据单元置换、 十六个数据单元置换以及三十二个数据单元置换的操作。
其中， 所述图样为： 输入矢量处理器的数据单元的位置经置换后最终 所得结果位置对应的参数。 例如： 设从左到右输入 e0,el,e2,e3 四个数据单 元， 需要输出的位置顺序为 e3,e0,el,e2, 那么配置图样的参数 p0,pl,p2,p3 的值分别为 1 , 2, 3 , 0。 当然， 图样中还包括数据交换时所需的交叉选择 器， 即二选一选择器对应的控制线的配置。
本发明是以三十二个数据单元为例进行描述， 但是本发明的方案并不 限于三十二个数据单元位置的任意置换， 其也可以类推到六十四个或更多 个数据单元的任意置换。 当然， 本发明一定支持四个、 八个和十六个数据 单元的任意置换， 具体实现方法见后续描述。
下面结合附图及具体实施例对本发明作进一步详细说明。 图 2 为本发明实施例支持三十二个数据单元任意置换的方法实现流程 示意图， 如图 2所示， 该方法的实现流程如下：
步驟 201 : 配置三十二个数据单元位置任意置换的图样；
具体为： 在配置三十二个数据单元任意置换的图样时， 依据实际需要 来进行， 即依据所需三十二个数据单元位置置换最终所得的结果进行配置。
例如： 以四个数据单元为例， 假设我们需要将输入矢量处理器中的先 后位置为 e0,el,e2,e3的四个数据单元置换为先后位置为 e3,el,e2,e0的顺序， 那么在配置图样时， 设图样对应的初始四个参数表示为 p0,pl,p2,p3 , 其对 应的数值分别设为： 3 , 1 , 2 , 0, 即以输入数据的输出位置配置图样； 数 值 3 , 1 , 2, 0进行两两比较大小， 按由小到大的顺序排列， 最后所得结果 必为 0, 1 , 2, 3 , 在两两比较时， 形成每个交叉选择器对应的控制线 ctl, 可设 ctl=l时， 表示两数据交换位置； ctl=0时， 表示两数据位置不变。 如图 3所示， 图 3中左侧表示置换图样， 右侧为实际输入到矢量处理器中的数据 单元位置置换流程， 其中横向箭头均表示控制线。 左侧 p0=3 , pl=l , p2=2 , p3=0四个参数排序共需进行三阶交换， 第一阶排序时， 因 ρθρΐ , ρ2>ρ3 , 所以对应交叉选择器的控制线分别为： ctl=l 和 ctl=l ; 第二阶排序时， 因 pl>p3 , p0>p2 , 所以对应交叉选择器的控制线分别为： ctl=l和 ctl=l ; 第三 阶排序时， 因 p2>pl , 所以对应交叉选择器的控制线为： ctl=l。 因此， 实际 输入的四个数据单元 e0,el,e2,e3 进行位置置换时， 每阶段的交换将在左侧 图样形成的控制线的控制下进行， 所得结果如图 3右侧所示。
三十二个数据单元位置任意置换的图样的配置方法与上述四个数据单 元位置任意置换的图样的配置方法相同， 主要是配置各交叉选择器的控制 线， 用于置换过程中控制两个数据单元的位置置换操作。 关于三十二个数 据单元位置置换分四步完成， 每一步对应相应的图样， 关于图样的具体形 式此处不再详述， 可对照后续步驟数据单元置换时的流程图。 步驟 202:根据图样的配置对三十二个数据单元执行四个数据单元置换 操作；
具体为： 首先将三十二个数据单元分成八组， 每组包括四个数据单元， 分别对各组中的四个数据单元均执行置换操作。 例如： 假设三十二个数据 单元分别为： a0,al,a2,...,a31 , 按先后顺序将每四个数据单元分为一组， 第 一组为 a0,al,a2,a3 ,按照已配置的图样， 即按图样中交叉选择器的控制线的 控制进行两数据位置互换， 实现对 a0,al,a2和 a3的置换操作， 其它七组按 同样的方法执行置换操作。
图 4为本发明实施例四个数据单元位置置换的流程图， 如图 4所示， 四个数据单元的置换共需执行三阶交换操作， 两个数据单元间的交换操作 均由对应图样中交叉选择器的控制线， 也就是由二选一选择器的控制线控 制。 从图 4中还可以看出， 四个数据单元的置换共需 5x2个交叉选择器， 即 10个二选一选择器， 那么八组， 即三十二个数据单元则共需 80个二选 一选择器。
图 5 为本发明数据单元位置置换时每个交叉连线对应的二选一选择器 电路图， 其中， 所述 inO和 inl为位置交换前的两个数据单元， outO和 outl 位置交换后的两个数据单元， ctl为控制线， 二选一选择器在控制线的控制 下确定是否交换两个数据单元的位置。 如图 5 所示， 两个数据单元交换位 置时， 形成的每个交叉线对应两个二选一选择器， 由此可得出各步驟共需 采用的二选一选择器的数目。
步驟 203 :根据图样的配置对三十二个数据单元执行八个数据单元置换 操作；
具体为： 将步驟 202 中两组已完成四个数据单元置换操作的八个数据 单元按图样置换成一个八个数据单元序列， 也就是将步驟 202 中经置换操 作的八组数据单元两两结合， 形成四组八个数据单元序列， 按图样的配置 对四组数据单元分别执行八个数据单元置换操作。
图 6为本发明实施例八个数据单元位置置换的流程图， 如图 6所示， 对数据单元 b0,bl,b2,b3,b4,b5,b6,b7进行置换，这里用 b0~b31表示该三十二 个数据单元的目的是与步驟 202 中未执行任何位置置换的三十二个数据单 元进行区分，对 a0,al,a2,...,a31三十二个数据单元执行四个数据单元置换操 作后的结果用 b0,bl,b2, ...,b31表示。
从图 6 中可以看出八个数据单元的置换同样需执行三阶交换操作， 还 可以看出， 八个数据单元的置换共需 9x2交叉选择器， 即 18个二选一选择 器， 那么四组， 即三十二个数据单元则共需 72个二选一选择器。
步驟 204:根据图样的配置对三十二个数据单元执行十六个数据单元置 换操作；
具体为： 将步驟 203 中两组已完成八个数据单元置换操作的十六个数 据单元按图样置换成一个十六个数据单元序列， 也就是将步驟 203 中经置 换操作的四组数据单元两两结合， 形成两组十六个数据单元序列， 按图样 的配置执行十六个数据单元置换操作。
图 7为本发明实施例十六个数据单元位置置换的流程图， 如图 7所示， 对数据单元 C0,cl,c2,...,cl5进行置换， 这里用 c0~c31表示该三十二个数据 单元的目的是与步驟 203 中执行八个数据单元位置置换的三十二个数据单 元进行区分， 对 b0,bl,b2,...,b31 三十二个数据单元执行八个数据单元置换 操作后的结果用 c0,cl,c2, ...,c31表示。
从图 7 中可以看出十六个数据单元的置换同样需执行四阶交换操作， 还可以看出， 十六个数据单元的置换共需 25x2个交叉选择器， 即 50个二 选一选择器， 那么两组， 即三十二个数据单元则共需 100个二选一选择器。
步驟 205:根据图样的配置对三十二个数据单元执行三十二个数据单元 置换操作； 具体为： 将步驟 204 中两组已完成十六个数据单元置换操作的三十二 个数据单元按图样置换成一个三十二个数据单元序列， 也就是将步驟 204 中经置换操作的两组数据单元相结合， 形成一组三十二个数据单元序列， 按图样的配置执行三十二个数据单元置换操作。
图 8为本发明实施例三十二个数据单元位置置换的流程图， 如图 8所 示， 对数据单元 d0,dl,d2,...,d31进行置换， 这里用 d0~d31表示该三十二个 数据单元的目的是与步驟 204 中执行十六个数据单元位置置换的三十二个 数据单元进行区分，对 c0,cl,c2,...,c31三十二个数据单元执行十六个数据单 元置换操作后的结果用 d0,dl,d2, ...,d31表示。
从图 8中可以看出三十二个数据单元的置换同样需执行五阶交换操作， 还可以看出， 三十二个数据单元的置换共需 65x2个交叉选择器， 即 130个 二选一选择器。
从上述步驟 202到步驟 205可以得出， 本发明三十二个数据单元的置 换共需 80+72+100+130=382个二选一选择器， 整个置换过程中共需执行 3+ 3+4+5=15阶交换操作。 如果推广开来， N个数据单元则共需执行 log22+log 24+log28+...+log2N阶的交换操作其中， 所述 N等于 2的 m次幂， m为正整 数。
此外， 本发明每阶交换操作中， 二选一选择器的延时大约为 0.2ns, 如 果需要工作在 500Mhz, 那么三十二个数据单元的任意置换操作则需两个时 钟周期完成， 速度非常快。
这里，对于一个二选一选择器，在 65ns的 TSMC低功耗库中的延时一 般小于 0.2ns, 而本发明共执行 15阶交换操作， 所以延时为 15x0.2=3ns, 而 500MHz的时钟周期是 2ns, 所以 3ns执行时间分成两个周期， 每个周期则 为 1.5ns, 所以本发明在 500MHz时钟下两个周期可以做完。
从现有技术可知， 现有实现方法对三十二个数据单元执行置换操作时 , 需要 992个二选一选择器， 远大于本发明所需二选一选择器的数目， 可见， 本发明可减小数据单元置换电路在芯片中占用的面积， 并相应减小功耗。
本发明还提供了一种支持多个数据单元任意置换的装置， 包括： 配置 模块和置换处理模块； 其中，
所述配置模块， 用于配置 N个数据单元位置任意置换的图样； 所述置换处理模块， 用于在数据单元置换过程中， 根据配置模块所配 置的图样对 N个数据单元执行置换操作。
下面以 N=32为例对本发明装置进行详细描述，如图 9所示，该装置包 括： 配置模块和置换处理模块； 其中，
所述配置模块， 用于配置三十二个数据单元位置任意置换的图样； 所述置换处理模块， 用于在数据单元置换过程中， 根据配置模块所配 置的图样对三十二个数据单元依次执行四个数据单元置换、 八个数据单元 置换、 十六个数据单元置换以及三十二个数据单元置换的操作。
本发明中， 所述配置模块可通过软件实现， 也可由硬件实现， 当然优 选通过软件实现， 这样可减小本装置在芯片中占用的面积。
所述置换处理模块还包括： 四数据单元置换模块、 八数据单元置换模 块、 十六数据单元置换模块以及三十二数据单元置换模块； 其中，
所述四数据单元置换模块， 用于根据配置模块所配置的图样对三十二 个数据单元执行四个数据单元置换操作， 并将置换结果发送给八数据单元 置换模块；
所述八数据单元置换模块， 用于根据配置模块所配置的图样对四数据 单元置换模块所发的三十二个数据单元执行八个数据单元置换操作， 并将 置换结果发送给十六数据单元置换模块；
所述十六数据单元置换模块， 用于根据配置模块所配置的图样对八数 据单元置换模块所发的三十二个数据单元执行十六个数据单元置换操作， 并将置换结果发送给三十二数据单元置换模块；
所述三十二数据单元置换模块， 用于根据配置模块所配置的图样对十 六数据单元置换模块所发的三十二个数据单元执行三十二个数据单元置换 操作。
本发明方案可以广泛用于矢量处理器中的矢量数据单元的位置任意置 换、 或其它应用中并行的多个数据单元的位置任意置换。
以上所述， 仅为本发明的较佳实施例而已， 并非用于限定本发明的保 护范围。

Claims

权利要求书
1、 一种支持多个数据单元任意置换的方法， 其特征在于， 配置 N个数 据单元位置任意置换的图样； 该方法还包括：
在数据单元置换过程中， 根据图样的配置对 N个数据单元执行置换操 作；
其中， 所述 N表示数据单元的个数， N等于 2的 m次幂， m为正整数。
2、 根据权利要求 1所述的支持多个数据单元任意置换的方法， 其特征 在于， 所述 N=32时， 该方法为：
配置三十二个数据单元位置任意置换的图样；
在数据单元置换过程中， 根据图样的配置对三十二个数据单元依次执 行四个数据单元置换、 八个数据单元置换、 十六个数据单元置换以及三十 二个数据单元置换的操作。
3、 根据权利要求 1或 1所述的支持多个数据单元任意置换的方法， 其 特征在于， 所述配置的图样包括： 交叉选择器对应的控制线的配置， 用于 置换过程中控制两个数据单元的置换操作。
4、 根据权利要求 2所述的支持多个数据单元任意置换的方法， 其特征 在于， 所述对三十二个数据单元执行四个数据单元置换操作， 为：
将三十二个数据单元分成八组， 每组包括四个数据单元， 分别对各组 中的四个数据单元均执行置换操作。
5、 根据权利要求 2所述的支持多个数据单元任意置换的方法， 其特征 在于， 所述对三十二个数据单元执行八个数据单元置换操作， 为：
将已执行四个数据单元置换操作的八组数据单元两两结合， 形成四组 八个数据单元序列， 并按已配置的图样对四组数据单元序列分别执行八个 数据单元置换操作。
6、 根据权利要求 2所述的支持多个数据单元任意置换的方法， 其特征 在于， 所述对三十二个数据单元执行十六个数据单元置换操作， 为： 将已执行八个数据单元置换操作的四组数据单元两两结合， 形成两组 十六个数据单元序列， 并按已配置的图样对两组数据单元序列分别执行十 六个数据单元置换操作。
7、 根据权利要求 2所述的支持多个数据单元任意置换的方法， 其特征 在于， 所述对三十二个数据单元执行三十二个数据单元置换操作， 为： 将已执行十六个数据单元置换操作的两组数据单元相结合， 形成一组 三十二个数据单元序列， 并按已配置的图样对该组数据单元序列执行三十 二个数据单元置换操作。
8、一种支持多个数据单元任意置换的装置，其特征在于，该装置包括： 配置模块和置换处理模块； 其中，
所述配置模块， 用于配置 N个数据单元位置任意置换的图样； 所述置换处理模块， 用于在数据单元置换过程中， 根据配置模块所配 置的图样对 N个数据单元执行置换操作；
其中， 所述 N表示数据单元的个数， N等于 2的 m次幂， m为正整数。
9、 根据权利要求 8所述的支持多个数据单元任意置换的装置， 其特征 在于， 所述 N=32时，
所述配置模块， 用于配置三十二个数据单元位置任意置换的图样； 所述置换处理模块， 用于在数据单元置换过程中， 根据配置模块所配 置的图样对三十二个数据单元依次执行四个数据单元置换、 八个数据单元 置换、 十六个数据单元置换以及三十二个数据单元置换的操作。
10、 根据权利要求 9所述的支持多个数据单元任意置换的装置， 其特 征在于， 所述置换处理模块还包括： 四数据单元置换模块、 八数据单元置 换模块、 十六数据单元置换模块和三十二数据单元置换模块； 其中，
所述四数据单元置换模块， 用于根据配置模块所配置的图样对三十二 个数据单元执行四个数据单元置换操作， 并将置换结果发送给八数据单元 置换模块；
所述八数据单元置换模块， 用于根据配置模块所配置的图样对四数据 单元置换模块所发的三十二个数据单元执行八个数据单元置换操作， 并将 置换结果发送给十六数据单元置换模块；
所述十六数据单元置换模块， 用于根据配置模块所配置的图样对八数 据单元置换模块所发的三十二个数据单元执行十六个数据单元置换操作， 并将置换结果发送给三十二数据单元置换模块；
所述三十二数据单元置换模块， 用于根据配置模块所配置的图样对十 六数据单元置换模块所发的三十二个数据单元执行三十二个数据单元置换 操作。
PCT/CN2011/082944 2011-04-07 2011-11-25 一种支持多个数据单元任意置换的方法和装置 Ceased WO2012136065A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP11863214.0A EP2696294B1 (en) 2011-04-07 2011-11-25 Method and device of supporting arbitrary replacement among multiple data units
US14/008,562 US9332495B2 (en) 2011-04-07 2011-11-25 Method and device of supporting arbitrary replacement of multiple data units

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201110086318.2A CN102737007B (zh) 2011-04-07 2011-04-07 一种支持多个数据单元任意置换的方法和装置
CN201110086318.2 2011-04-07

Publications (1)

Publication Number Publication Date
WO2012136065A1 true WO2012136065A1 (zh) 2012-10-11

Family

ID=46968571

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/082944 Ceased WO2012136065A1 (zh) 2011-04-07 2011-11-25 一种支持多个数据单元任意置换的方法和装置

Country Status (4)

Country Link
US (1) US9332495B2 (zh)
EP (1) EP2696294B1 (zh)
CN (1) CN102737007B (zh)
WO (1) WO2012136065A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109614365B (zh) * 2018-11-14 2022-12-27 青岛海信日立空调系统有限公司 Hbs芯片数据接收兼容电路
CN109582635B (zh) * 2018-11-14 2022-12-27 青岛海信日立空调系统有限公司 Hbs芯片数据发送兼容电路
CN110320957B (zh) * 2019-08-05 2022-01-07 北京中科银河芯科技有限公司 一种电压选择电路

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1558588A (zh) * 2004-01-20 2004-12-29 海信集团有限公司 可重构密码协处理器的可重构置换模块的设计方法
WO2007015941A2 (en) * 2005-07-26 2007-02-08 Invensys Systems, Inc. Dead band tiltering of data to database
CN101453292A (zh) * 2007-11-30 2009-06-10 大唐移动通信设备有限公司 信道交织方法及装置
US20090254718A1 (en) * 2007-10-31 2009-10-08 Texas Instruments Incorporated Local Memories with Permutation Functionality for Digital Signal Processors
CN101874354A (zh) * 2007-11-26 2010-10-27 摩托罗拉公司 用于矢量化turbo解码器的数据交织电路和方法

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4078260A (en) * 1976-05-12 1978-03-07 International Business Machines Corporation Apparatus for transposition sorting of equal length records in overlap relation with record loading and extraction
JP3344383B2 (ja) * 1999-10-04 2002-11-11 日本電気株式会社 スケジューラ
US6622242B1 (en) * 2000-04-07 2003-09-16 Sun Microsystems, Inc. System and method for performing generalized operations in connection with bits units of a data word
CA2375058A1 (en) * 2000-05-05 2001-11-22 Ruby B. Lee A method and system for performing permutations using permutation instructions based on modified omega and flip stages
JP2003007065A (ja) * 2001-06-21 2003-01-10 Nec Microsystems Ltd データ記憶回路、データ処理装置
US7631025B2 (en) * 2001-10-29 2009-12-08 Intel Corporation Method and apparatus for rearranging data between multiple registers
US6990552B2 (en) * 2002-10-31 2006-01-24 Mosaid Technologies, Inc. Sorting method and apparatus using a CAM
DE112004000026D2 (de) * 2003-04-04 2006-06-14 Pact Xpp Technologies Ag Verfahren und Vorrichtung für die Datenverarbeitung
EP1769391A1 (en) * 2004-07-08 2007-04-04 Asocs Ltd. A method of and apparatus for implementing fast orthogonal transforms of variable size
JP4911223B2 (ja) * 2007-06-06 2012-04-04 富士通株式会社 中継装置および端末装置
US20090106525A1 (en) * 2007-10-23 2009-04-23 David Arnold Luick Design structure for scalar precision float implementation on the "w" lane of vector unit
JP2011023975A (ja) * 2009-07-15 2011-02-03 Seiko Epson Corp データ転送装置、データ転送プログラム、データ転送システム、画像表示システム及びデータ転送方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1558588A (zh) * 2004-01-20 2004-12-29 海信集团有限公司 可重构密码协处理器的可重构置换模块的设计方法
WO2007015941A2 (en) * 2005-07-26 2007-02-08 Invensys Systems, Inc. Dead band tiltering of data to database
US20090254718A1 (en) * 2007-10-31 2009-10-08 Texas Instruments Incorporated Local Memories with Permutation Functionality for Digital Signal Processors
CN101874354A (zh) * 2007-11-26 2010-10-27 摩托罗拉公司 用于矢量化turbo解码器的数据交织电路和方法
CN101453292A (zh) * 2007-11-30 2009-06-10 大唐移动通信设备有限公司 信道交织方法及装置

Also Published As

Publication number Publication date
EP2696294B1 (en) 2019-11-13
EP2696294A4 (en) 2017-03-01
EP2696294A1 (en) 2014-02-12
US9332495B2 (en) 2016-05-03
CN102737007A (zh) 2012-10-17
US20140022971A1 (en) 2014-01-23
CN102737007B (zh) 2015-01-28

Similar Documents

Publication Publication Date Title
Lin et al. Wireless network cloud: Architecture and system requirements
EP1415399B1 (en) Adaptive integrated circuitry with heterogeneous and reconfigurable matrices of diverse and adaptive computational units having fixed, application specific computational elements
US20120173864A1 (en) Flexible multi-processing system
US20090037692A1 (en) Adaptive integrated circuitry with heterogeneous and reconfigurable matrices of diverse and adaptive computational units having fixed, application specific computational elements
GB2459939A (en) DMA engine for high bandwidth real-time critical applications
JP4712090B2 (ja) 物理層に再設定可能な処理エンジンを備えたワイヤレス通信装置
WO2012136065A1 (zh) 一种支持多个数据单元任意置换的方法和装置
JP5264770B2 (ja) 異なる通信標準間のハンドオフのシステムおよび方法
Wu et al. Accelerating collective communications with mutual benefits of optical rackless DC and in-network computing
JP4856053B2 (ja) コンピュータ・プロセッサ・アレイ
EP2051435B1 (en) High-efficient encryption and decryption processing method for implementing SMS4 algorithm
CN102508815B (zh) 一种数据处理方法和系统
US20090132794A1 (en) Method and apparatus for performing complex calculations in a multiprocessor array
CN109039608A (zh) 一种基于双S核的8-bitAES电路
Yang et al. Network-on-chip for turbo decoders
CN113647023A (zh) 实现宽带和高效的5g发射机的带符号的rfdac架构
CN108738035B (zh) 一种多制式基带芯片的数据处理方法及装置、处理设备
CN116360852A (zh) 电子装置、存储地址产方法及终端设备
US8694874B2 (en) Circuit and method for parallel perforation in rate matching
CN105847191B (zh) 一种基于重排序算法的交换设备
CN104123959B (zh) 置换规则配置简洁的通用置换电路结构
CN100438345C (zh) 一种交织装置
CN100596136C (zh) 一种数据通道分组选择方法及装置
CN101272391A (zh) 基于以太网交换的软件无线电系统
US9270397B2 (en) Cascaded communication of serialized data streams through devices and their resulting operation

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11863214

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 14008562

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2011863214

Country of ref document: EP