CN201156248Y - Hardware construction for computing number of multiple data ordering - Google Patents
Hardware construction for computing number of multiple data ordering Download PDFInfo
- Publication number
- CN201156248Y CN201156248Y CNU200820092094XU CN200820092094U CN201156248Y CN 201156248 Y CN201156248 Y CN 201156248Y CN U200820092094X U CNU200820092094X U CN U200820092094XU CN 200820092094 U CN200820092094 U CN 200820092094U CN 201156248 Y CN201156248 Y CN 201156248Y
- Authority
- CN
- China
- Prior art keywords
- data
- sequence number
- row
- digital comparator
- input
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The utility model discloses a computer multi-data sequencing number hardware structure. When the sequencing number of each number is calculated in a data sample of N numbers, the structure consists of N calculating units, wherein, every calculating unit calculates one data sequencing number; the calculating unit comprises N minus one numerical value comparators and one adder; the data of the sequence number to be calculated is respectively input to the same input end of N minus one numerical value comparators, the other N minus one data are respectively input to the other input end of the numerical value comparators, and a comparison result is input to the adder to be added to obtain the sequencing number of the data. The structure uses the hardware structure to realize the calculation of data sequencing number, the data to be compared is not restricted in number, the calculations of sequencing data and getting the mid value, and the like, can be realized just by adding the simple hardware structure, and the hardware structure has wide application range; in addition, the structure is convenient for the hardware to be realized and reduces the area of the hardware, the operation speed is better than software, and the speed advantage of the structure is more obvious for more data.
Description
Technical field
The utility model relates to a kind of structure of computational data ordering sequence number.
Background technology
When data image signal is handled, need after the computational data ordering sequence number and find intermediate value, medium filtering in the noise processed finds intermediate value after just needing to calculate the sequence number that sorts, medium filtering is generally opened 3 * 3 windows, or 5 * 5 windows, finds out intermediate value then from these data.
The ordering sequence number is exactly that logarithm is by the sequence number that is arranged in order from big to small, the sequence number that perhaps is arranged in order from small to large, just can obtain the ordering of data after the calculating ordering sequence number, ordering has a lot in the software algorithm the inside, as bubble sort, selection sort, insertion ranking method, quick sort, sequencing by merging method, counting sorting etc.These methods are easy to realization in software program, but are difficult to realize these algorithms and thought in hardware, and the ordering that software is realized is difficult to satisfy the requirement of real-time system to speed.
Summary of the invention
Technical problem to be solved in the utility model is to need hardware to realize the situation of calculating multidata ordering sequence number at a high speed for sound, Video processing and other, design a kind of hardware of being convenient to and realizes the sort structure of sequence number of the multidata of calculating, this structure composed as follows:
A kind of hardware configuration that calculates multidata ordering sequence number is when calculating the ordering sequence number of N number wherein in the data sample of N number; This structure is made up of N computing unit;
Wherein each described computing unit calculates a data ordering sequence number, described arbitrary computing unit comprises N-1 digital comparator and a totalizer, the data of sequence number to be calculated are imported the same input end of N-1 digital comparator respectively, another input end of other N-1 data difference input value comparer, the result after comparing inputs to totalizer and carries out the ordering sequence number that addition obtains these data.
In structure described in the utility model,, only need this comparative result by reverser, to substitute required digital comparator if two data compare in described other computing unit.
In structure described in the utility model, can replace with reverser to N (N-1)/2 a described digital comparator at most.
Digital comparator described in the utility model can be one and is used to realize that substracting unit that two input data are subtracted each other and one are used to judge whether subtract each other the result forms greater than zero judgment means; Two number input substracting units to be compared subtract each other, and the difference input judgment means that obtains is judged, result's output of this judgement.
This structure uses hardware to realize the computing of data sorting sequence number, and the data of sample are not had restriction on the number, only needs to add the ordering that simple hardware construction just can realize data, gets computings such as intermediate value, has application space widely.Realize because algorithm of the present utility model is particularly suitable for hardware, reduced the area of hardware, compared the obvious speed advantage with software, and how of the present utility model data speed advantage more be just obvious more.
Description of drawings
Fig. 1 is the synoptic diagram of digital comparator input/output port;
Fig. 2 is the structural drawing of the utility model embodiment 2;
Fig. 3 is the structural drawing of the utility model embodiment 3;
Fig. 4 is the structural drawing of the utility model embodiment 4.
Embodiment
Below in conjunction with accompanying drawing embodiment of the present utility model is elaborated.
Embodiment 1
Used digital comparator has two input ends to be respectively side a and b as shown in Figure 1 in the utility model, and two numbers to be compared are respectively from the side a and b input, and output terminal is exported the number of a 1bit.If the number of A end is more than or equal to the number of B end, then output terminal exports 1, if the number that the number of input A end is held less than input B, then output terminal C output 0.
1 data in calculating N data sample when being made as the ordering sequence number of a, need a computing unit.It comprises N-1 digital comparator and 1 totalizer.Data a imports the same input end of all digital comparators respectively, and other another input end of data input value comparer compares, and its comparative result is input to addition in the totalizer, obtains the ordering sequence number of data a.The ordering sequence number that the utility model Structure Calculation goes out is since 0.
Embodiment 2
To get N=3 is example, when 3 data ordering sequence number to be calculated is promptly arranged, needs the computing units described in 3 as the example 1 in real time.Each computing unit is arranged in one lists, be illustrated in figure 2 as 3 row, every row comprise 2 digital comparators and 1 totalizer.
Wherein, data 1 are imported the digital comparator 211 of 1 row, the A end of digital comparator 212 respectively, and other two data are imported the B end of digital comparator in 1 row respectively.
Data 2 are imported the digital comparator 221 of 2 row, the A end of digital comparator 222 respectively, and other data are imported the B end of digital comparator in 2 row respectively.
Data 3 are imported the digital comparator 231 of 3 row, the A end of digital comparator 232 respectively, and other data are imported the B end of digital comparator in 3 row respectively.
Data 1 and data 2 are the output result after relatively in digital comparator 211, and data 1 and data 3 are at the comparative result input summer 210 of digital comparator 212.Addition obtains the sequence number that data 1 sort in 3 numbers in totalizer 210.
Data 2 and data 1 are the output result after relatively in digital comparator 221, with data 2 and the data 3 comparative result input summer 220 at digital comparator 222.Addition obtains the sequence number that data 2 sort in 3 numbers in totalizer 220.
The same, data 3 and data 1 are the output result after relatively in digital comparator 231, with data 3 and the data 2 comparative result input summer 230 at digital comparator 232.Addition obtains the sequence number that data 3 sort in 3 numbers in totalizer 230.
Embodiment 3
In Fig. 2, the 2nd columns value comparer 221 data 2 and data 1 relatively compare in the 1st columns value comparer 211, and just the result is opposite.The data 3 and the data 1 that compare in the same digital comparator 231, digital comparator 232 data 3 and data 2 relatively, all row relatively have been that result relatively is opposite in front, therefore can be undertaken negate to front result relatively by reverser and obtain these comparative results, thereby remove these digital comparators to save hardware space.Can calculate simply that there being the N number relatively be divided under the situation of N row, second shows one group of data is listed as relatively mistake in front; The 3rd shows two groups of data is listed as relatively mistake in front; In the N row, there are N-1 group data to be listed as relatively mistake in front by that analogy, therefore in the N row, can save N (N-1)/2 digital comparator altogether.
Save later structure as shown in Figure 3, in data 1 and the data 2 input value comparers 311, in data 1 and the data 3 input value comparers 312, output result addition in totalizer 310 of digital comparator 311 and digital comparator 312 obtains the sequence number of data 1.
Data 2 and data 3 input comparators 322, the comparative result of digital comparator 311 through behind the reverser with comparative result addition in totalizer 320 of digital comparator 322, obtain the sequence number of data 2.
The output result of digital comparator 312 and the output result of digital comparator 322 obtain the sequence number of data 3 respectively by addition in totalizer 330 behind the reverser.
Embodiment 4
To get N=3 is example, promptly comprises 9 numerals at data sample, needs to calculate the ordering sequence number of each data as shown in Figure 4, and data 1 are imported the A end of the 1st all digital comparators of row respectively.Other 8 data are imported the B end of all digital comparators in 1 row respectively, and the output result of the 1st columns value comparer addition in totalizer 410 obtains the ordering sequence number of data 1.
Data 2 are imported the digital comparator A end of the 2nd row 2 to 8 row respectively, and other 7 data are imported the B end of the 2nd row 2 to 8 line number value comparers respectively except that data 1 and 2; The digital comparator of 1 row 1 row by reverser after and the output of 2 all digital comparators of row as a result in the input summer 420 addition obtain the ordering sequence number of data 2.
Data 3 are imported the digital comparator A end of the 3rd row 3 to 8 row respectively, and other 6 data are imported the B end of the 3rd row 3 to 8 line number value comparers respectively except that data 1,2,3; The output result of 2 row, 1 columns value comparer, 2 row, 2 columns value comparers by reverser after and the output of all digital comparators of 3 row as a result in the input summer 430 addition obtain the ordering sequence number of data 3.
Data 4 are imported the digital comparator A end of the 4th row 4 to 8 row respectively, and the B that imports the 4th row 4 to 8 line number value comparers except that data 1 other 5 data to 4 respectively holds; 3 row, 1 columns value comparer, 3 row, 2 columns value comparers, 3 row, 3 columns value comparers by reverser after and the output of all digital comparators of 4 row as a result in the input summer 440 addition obtain the ordering sequence number of data 4.
Can extrapolate the concrete structure of the 5th to 7 row according to accompanying drawing and the structure law that is listed as above, not be described in detail at this.
Eighth row 1 columns value comparer, 8 row, 2 columns value comparers, 8 row, 3 columns value comparers, 8 row, 4 columns value comparers, 8 row, 5 columns value comparers, 8 row, 6 columns value comparers, 8 row, 7 columns value comparers, 8 row, 8 columns value comparers, obtain the ordering sequence number of data 9 by addition in the input summer behind the reverser 490.
Embodiment 5
Comprise N numeral at data sample, need to calculate the ordering sequence number of each data, data 1 are imported the A end of the 1st all digital comparators of row respectively.Other 2～N data are imported the B end of corresponding row digital comparator respectively, and the result of the 1st columns value comparer output addition in the totalizer of the 1st row obtains the ordering sequence number of data 1.
Data 2 are imported the digital comparator A end of capable 2 row of the 2nd～N-1 respectively, and 3～N data are imported the B end of the capable respective value comparer of the 2nd row 2～N-1 respectively.The digital comparator of 1 row 1 row by reverser after and the output result of all digital comparators of 2 row import the ordering sequence number that the addition of secondary series totalizer obtains data 2.
Data 3 are imported the digital comparator A end of capable 3 row of the 3rd～N-1 respectively, and the 4th～N data are imported the B end of the capable digital comparator of the 3rd row 3～N-1 respectively.2 row, 1 columns value comparer and 2 row, 2 columns value comparers by reverser after and the output result of all digital comparators of 3 row import the ordering sequence number that the 3rd row totalizer addition obtains data 3.
Data 4 are imported the digital comparator A end of capable 4 row of the 4th～N-1 respectively, and the 5th～N data are imported the B end of the digital comparator of the 4th row 4 to 8 row respectively.3 row, 1 columns value comparer, 3 row, 2 columns value comparers and 3 row, 3 columns value comparers by reverser after and the output result of all digital comparators of 4 row import the ordering sequence number that the 3rd row totalizer addition obtains data 4.
Can extrapolate the 5th to the N-2 concrete structure that is listed as according to the structure law of top every row, not be described in detail at this.
Data N-1 imports the capable digital comparator A end of N-1 row N-1 respectively, and N data are imported the B end of the digital comparator of the capable N-1 row of N-1.With 1 row N-2 line number value comparer, 2 row N-2 line number value comparers and 3 row N-2 line number value comparers to the output result of the export structure of N-3 row N-2 line number value comparer and N-2 row N-2 line number value comparer by reverser after and the output result of N-1 row N-1 line number value comparer import the ordering sequence number that the addition of N-1 row totalizer obtains data N-1.
The capable 1 columns value comparer of N-1, the capable 2 columns value comparers of N-1, the capable 3 columns value comparers of N-1, the capable 4 columns value comparers of N-1, the capable 5 columns value comparers of N-1, the capable 6 columns value comparers of N-1 to the capable N-1 columns of N-1 value comparer by reverser after the addition of input N row totalizer obtain the ordering sequence number of data N.
Digital comparator is also replaceable to be one and to be used to realize the device of subtraction and to be used to judge that comparative result is whether greater than zero device.Need relatively two numbers respectively input be used to realize that result's input of the device output of subtraction is used for judged result whether greater than zero device, its judged result output.
Digital comparator also can be one and is used to realize the device of subtraction and is used to judge whether comparative result is formed greater than zero device.Need relatively two numbers respectively input be used to realize that result's input of the device output of subtraction is used for judged result whether greater than zero device, its judged result output.
This structure uses hardware to realize the calculating of data sorting sequence number, the data of sample is not had restriction on the number, and only need add the ordering that simple hardware construction just can realize data, gets computings such as intermediate value, has application space widely.Realize because algorithm of the present utility model is particularly suitable for hardware, reduced the area of hardware, compared the obvious speed advantage with software.And the speed advantage of the many more orderings of data is just obvious more.
Certainly; above-mentioned explanation is not to be to restriction of the present utility model; the utility model also be not limited in above-mentioned for example, variation, remodeling, interpolation or replacement that the those of ordinary skill of present technique is made in essential scope of the present utility model also should belong to the protection domain of utility model.
Claims (4)
1, a kind of hardware configuration that calculates multidata ordering sequence number is characterized in that, when calculating ordering sequence number of each number wherein in the data sample of N number; This structure is made up of N computing unit;
Wherein each described computing unit calculate a number according to the ordering sequence number, described arbitrary computing unit comprises N-1 digital comparator and a totalizer, the data of sequence number to be calculated are imported the same input end of N-1 digital comparator respectively, another input end of other N-1 data difference input value comparer, the result after comparing inputs to totalizer and carries out the ordering sequence number that addition obtains these data.
2, a kind of hardware configuration that calculates multidata ordering sequence number according to claim 1 is characterized in that, if two data compare in described other computing unit, only needs this comparative result by reverser, to substitute required digital comparator.
3, a kind of hardware configuration that calculates multidata ordering sequence number according to claim 2 is characterized in that, can replace with reverser to N (N-1)/2 a described digital comparator at most.
4, according to the arbitrary described a kind of hardware configuration that calculates multidata ordering sequence number of claim 1 to 3, it is characterized in that described digital comparator can be one and is used to realize that substracting unit that two input data are subtracted each other and one are used to judge whether subtract each other the result forms greater than zero judgment means; Two number input substracting units to be compared subtract each other, and the difference input judgment means that obtains is judged, result's output of this judgement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU200820092094XU CN201156248Y (en) | 2008-02-03 | 2008-02-03 | Hardware construction for computing number of multiple data ordering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU200820092094XU CN201156248Y (en) | 2008-02-03 | 2008-02-03 | Hardware construction for computing number of multiple data ordering |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201156248Y true CN201156248Y (en) | 2008-11-26 |
Family
ID=40104027
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNU200820092094XU Expired - Fee Related CN201156248Y (en) | 2008-02-03 | 2008-02-03 | Hardware construction for computing number of multiple data ordering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201156248Y (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101567781A (en) * | 2009-06-09 | 2009-10-28 | 刘海云 | Sequence number encrypting method and cipher machine generated by sequencing |
| CN102207846A (en) * | 2010-03-31 | 2011-10-05 | 国际商业机器公司 | Circuit and method for realizing data sorting |
| CN104317549A (en) * | 2014-10-15 | 2015-01-28 | 中国航天科技集团公司第九研究院第七七一研究所 | Cascade structure circuit and method for realizing data sorting |
| CN109426484A (en) * | 2017-08-28 | 2019-03-05 | 华为技术有限公司 | A kind of data sorting device, method and chip |
-
2008
- 2008-02-03 CN CNU200820092094XU patent/CN201156248Y/en not_active Expired - Fee Related
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101567781A (en) * | 2009-06-09 | 2009-10-28 | 刘海云 | Sequence number encrypting method and cipher machine generated by sequencing |
| CN102207846A (en) * | 2010-03-31 | 2011-10-05 | 国际商业机器公司 | Circuit and method for realizing data sorting |
| CN104317549A (en) * | 2014-10-15 | 2015-01-28 | 中国航天科技集团公司第九研究院第七七一研究所 | Cascade structure circuit and method for realizing data sorting |
| CN109426484A (en) * | 2017-08-28 | 2019-03-05 | 华为技术有限公司 | A kind of data sorting device, method and chip |
| CN109426484B (en) * | 2017-08-28 | 2021-08-31 | 华为技术有限公司 | Data sorting device, method and chip |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110688088B (en) | General nonlinear activation function computing device and method for neural network | |
| CN201156248Y (en) | Hardware construction for computing number of multiple data ordering | |
| CN109558400A (en) | Data processing method, device, equipment and storage medium | |
| CA2637975A1 (en) | Method and system for determining topical on-line influence of an entity | |
| CN102724219A (en) | A network data computer processing method and a system thereof | |
| CN109635084B (en) | Real-time rapid duplicate removal method and system for multi-source data document | |
| CN101477552A (en) | Website user rank division method | |
| CN102707931A (en) | Digital signal processor based on parallel data channel | |
| CN107895342B (en) | Implementation method of high-speed color screening method based on FPGA | |
| CN101425794B (en) | Digital filter with fixed coefficient | |
| CN105096101A (en) | Analyzing method for obtaining evidence from networked mails based on LDA model and centrality degree algorithm | |
| CN103902798A (en) | Data preprocessing method | |
| CN106909490B (en) | Monitoring equipment data flow evaluation and noise elimination method | |
| CN103001605B (en) | DA-RNS (distributed arithmetic-residue number system) algorithm based FIR (finite impulse response) filter realizing method | |
| CN116975770A (en) | Cast steel defect detection method and system | |
| CN117113247A (en) | Drainage system abnormality monitoring method, equipment and storage medium based on two-classification and clustering algorithm | |
| CN100394797C (en) | Method of realizing VLSI of brightness interpolator based on AVS movement compensation | |
| CN108255463A (en) | A kind of digital logical operation method, circuit and fpga chip | |
| CN113079176B (en) | High-speed network flow abnormity detection system suitable for mass data | |
| CN109634556A (en) | A kind of multiply-accumulator and cumulative output method | |
| CN112416920B (en) | MES-oriented data cleaning method and system | |
| CN109343824B (en) | CFAR reference unit summation and constant false alarm rate detection method and device based on FPGA | |
| US8990277B2 (en) | Method for searching a lookup table | |
| KR101900576B1 (en) | Triangles counting method for graph stream | |
| Roskind | A fast sort-selection filter chip with effectively linear hardware complexity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20081126 Termination date: 20140203 |