CN110489267A - Memory and the method for reinforcing data to be stored - Google Patents
Memory and the method for reinforcing data to be stored Download PDFInfo
- Publication number
- CN110489267A CN110489267A CN201910618491.9A CN201910618491A CN110489267A CN 110489267 A CN110489267 A CN 110489267A CN 201910618491 A CN201910618491 A CN 201910618491A CN 110489267 A CN110489267 A CN 110489267A
- Authority
- CN
- China
- Prior art keywords
- data
- subdata
- check code
- stored
- error correcting
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/08—Error detection or correction by redundancy in data representation, e.g. by using checking codes
- G06F11/10—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
- G06F11/1008—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices
- G06F11/1044—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices with specific ECC/EDC distribution
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Techniques For Improving Reliability Of Storages (AREA)
- Error Detection And Correction (AREA)
Abstract
A kind of method that the invention is related to memory and reinforces data to be stored, wherein the memory, comprising: data reinforce module, for reinforcing data to be stored, comprising: error correcting and detecting unit, for carrying out error correcting and detecting coding and decoding to data to be stored;Check code acquiring unit is connected to the error correcting and detecting unit, for obtaining the check code used during error correcting and detecting encodes;Check code copied cells are connected to the check code acquiring unit, for replicating the check code, make 3 parts of number of check code;Data splitting and reorganizing unit, for carrying out splitting and reorganizing to data;Data concatenation unit is connected to the check code copied cells, and is connected to the data splitting and reorganizing unit, for splicing the check code after the data and duplication after splitting and reorganizing, obtains the reinforcing data of storage.
Description
Technical field
The present invention relates to memory area, in particular to a kind of memory and the method for reinforcing data to be stored.
Background technique
With the high speed development of integrated circuit, memory occupies increasingly consequence, wherein static memory
(SRAM) because its speed it is fast, it is low in energy consumption be widely used in various high-speed memories, meanwhile, SRAM is in all kinds of spacecrafts
Also have in the circuit system of part and is more widely applied, according to statistics, in space radiation environment, single particle effect (SEE,
Single Event Effects) it is to cause memory stores data that one of the main reason for wrong occurs.SEE is mainly universe
In high energy particle such as heavy nucleus particle, α particle incidence device when, caused by the charge that deposits on track is collected by sensitive nodes
The signal of memory is overturn.Therefore, the SRAM that space flight is used will not only guarantee the characteristic of itself, should also have anti-well
The characteristic of single particle effect.
For the reinforcing of memory, in the prior art usually using the method for duplication redundancy, Hamming is carried out to data first
Code coding, uses triplication redundancy (TMR, Triple Modular Redundancy) to the new data of generation later.But in this way
It does, on the one hand former data and check code is replicated simultaneously when triplication redundancy replicates, need to occupy memory biggish
Area can only correct one mistake on the other hand due to the characteristic of Hamming code, the continuous multidigit for caused by high energy particle
The case where reversion, effect are poor.
Referring to Fig. 1, the standard D flip-flop (D-flip flop, DFF) that traditional TMR design mainly will be used to reinforce
Unit is copied into 3 parts, for convenience of description for the sake of, three d type flip flops are respectively designated as A, B, C.And then to 3 d type flip flop lists
The output of member carries out 3 and selects 2 votings, and this method, which is reduced from probability when transmitting data using standard D flip-flop, soft error occurs
Probability accidentally can select 2 voting circuits to filter a signal overturning, the circuit structure of voting circuit such as Fig. 2 institute by 3
Show.The output F value of voting circuit is equal to the majority (such as A=1, B=1, C=0, then F=1) in A, B, C, utilizes space
Redundancy improves reliability.
Although technical solution in the prior art can reach preferable performance, but need to occupy biggish memory surface
Product.
Summary of the invention
The purpose of the present invention is to provide a kind of memory and the methods for reinforcing data to be stored, can reinforce need meeting
Reduce the occupancy of memory while asking.
In order to solve the above-mentioned technical problems, the present invention provides a kind of memories, comprising: data reinforce module, for adding
Gu data to be stored, comprising: error correcting and detecting unit, for carrying out error correcting and detecting coding and decoding to data to be stored;Check code
Acquiring unit is connected to the error correcting and detecting unit, for obtaining the check code used during error correcting and detecting encodes;Verification
Code copied cells, are connected to the check code acquiring unit, for replicating the check code, make 3 parts of number of check code;Number
According to splitting and reorganizing unit, it is connected to the error correcting and detecting unit, for carrying out splitting and reorganizing to data;Data concatenation unit, even
The check code copied cells and the data splitting and reorganizing unit are connected to, for will be after the data and duplication after splitting and reorganizing
Check code splicing, obtains the reinforcing data of storage.
Optionally, the check code copied cells include: standard D flip-flop, and number is wanted with the check code copied cells
The number of duplication is consistent, and each standard D flip-flop is connected to the check code acquiring unit, the school after exporting a duplication
Test code.
Optionally, the data splitting and reorganizing unit includes: the first fractionation subelement, for from highest order to lowest order according to
It is secondary that the data to be stored is split, get 32 every part of the first subdata;Second splits subelement, is connected to institute
The first fractionation subelement is stated, and is connected to the error correcting and detecting unit, is used for from highest order to lowest order successively to described first
Subdata is split, and 8 every part of the first to fourth the second subdatas are got;Third splits subelement, is connected to described
Second fractionation subelement gets 2 every part for successively splitting to second subdata from highest order to lowest order
First to fourth third subdata.Data to be stored third subdata.
Optionally, the data splitting and reorganizing unit further include: recombination subelement is connected to the third and splits son list
Member is recombinated for first subdata, reformulation rule include: by be subordinated to same first subdata, sequence it is identical
Third subdata be integrated together, form first to fourth recombination subdata, and the first to fourth recombination subdata point
Not Dui Yingyu first subdata all first to fourth third subdatas combination;And in the recombination subdata, by institute
The ascending order for stating the serial number for the second subdata that third subdata is slave to, by the third subdata with same serial number from most
A high position is aligned to lowest order.
Optionally, the rule that the data concatenation unit is spliced is as follows: recombinating subdata described first to the 4th
Lowest order after respectively correspond addition first to fourth group of second subdata check code duplication as a result, complete splicing.
Optionally, the error correcting and detecting unit is Hamming code error correcting and detecting unit, for carrying out to second subdata
Hamming code error correcting and detecting coding and decoding, and the number of the check code of every a second subdata generation is five.
In order to solve the above-mentioned technical problems, the present invention provides a kind of methods for reinforcing data to be stored, including following step
It is rapid: error correcting and detecting coding is carried out to data to be stored;The check code used in error correcting and detecting cataloged procedure is obtained, and will be described
Check code copies as three parts;Three parts of check codes after duplication are spliced with the data to be stored, is formed and reinforces data.
Optionally, further comprising the steps of before carrying out error correcting and detecting coding to data to be stored: from highest order to minimum
Position successively splits the data to be stored, gets 32 every part of the first subdata;From highest order to lowest order according to
It is secondary that first subdata is split, get 8 every part of the first to fourth the second subdatas.
Optionally, when carrying out error correcting and detecting coding and decoding to data to be stored, comprising the following steps: to second son
Data carry out Hamming code error correcting and detecting coding, and every one 8 the second subdatas generate five bit check codes.
Optionally, before three parts of check codes after duplication being spliced with the data to be stored, comprising the following steps: from most
A high position successively splits second subdata to lowest order, gets 2 every part of first to fourth third subdata;
By be subordinated to same first subdata, the identical third subdata of sequence be integrated together, form first to fourth recon
Data, the first to fourth recombination subdata correspond respectively to all first to fourth third subdatas of first subdata
Combination;The ascending order of the serial number for the second subdata being slave to by the third subdata, by described with same serial number
Three subdatas are aligned to lowest order from highest order, form the recombination subdata;In the first to fourth recombination subdata
The duplication plus the check code of first to fourth the second subdata is respectively corresponded after lowest order as a result, completing splicing.
Memory of the invention and the method for reinforcing data to be stored first reinforce data when writing data, right first
Data to be stored carry out Hamming code coding generate check code, later only duplication redundancy is carried out to check code, radiation resistance with
Equilibrium is achieved between memory area shared by coded data.Check code and data to be stored after reusing redundancy later are spelled
It connects, in this way when high energy particle causes multidigit to invert, still can restore original data to a certain extent.Further, In
When reading the reinforcing data stored in memory, three parts of check codes can be obtained into a check code by voting circuit first,
Original data are restored according to the rule of splicing later, obtained a check code is stitched together with pending data then and is entangled
False retrieval false retrieval is tested, in case of the mistake for being less than or equal to one, the available correct data after error correcting and detecting, if hair
The raw mistake for being greater than one then returns to an error flag position, at least occurs two in the data for notifying user currently stored
Mistake.
Detailed description of the invention
Fig. 1 is duplication redundancy structure schematic diagram in the prior art.
Fig. 2 is voting circuit structural schematic diagram in the prior art.
Fig. 3 is the flow diagram in the prior art for reinforcing data to be stored.
Fig. 4 is the structural schematic diagram of memory in a kind of specific embodiment of the invention.
Fig. 5 is the step schematic diagram that the method for data to be stored is reinforced in a kind of specific embodiment of the invention.
Specific embodiment
To a kind of memory proposed by the present invention and data to be stored is reinforced below in conjunction with the drawings and specific embodiments
Method is described in further detail.
Referring to Fig. 4, for the structural schematic diagram of memory in a kind of specific embodiment of the invention.
In this specific embodiment, a kind of memory is provided, comprising: data reinforce module, for reinforcing wait store
Data, comprising: error correcting and detecting unit 401, for carrying out error correcting and detecting coding and decoding to data to be stored;Check code obtains single
Member 402, is connected to the error correcting and detecting unit 401, for obtaining the check code used during error correcting and detecting encodes;Verification
Code copied cells 403, are connected to the check code acquiring unit 402, for replicating the check code, make the number of check code
3 parts;Data splitting and reorganizing unit 404 is connected to the error correcting and detecting unit 401, for carrying out splitting and reorganizing to data;Data
Concatenation unit 405 is connected to the check code copied cells 403 and the data splitting and reorganizing unit 404, for that will split weight
Check code splicing after data and duplication after group, obtains the reinforcing data of storage.
In a specific embodiment, the check code copied cells 403 include: standard D flip-flop, number with it is described
The number to be replicated of check code copied cells 403 is consistent, and each standard D flip-flop is connected to the check code acquiring unit
402, the check code after exporting a duplication.
In a specific embodiment, the data splitting and reorganizing unit 404 includes: the first fractionation subelement, is used for
Successively the data to be stored is split from highest order to lowest order, gets 32 every part of the first subdata;Second
Subelement is split, described first is connected to and splits subelement, and be connected to the error correcting and detecting unit, for from highest order to most
Low level successively splits first subdata, gets 8 every part of the first to fourth the second subdatas;Third is split
Subelement is connected to described second and splits subelement, for successively carrying out to second subdata from highest order to lowest order
It splits, gets 2 every part of first to fourth third subdata.Data to be stored third subdata.
The data splitting and reorganizing unit further include: recombination subelement is connected to the third and splits subelement, is used for institute
It states the first subdata to be recombinated, reformulation rule includes: that will be subordinated to same first subdata, the identical third subnumber of sequence
According to being integrated together, first to fourth recombination subdata is formed, and the first to fourth recombination subdata corresponds respectively to this
The combination of all first to fourth third subdatas of first subdata;And in the recombination subdata, by the third subnumber
According to the ascending order of the serial number for the second subdata being slave to, the third subdata with same serial number is aligned to from highest order
Lowest order.
In a specific embodiment, the rule that the data concatenation unit 405 is spliced is as follows: described first
The duplication of the check code of first to fourth group of second subdata of addition is respectively corresponded after to the lowest order of the 4th recombination subdata
As a result, completing splicing.
In a specific embodiment, the error correcting and detecting unit 401 is Hamming code error correcting and detecting unit, for institute
It states the second subdata and carries out Hamming code coding and decoding, while the number for the check code that every a second subdata generates is five
Position in this way, can correct a bit-errors present in read data in decoding stage, and detects two existing
The situation of mistake.
In a specific embodiment, the verification code bit number that the Hamming code error correcting and detecting unit 401 uses is 5,
After five bit check codes are added to second subdata, 3 power of 20 power to 2 of the data after occupying addition check code
The highest order of data behind position and the addition check code.
In this way, the digit of the reinforcing data of formation is 92, and digit is less, and due to making to one group of first subdata
It is encoded with Hamming code error correcting and detecting, every 8 five bit check codes of corresponding generation, therefore in decoding stage, can correct and be read
Data present in a bit-errors, it is simple and convenient and the case where detect existing two bit-errors.
In a specific embodiment, the memory further includes voting circuit, and it is single to be connected to the check code duplication
Member 403 is put to the vote carry for the check code after replicate, and the check code of output is the majority of the check code after replicating, in this way,
Make user when being decoded using check code to reinforcing data, the check code after using voting, the probability of mistake is smaller,
Be conducive to reinforce the smooth decoding of data.
In a specific embodiment, the voting circuit selects 2 voting circuits for 3, straight with the number of copies of check code
Connect correlation.
Referring to Fig. 5, illustrating the step of the method for data to be stored to be reinforced in a kind of specific embodiment of the invention
Figure.
In this specific embodiment, a kind of method for reinforcing data to be stored is additionally provided, comprising the following steps: S51
Error correcting and detecting coding is carried out to data to be stored;S52 obtains the check code used in error correcting and detecting cataloged procedure, and will be described
Check code copies as three parts;S53 splices three parts of check codes after duplication with data, is formed and reinforces data.
In a specific embodiment, further comprising the steps of before carrying out error correcting and detecting coding to data to be stored:
Successively the data to be stored is split from highest order to lowest order, gets 32 every part of the first subdata;From most
A high position successively splits first subdata to lowest order, gets 8 every part of the first to fourth the second subdatas.
In a specific embodiment, when carrying out error correcting and detecting coding and decoding to data to be stored, including following step
It is rapid: Hamming code error correcting and detecting coding to be carried out to second subdata, and every one 8 the second subdatas generate five bit check codes.
In a specific embodiment, three parts of check codes after duplication are wrapped with before data to be stored splicing
Include following steps: successively second subdata being split from highest order to lowest order, get the first of 2 every part to
4th third subdata;By be subordinated to same first subdata, the identical third subdata of sequence be integrated together, form the
One to the 4th recombination subdata, the first to fourth recombination subdata correspond respectively to all the first of first subdata to
The combination of 4th third subdata;The ascending order of the serial number for the second subdata being slave to by the third subdata will have same
The third subdata of one serial number is aligned to lowest order from highest order, forms the recombination subdata.Pass through the fractionation of data
Recombination, forms third subdata to be stored, and causes to reinforce the reversion of data multidigit (MBU, Multiple when high energy particle in this way
Bit Upset) when, it still can restore original data to a certain extent.
;It is respectively corresponded after the lowest order of the first to fourth recombination subdata plus first to fourth the second son
The duplication of the check code of data is as a result, complete splicing.
The method of the reinforcing data to be stored of the specific embodiment allows data to be compiled by Hamming code first when writing data
Code generates check code, only carries out triplication redundancy duplication to check code later.In this way, shared by radiation resistance and the coded data
Equilibrium is achieved between memory area.
In a specific embodiment, when reading reinforcing data, the check code after three parts of redundancies are replicated first leads to
It crosses three and selects two voting circuits, the majority of the check code after output duplication, later according to the splitting and reorganizing rule for reinforcing data
Restore original data, to relevant splitting and reorganizing the step of carries out inverse operation, then obtained a check code with it is original
Data, which are stitched together, carries out error correcting and detecting inspection, in case of the mistake for being less than or equal to one, can obtain after error correcting and detecting
To correct data, in case of the mistake for being greater than one, then an error flag position is returned to, notifier processes device data have occurred
Mistake.
Embodiment one:
Since in computer, the data length of processing is generally 32 or 32 integral multiples, when data length is not
At 32, initial data can be split as 32 processing by remaining, therefore here to iting is assumed that 32 initial data make
Following processing:
(1) initial data is split into four parts, every part is 8, remembers that this four parts of data are respectively a [7:0], b [7:0], c
[7:0], d [7:0];
(2) check code is generated to 8 data after fractionation, by the Hamming code coding rule of 13_8, every 8 generate 5
Check code, 5 bit check codes of every 8 generations can correct number so that after encoding using Hamming code to initial data
A bit-errors present in detect two bit-errors present in data.For 5 bit check codes of generation, we remember respectively
For parity_a [4:0], parity_b [4:0], parity_c [4:0], parity_d [4:0];
(3) every part of check code is replicated three parts, and four parts 8 data is recombinated, it is assumed that the data after recombination are denoted as A
[7:0], B [7:0], C [7:0], D [7:0], the rule of recombination include:
A [7:0]={ a [7:6], b [7:6], c [7:6], d [7:6] };
B [7:0]={ a [5:4], b [5:4], c [5:4], d [5:4] };
C [7:0]={ a [3:2], b [3:2], c [3:2], d [3:2] };
D [7:0]={ a [1:0], b [1:0], c [1:0], d [1:0] };
(4) check code of data and redundancy after recombination is stitched together, remember spliced data be DATA_A,
Memory is written in DATA_B, DATA_C, DATA_D.The rule of splicing includes:
DATA_A={ A [7:0], parity_a [4:0], parity_a [4:0], parity_a [4:0] };
DATA_B={ B [7:0], parity_b [4:0], parity_b [4:0], parity_b [4:0] };
DATA_C={ C [7:0], parity_c [4:0], parity_c [4:0], parity_c [4:0] };
DATA_D={ D [7:0], parity_d [4:0], parity_d [4:0], parity_d [4:0] };
In this embodiment, 32 original data are after coding, it is only necessary to 92.
It is if being encoded using method in the prior art, and to data by the Hamming code coding rule of 13-8, then right
Original 32 data processings, after being formed and reinforcing data, the final data that memory module is written are as follows:
DATA_A=a [7:0], parity_a [4:0], a [7:0], parity_a [4:0], a [7:0], parity_a [4:
0]};
DATA_B=b [7:0], parity_b [4:0], b [7:0], parity_b [4:0], b [7:0], parity_b [4:
0]};
DATA_C=c [7:0], parity_c [4:0], c [7:0], parity_c [4:0], c [7:0], parity_c [4:
0]};
DATA_D=d [7:0], parity_d [4:0], d [7:0], parity_d [4:0], d [7:0], parity_d [4:
0]};
At this point, 32 data are becoming 156 after reinforcing coding, the memory of occupancy is larger.
Therefore, using the method in the embodiment, the Hamming code coding rule of traditional method combination 13-8 than using, section
41% digit has been saved, the area of memory occupied in reinforcing process is greatly saved.
Assuming that memory, under the influence of high energy particle, continuous two mistakes have occurred in A [6:5], if do not treated
Storing data carries out splitting and reorganizing and stores at this time then since Hamming code decoder module cannot correct continuous two mistakes
Data malfunctioned, at this time can only notifier processes device re-write correct data.But if use new fractionation weight
The method of group carries out splitting and reorganizing to data to be stored, and A [6] and A [5] are respectively a [6] and b [7], correspond to original data block a
And b, therefore for original data block, one error in data only has occurred respectively, Hamming code solution can be passed through at this time
Code module restores correct data, therefore, by the way that data are carried out with the processing of splitting and reorganizing, in the premise for reducing memory area
Under, improve the performance of the anti-Multiple-bit upsets of memory module.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
Member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications also should be regarded as
Protection scope of the present invention.
Claims (10)
1. a kind of memory characterized by comprising
Data reinforce module, for reinforcing data to be stored, comprising:
Error correcting and detecting unit, for carrying out error correcting and detecting coding and decoding to data to be stored;
Check code acquiring unit is connected to the error correcting and detecting unit, for obtaining error correcting and detecting coding during use
Check code;
Check code copied cells are connected to the check code acquiring unit, for replicating the check code, make the number of check code
It is 3 parts;
Data splitting and reorganizing unit is connected to the error correcting and detecting unit, for carrying out splitting and reorganizing to data;
Data concatenation unit is connected to the check code copied cells and the data splitting and reorganizing unit, for that will split weight
Check code splicing after data and duplication after group, obtains the reinforcing data of storage.
2. memory according to claim 1, which is characterized in that the check code copied cells include:
Standard D flip-flop, number is consistent with the check code copied cells number to be replicated, and each standard D flip-flop is equal
It is connected to the check code acquiring unit, the check code after exporting a duplication.
3. memory according to claim 1, which is characterized in that the data splitting and reorganizing unit includes:
First fractionation subelement gets 32 for successively splitting to the data to be stored from highest order to lowest order
First subdata of every part of position;
Second splits subelement, is connected to described first and splits subelement, and be connected to the error correcting and detecting unit, for from most
A high position successively splits first subdata to lowest order, gets 8 every part of the first to fourth the second subdatas;
Third splits subelement, is connected to described second and splits subelement, for from highest order to lowest order successively to described the
Two subdatas are split, and 2 every part of first to fourth third subdata is got.Data to be stored third subdata.
4. memory according to claim 3, which is characterized in that the data splitting and reorganizing unit further include:
Subelement is recombinated, the third is connected to and splits subelement, recombinated for first subdata, reformulation rule packet
It includes:
By be subordinated to same first subdata, the identical third subdata of sequence be integrated together, form first to fourth weight
Group subdata, and the first to fourth recombination subdata corresponds respectively to all first to fourth thirds of first subdata
The combination of subdata;
And in the recombination subdata, the ascending order of the serial number for the second subdata being slave to by the third subdata will have
The third subdata of same serial number is aligned to lowest order from highest order.
5. memory according to claim 4, which is characterized in that the rule that the data concatenation unit is spliced is such as
Under:
First to fourth group of second subdata of addition is respectively corresponded after the lowest order of the described first to the 4th recombination subdata
Check code duplication as a result, complete splicing.
6. memory according to claim 3, which is characterized in that the error correcting and detecting unit is Hamming code error correcting and detecting list
Member, for carrying out Hamming code error correcting and detecting coding and decoding to second subdata, and every a second subdata generates
The number of check code is five.
7. a kind of method for reinforcing data to be stored, which comprises the following steps:
Error correcting and detecting coding is carried out to data to be stored;
The check code used in error correcting and detecting cataloged procedure is obtained, and the check code is copied as three parts;
Three parts of check codes after duplication are spliced with the data to be stored, is formed and reinforces data.
8. the method according to claim 7 for reinforcing data to be stored, which is characterized in that carry out error correction to data to be stored
It is further comprising the steps of before error detection coding:
Successively the data to be stored is split from highest order to lowest order, gets 32 every part of the first subdata;
Successively first subdata is split from highest order to lowest order, gets first to fourth the of 8 every part
Two subdatas.
9. the method according to claim 8 for reinforcing data to be stored, which is characterized in that carry out error correction to data to be stored
When error detection coding and decoding, comprising the following steps:
Hamming code error correcting and detecting coding is carried out to second subdata, and every one 8 the second subdatas generate five bit check codes.
10. the method according to claim 9 for reinforcing data to be stored, which is characterized in that by three parts of verifications after duplication
When code splices with the data to be stored, comprising the following steps:
Successively second subdata is split from highest order to lowest order, gets first to fourth the of 2 every part
Three subdatas;
By be subordinated to same first subdata, the identical third subdata of sequence be integrated together, form first to fourth weight
Group subdata, the first to fourth recombination subdata correspond respectively to all first to fourth thirds of first subdata
The combination of data;
The ascending order of the serial number for the second subdata being slave to by the third subdata, by the third with same serial number
Data are aligned to lowest order from highest order, form the recombination subdata;
It respectively corresponds after the lowest order of the first to fourth recombination subdata plus first to fourth the second subdata
Check code duplication as a result, complete splicing.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910618491.9A CN110489267B (en) | 2019-07-10 | 2019-07-10 | Memory and method for reinforcing data to be stored |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910618491.9A CN110489267B (en) | 2019-07-10 | 2019-07-10 | Memory and method for reinforcing data to be stored |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110489267A true CN110489267A (en) | 2019-11-22 |
CN110489267B CN110489267B (en) | 2021-10-29 |
Family
ID=68546976
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910618491.9A Active CN110489267B (en) | 2019-07-10 | 2019-07-10 | Memory and method for reinforcing data to be stored |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110489267B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1858713A (en) * | 2005-04-30 | 2006-11-08 | 广东科龙电器股份有限公司 | Method for improving E/EPROM storage storaged dada reliability |
US20180143877A1 (en) * | 2016-04-27 | 2018-05-24 | Silicon Motion Inc. | Flash memory apparatus and storage management method for flash memory |
CN108132854A (en) * | 2018-01-15 | 2018-06-08 | 成都信息工程大学 | A kind of correcting and eleting codes coding/decoding method that can restore data element and redundant elements simultaneously |
CN109802760A (en) * | 2018-12-29 | 2019-05-24 | 北京行易道科技有限公司 | The storage method and device of data, the coding/decoding method of data |
-
2019
- 2019-07-10 CN CN201910618491.9A patent/CN110489267B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1858713A (en) * | 2005-04-30 | 2006-11-08 | 广东科龙电器股份有限公司 | Method for improving E/EPROM storage storaged dada reliability |
US20180143877A1 (en) * | 2016-04-27 | 2018-05-24 | Silicon Motion Inc. | Flash memory apparatus and storage management method for flash memory |
CN108132854A (en) * | 2018-01-15 | 2018-06-08 | 成都信息工程大学 | A kind of correcting and eleting codes coding/decoding method that can restore data element and redundant elements simultaneously |
CN109802760A (en) * | 2018-12-29 | 2019-05-24 | 北京行易道科技有限公司 | The storage method and device of data, the coding/decoding method of data |
Non-Patent Citations (1)
Title |
---|
杨卫平: "CRC 计算实现方法", 《数据库技术》 * |
Also Published As
Publication number | Publication date |
---|---|
CN110489267B (en) | 2021-10-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107403646B (en) | Flash memory device and flash memory management method | |
US7644348B2 (en) | Method and apparatus for error detection and correction | |
US6604222B1 (en) | Block code to efficiently correct adjacent data and/or check bit errors | |
CN111338840B (en) | Space data protection method, storage medium, computer program, system and terminal | |
Sanchez-Macian et al. | Enhanced detection of double and triple adjacent errors in hamming codes through selective bit placement | |
CN111459712B (en) | SRAM type FPGA single event upset error correction method and single event upset error correction circuit | |
US20190325962A1 (en) | Error Detection and Correction Circuitry | |
CN103413571B (en) | Storer and utilize this storer to realize the method for error-detection error-correction | |
CA1203027A (en) | Computer memory system which detects errors, address errors and operation errors | |
Silva et al. | CLC-A: An adaptive implementation of the Column Line Code (CLC) ECC | |
CN106531224A (en) | SRAM type FPGA-based RAM data reliability reinforcement apparatus and method | |
CN105027084B (en) | The apparatus and method of control memory in mobile communication system | |
CN110489267A (en) | Memory and the method for reinforcing data to be stored | |
CN107301881A (en) | It is a kind of based on 4 adjacent and 3 burst error correction codes SRAM memory radiation hardening methods and accumulator system | |
Satyanarayana et al. | Design and implementation of error detection and correction system for semiconductor memory applications | |
US7904786B2 (en) | Assisted memory system | |
Ghosh et al. | Selecting error correcting codes to minimize power in memory checker circuits | |
Fuja et al. | Linear sum codes for random access memories | |
CN114220474A (en) | Data processing method, device and storage medium | |
CN104378120A (en) | Hsiao coding check matrix generation method for continuous MBU detection | |
Gupta et al. | Utilization of on-line (concurrent) checkers during built-in self-test and vice versa | |
CN202838976U (en) | EDAC (Error Detection And Correction)-module-based fault-tolerant device of triple modular redundancy to storage | |
Lo | Fault-tolerant content addressable memory | |
CA1184308A (en) | True single error correction system | |
Nicolaidis et al. | Transparent BIST for ECC-based memory repair |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |