CN108563533A - A method of it obtaining ICI impact factors and promotes error correcting capability - Google Patents

A method of it obtaining ICI impact factors and promotes error correcting capability Download PDF

Info

Publication number
CN108563533A
CN108563533A CN201810258934.3A CN201810258934A CN108563533A CN 108563533 A CN108563533 A CN 108563533A CN 201810258934 A CN201810258934 A CN 201810258934A CN 108563533 A CN108563533 A CN 108563533A
Authority
CN
China
Prior art keywords
storage unit
cell
ici
impact factors
interference
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.)
Pending
Application number
CN201810258934.3A
Other languages
Chinese (zh)
Inventor
胡颖颖
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.)
Shenzhen Union Memory Information System Co Ltd
Original Assignee
Shenzhen Union Memory Information System Co Ltd
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 Shenzhen Union Memory Information System Co Ltd filed Critical Shenzhen Union Memory Information System Co Ltd
Priority to CN201810258934.3A priority Critical patent/CN108563533A/en
Publication of CN108563533A publication Critical patent/CN108563533A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/08Error detection or correction by redundancy in data representation, e.g. by using checking codes
    • G06F11/10Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
    • G06F11/1008Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices
    • G06F11/1048Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices using arrangements adapted for a specific error detection or correction feature
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/06Auxiliary circuits, e.g. for writing into memory
    • G11C16/34Determination of programming status, e.g. threshold voltage, overprogramming or underprogramming, retention
    • G11C16/3418Disturbance prevention or evaluation; Refreshing of disturbed memory data
    • G11C16/3427Circuits or methods to prevent or reduce disturbance of the state of a memory cell when neighbouring cells are read or written
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C29/00Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/38Response verification devices
    • G11C29/42Response verification devices using error correcting codes [ECC] or parity check

Abstract

The invention discloses a kind of acquisition ICI impact factors and the methods for promoting error correcting capability, it is characterized in that adjacent interference storage unit, which is obtained ahead of time, according to experiment stores different data to storage unit Cell influence degrees, the data that interference storage unit by reading storage unit Cell is actually written into, the sub- impact factor for obtaining each interference storage unit respectively, is finally overlapped and obtains final ICI impact factors.The present invention is directed to the influence of the crosstalk ICI between consecutive storage unit in NAND Flash, obtain ICI impact factors, and it applies it in the LLR of Cell to generate the LUT tables more fully and refined, to promote LDPC error correcting capabilities, the method to promote NAND Flash service lifes.This method can solve the problems, such as that modification threshold voltage can not judge that overlapping part is distributed, and ICI influences are applied, and promote error correcting capability.

Description

A method of it obtaining ICI impact factors and promotes error correcting capability
Technical field
The present invention relates to technical field of memory more particularly to a kind of sides for obtaining ICI impact factors and promoting error correcting capability Method.
Background technology
Nand Flash are a kind of non-volatile memories semiconductors, by injecting electronics to storage unit Cell floating gate layers Mode store data.As electron number increases in storage unit floating gate layer, corresponding voltage also can gradually increase.Flash's Erasing can damage the oxidation raceway groove of floating transistor, and the fluctuation and offset of voltage value, interfacial state trap is caused to restore and electronics Escape will also result in the reduction of voltage, and usual Cell voltages can be approximately Gaussian Profile.By Cell voltages and judge voltage Vth (also referred to as threshold voltage) comparison, may thereby determine that the data Level of its storage.
Fig. 1 is two kinds of contiguous memory locations voltage's distribiuting schematic diagrames, and wherein A is point of ideal two neighboring storage unit Overlapping cases are not present in the voltage of cloth, two storage units, therefore default judgement voltage Default Vref is taken to can be realized The data of two storage units judge that there is no influence each other.But the data of Nand Flash storages can be by adjacent The interference of Cell is detained influence and the random noise disturbance of Retention characteristics, causes the offset and broadening of voltage's distribiuting. Judge that voltage Vth range will result in erroneous judgement when part Cell voltage's distribiutings exceed, mistake Bit numbers is caused to increase.B is adjacent two There is the distribution for overlapping situation in a storage unit;But when adjacent Level distributions overlap, simple adjustment judges electricity Pressure not can determine that the voltage's distribiuting at overlapping place.Targetedly offset judges that voltage can reduce erroneous judgement, this is current universal Error correction method.
LDPC (Low-density Parity-check, low-density checksum) algorithm is a kind of loop iteration algorithm, It is mainly judged by carrying out reliability (Log Likelihood Ratio) to each bit, ultimately generates LUT tables (Look Up Table. look-up table, exactly each bit can have the reliability metric form of oneself in a lookup table, for characterizing corresponding bit Whether the reliable and degree of reliability), overturn the mode of iteration error correction to carrying out part bit to the data of read error and carry out.Its During generating LUT tables, required average information is known as Soft Inform ation.The acquisition of Soft Inform ation, it is contemplated that factor it is more, Information generation more refines, and LUT tables are more accurate, to which LDPC error correcting capabilities are promoted.
Invention content
LUT tables are refined last real for disadvantages described above present invention aims at how to obtain more detailed Soft Inform ation Now promote the purpose of LDPC error correcting capabilities.
To achieve the goals above, the present invention provides a kind of methods obtaining ICI impact factors, it is characterised in that according to Adjacent interference storage unit storage different data is obtained ahead of time to storage unit Cell influence degrees in experiment, is stored by reading The data that the interference storage unit of unit Cell is actually written into obtain the sub- impact factor of each interference storage unit, most respectively It is overlapped afterwards and obtains final ICI impact factors.
The method of the acquisition ICI impact factors, it is characterised in that only choose maximum on storage unit Cell influences Two adjacent storage units only read the number that the two interference storage units selected are actually written into as interference storage unit According to, and two sub- impact factors are obtained, final ICI impact factors are obtained after the two sub- impact factors are overlapped.
The method of the acquisition ICI impact factors, it is characterised in that Nand Flash are MLC types, and interference storage is single " 10 " are pressed in influence of the data of member storage to storage unit Cell>“00”>“01”>" 11 " are distributed from big to small;Storage is single The ICI impact factors selection Cell [WLi-1, BLj] and two storage units of Cell [WLi+1, BLj] of first Cell [WLi, BLj] As interference storage unit.
A method of promoting error correcting capability, reliability of the feature in calculating LDPC algorithms according to ICI impact factors LLR (Log Likelihood Ratio) and LUT tables, error correction is iterated using LLR and LUT tables;The acquisition ICI influence because Son is obtained according to following method, is that adjacent interference storage unit storage different data is obtained ahead of time to storage list according to experiment First Cell influence degrees, the data that the interference storage unit by reading storage unit Cell is actually written into, obtain each respectively The sub- impact factor for interfering storage unit is finally overlapped and obtains final ICI impact factors.
The method of the promotion error correcting capability, it is characterised in that only choose influences maximum two to storage unit Cell Adjacent storage unit only reads the data that the two interference storage units selected are actually written into as interference storage unit, and Two sub- impact factors are obtained, final ICI impact factors are obtained after the two sub- impact factors are overlapped.
The method of the promotion error correcting capability, it is characterised in that Nand Flash are MLC types, and interference storage unit is deposited " 10 " are pressed in influence of the data of storage to storage unit Cell>“00”>“01”>" 11 " are distributed from big to small;Storage unit Two storage units of ICI impact factors selection Cell [WLi-1, BLj] and Cell [WLi+1, BLj] of Cell [WLi, BLj] are made To interfere storage unit.
The present invention is directed to the influence of the crosstalk ICI between consecutive storage unit in NAND Flash, obtains ICI impact factors, And apply it to generate the LUT tables more fully and refined in the LLR of Cell, to promote LDPC error correcting capabilities, to The method for promoting NAND Flash service lifes.This method, which can solve modification threshold voltage, can not judge overlapping part distribution Problem, and ICI influences are applied, promote error correcting capability.
Description of the drawings
Fig. 1 is two kinds of contiguous memory locations voltage's distribiuting schematic diagrames;
Fig. 2 is the structural schematic diagram of NAND Flash storage units;
Fig. 3 is the voltage's distribiuting schematic diagram for MLC Nand;
Fig. 4 is Flash consecutive storage unit schematic diagrames;
Fig. 5 is that data distribution influence schematic diagram is written with it in the impacted degree of WLi;
Fig. 6 is the influence schematic diagram that the different Cell states of WLi are subject in different adjacent WL under Cell states respectively.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, those of ordinary skill in the art are obtained every other without creative efforts Embodiment shall fall within the protection scope of the present invention.
Fig. 2 is the structural schematic diagram of NAND Flash storage units, mainly include source electrode (Source), drain electrode (Drain), Grid (Control Gate) and floating gate (Floating Gate) are controlled, stores the size of data by injecting electron institute The voltage swing of generation indicates.To must first be wiped it before the write-in of NAND Flash, nothing in floating gate after erasing Electronics is expressed as " 1 ", it is written, then is that electronics, different electron amounts difference are injected to floating gate layer according to the data of write-in Indicate different data informations.
Nand Flash interfacial state traps restore and the obedience Poisson distribution of the detrapping process approximation of electronics, therefore its voltage Distribution can be approximated to be Gaussian Profile.Fig. 3 is the voltage's distribiuting schematic diagram for MLC Nand, is that Low Page sentence with wherein VL Power-off pressure, VH1 and VH2 are that High Page judge voltage.
Fig. 4 is Flash consecutive storage unit schematic diagrames, in Nand Flash, there are a kind of adjacent interval interference phenomenon, I.e. arbitrary storage unit can be caused storage charge number to change by the crosstalk of consecutive storage unit.Intermediate storage unit is known as Disturbed storage unit VictimCell, the storage unit for generating interference around remaining to Victim Cell are known as interfering storage Unit Aggressor Cell.Any one storage unit is both that Victim Cell are interfered by surrounding storage unit, and Aggressor Cell disturb the storage unit near it.
Interference of the Aggressor Cell to Victim Cell, interfere between mainly adjacent Cell it is larger, in Fig. 4 For, it is Cell [WLi-1, BLj] respectively to interfere larger Cell [WLi, BLj], Cell [WLi+1, BLj], Cell [WLi, BLj-1] and Cell [WLi, BLj+1], wherein adjacent word line WL Cell [WLi-1, BLj], Cell [WLi+1, BLj] influence compared with Greatly, the influence of the two is mainly considered herein.
Influences of the Aggressor Cell of adjacent WL to Victim Cell, essentially consists in the ablation process to WLi+1 In, it needs to apply WLi certain voltage, this voltage causes in some electron injections to the floating gate layer of WLi, this effect is referred to as soft It programs (Soft Program), under the action of soft programming, the voltage's distribiuting of Cell generates variation.
The size that WLi+1 acts on the soft programming of WLi, it is related with the write-in data of WLi+1, from the acquiescence after erasing " The span difference of 11 " states to end-state is bigger, bigger to the soft programming effect of WLi, that is to say, that by taking MLC as an example, in Cell In the case that [WLi, BLj] data with existing is written and state distribution has been determined, when data are written in Cell [WLi+1, BLj], write-in Different data be " 10 " to the influence degree situations of data in Cell [WLi, BLj]>“00”>“01”>“11”.
Degree impacted WLi is written data distribution with it and further relates to, the high state of spread voltage about susceptible, That is, by taking MLC as an example, if Fig. 5 is that the impacted degree of WLi is written with it shown in data distribution influence schematic diagram, F1 is Arrow initial position is distributed:Electronic voltage distribution situation after program in Cell;F2 is distributed for arrow final position: Electronic voltage distribution situation after influences of the Aggressor Cell to different Victim Cell electronic voltages.Cell [WLi+1, BLj] write-in data it is the same in the case of, the impacted degree of different distributions is in Cell [WLi, BLj]:“10”>“00”>“01” >“11”。
Shown in Fig. 6, characterization is that the different Cell states of WLi are subject in different adjacent WL under Cell states respectively It influences.Solid line:Electronic voltage distribution after data program;Dotted line:Victim when corresponding states being written in adjacent WL Electronic voltage distribution after cell is impacted.
In NAND Flash in error correction procedure, by taking High Page as an example, the impact factor of Aggressor Cell obtain and Applying step is as follows:
Step 1:When being read out the mistake for occurring can not correcting to WLi data, the Low of adjacent WLi-1 is read Page data obtain data set { D10, D11, D12, D13 ... }, and wherein D10 is first Bit data, and D11 is second Bit Data, and so on;And the High Page data of adjacent WLi-1 are read, data set { D20, D21, D22, D23 ... } is obtained, Middle D20 is first Bit data, and D21 is second Bit data, and so on.Thus, it is possible to obtain the 2bit numbers in Cell According to collection, i.e., { (D10D20), (D11D21), (D12D22), (D13D23) ... }, wherein (D10D20) is in first Cell 2bit data, (D11D21) they are the 2bit data in second Cell, and so on.
Step 2:If there are valid data in WLi+1, obtained in WLi+1 in each Cell according to method in step 1 2bit data such as obtain data set { (D30D40), (D31D41), (D32D42), (D33D43) ... }, wherein (D30D40) is the 2bit data in one Cell, (D31D41) they are the 2bit data in second Cell, and so on.
Step 3:The data of Cell in WLi-1 are indicated with impact factor herein according to the interval of itself and " 11 " distribution, are divided It is 4 grades, is indicated respectively with 1,2,3,4, numerical value is bigger indicates that its distance " 11 " state is remoter.When there is valid data in WLi+1, Its impact factor is superimposed with the impact factor of WLi-1, obtain all Cell in WL impact factor collection L0, L1, L2, L3 ... }, wherein L0 is the impact factor of first Cell, and L1 is the impact factor of second Cell, and so on.
Step 4:It can be with influence factor value is bigger, indicates that its influence to WLi is bigger from previous step and analysis.
Step 5:If the reading of High Page in the WLi of error reads the Low Page in WLi, to obtain at this time The 2bit data sets in each Cell of current WLi are got, according to obtained Cell distribution situations, described in step 3 Degree of susceptibility, be added to impacted degree as impact factor addition in acquired impact factor.
Step 6:After the impact factor for obtaining Cell, the data of the position of the bigger Cell of impact factor more can not It leans on, that is, the probability that malfunctions is higher.In LDPC algorithms, LLR is smaller, i.e., when corrupt data, positive exact figures are obtained after being overturn According to probability it is higher.
Step 7:According to the correspondence of the LLR and LUT tables of LDPC algorithms in used controller.Applying to LDPC When, according to the difference of each controller realization method, this likelihood ratio can be mapped, being mapped to oneself controller can Identify the value used.
When final error correction in the controller, LUT tables, that is, the value after mapping are used.Error correction principles are:When going out When existing UECC mistakes, check that the reliability of each bit, the low bit of reliability are the portion that bit overturnings have occurred in maximum possible Point, which is overturn, i.e., 1 overturning is that 0,0 overturning is 1, then goes to judge current data correctness again, if still not just Really, then the minimum bit of reliability is overturn, is iterated by the reliability for calculating each bit again, until data are whole just Iteration maximum times that are true or reaching setting.
Rational LLR and LUT tables can be calculated according to this impact factor for iteration error correction, are improved error correcting capability, are obtained To correct data.
Above disclosed is only an embodiment of the present invention, cannot limit the right model of the present invention with this certainly It encloses, those skilled in the art can understand all or part of the processes for realizing the above embodiment, and is wanted according to right of the present invention Equivalent variations made by asking still fall within the range that the present invention is covered.

Claims (6)

1. a kind of method obtaining ICI impact factors, it is characterised in that adjacent interference storage unit is obtained ahead of time according to experiment Different data is stored to storage unit Cell influence degrees, the interference storage unit by reading storage unit Cell is actually written into Data, obtain the sub- impact factor of each interference storage unit respectively, be finally overlapped obtain final ICI influence because Son.
2. the method according to claim 1 for obtaining ICI impact factors, it is characterised in that only choose to storage unit Cell Maximum two adjacent storage units are influenced as interference storage unit, and it is practical only to read the two interference storage units selected The data of write-in, and two sub- impact factors are obtained, final ICI is obtained after the two sub- impact factors are overlapped to be influenced The factor.
3. the method according to claim 2 for obtaining ICI impact factors, it is characterised in that the number of interference storage unit storage " 10 " are pressed according to the influence to storage unit Cell>“00”>“01”>" 11 " are distributed from big to small;Storage unit Cell [WLi, BLj] two storage units of ICI impact factors selection Cell [WLi-1, BLj] and Cell [WLi+1, BLj] as interference storage Unit.
4. a kind of method promoting error correcting capability, reliability LLR of the feature in calculating LDPC algorithms according to ICI impact factors (Log Likelihood Ratio) and LUT tables, error correction is iterated using LLR and LUT tables;The acquisition ICI impact factors It is obtained according to following method, is that adjacent interference storage unit, which is obtained ahead of time, according to experiment stores different data to storage unit Cell influence degrees, the data that the interference storage unit by reading storage unit Cell is actually written into obtain each dry respectively The sub- impact factor for disturbing storage unit is finally overlapped and obtains final ICI impact factors.
5. the method according to claim 4 for promoting error correcting capability, it is characterised in that only choose to storage unit Cell shadows Maximum two adjacent storage units are rung as interference storage units, only read two interference selecting storage unit is practical and write Data entered, and obtain two sub- impact factors, obtained after the two sub- impact factors are overlapped final ICI influence because Son.
6. the method according to claim 5 for promoting error correcting capability, it is characterised in that Nand Flash are MLC types, are done Influence of the data of storage unit storage to storage unit Cell is disturbed by " 10 ">“00”>“01”>" 11 " are divided from big to small Cloth;The ICI impact factors selection Cell [WLi-1, BLj] and Cell [WLi+1, BLj] two of storage unit Cell [WLi, BLj] Storage unit is as interference storage unit.
CN201810258934.3A 2018-03-27 2018-03-27 A method of it obtaining ICI impact factors and promotes error correcting capability Pending CN108563533A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810258934.3A CN108563533A (en) 2018-03-27 2018-03-27 A method of it obtaining ICI impact factors and promotes error correcting capability

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810258934.3A CN108563533A (en) 2018-03-27 2018-03-27 A method of it obtaining ICI impact factors and promotes error correcting capability

Publications (1)

Publication Number Publication Date
CN108563533A true CN108563533A (en) 2018-09-21

Family

ID=63533522

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810258934.3A Pending CN108563533A (en) 2018-03-27 2018-03-27 A method of it obtaining ICI impact factors and promotes error correcting capability

Country Status (1)

Country Link
CN (1) CN108563533A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102171767A (en) * 2008-09-30 2011-08-31 Lsi公司 Methods and apparatus for soft data generation for memory devices based on performance factor adjustment
US20130094293A1 (en) * 2011-10-18 2013-04-18 Samsung Electronics Co., Ltd. Memory device and method of reading data from memory device
CN103226974A (en) * 2011-12-15 2013-07-31 马维尔国际贸易有限公司 Inter-cell interference cancellation
CN105679364A (en) * 2014-12-08 2016-06-15 桑迪士克技术有限公司 Rewritable multibit non-volatile memory with soft decode optimization

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102171767A (en) * 2008-09-30 2011-08-31 Lsi公司 Methods and apparatus for soft data generation for memory devices based on performance factor adjustment
US20130094293A1 (en) * 2011-10-18 2013-04-18 Samsung Electronics Co., Ltd. Memory device and method of reading data from memory device
CN103226974A (en) * 2011-12-15 2013-07-31 马维尔国际贸易有限公司 Inter-cell interference cancellation
CN105679364A (en) * 2014-12-08 2016-06-15 桑迪士克技术有限公司 Rewritable multibit non-volatile memory with soft decode optimization

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
MINCHEOL PARK .ETC: "Direct Field Effect of Neighboring Cell Transistor on Cell-to-Cell Interference of NAND Flash Cell Arrays", 《IEEE ELECTRON DEVICE LETTERS》 *
VEERESH TARANALLI .ETC: "Error analysis and inter-cell interference mitigation in multi-level cell flash memories", 《2015 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC)》 *

Similar Documents

Publication Publication Date Title
TWI574277B (en) Error correction operations in a memory device
US9087598B2 (en) Reliability metrics management for soft decoding
TWI569273B (en) Method for read disturbance management in non-volatile memory devices
US8036035B2 (en) Erase cycle counter usage in a memory device
US11163483B2 (en) Robust detection techniques for updating read voltages of memory devices
US8689082B2 (en) Method of operating memory controller, and memory system, memory card and portable electronic device including the memory controller
US7894270B2 (en) Data restoration method for a non-volatile memory
US9478298B2 (en) Memory system and method of reading data thereof
CN108647109A (en) A kind of method that solid state disk promotes LDPC error correcting capabilities
US9342401B2 (en) Selective in-situ retouching of data in nonvolatile memory
US11557353B2 (en) Optimal detection voltage obtaining method, reading control method and apparatus of memory
US9490024B1 (en) Solid state storage device and reading control method thereof
US9430339B1 (en) Method and apparatus for using wear-out blocks in nonvolatile memory
US9053011B2 (en) Selective protection of lower page data during upper page write
CN108154900B (en) Method for relieving write interference problem of MLC flash memory
CN108595286A (en) A kind of method and solid state disk promoting reliability of flash memory
US11513879B2 (en) Detection and mitigation for solid-state storage device read failures due to weak erase
CN108563533A (en) A method of it obtaining ICI impact factors and promotes error correcting capability
JP2022525273A (en) A method for writing to a non-volatile memory device by applying multiple bit line bias voltages
JP5039193B2 (en) Semiconductor memory device and control method
CN117296098A (en) Voltage block group calibration based on temporary voltage shift offset
CN110908825B (en) Data reading method and device, storage equipment and storage medium
US9754685B2 (en) Memory control device and memory control method
JP2022524535A (en) Methods and equipment for the operation of the non-volatile memory mechanism
CN117724878A (en) Data processing method, electronic equipment and computer readable storage device

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20180921

RJ01 Rejection of invention patent application after publication