CN109743062A - A kind of early method of shutting down of low density parity check code ADMM-LP decoding - Google Patents
A kind of early method of shutting down of low density parity check code ADMM-LP decoding Download PDFInfo
- Publication number
- CN109743062A CN109743062A CN201910003123.3A CN201910003123A CN109743062A CN 109743062 A CN109743062 A CN 109743062A CN 201910003123 A CN201910003123 A CN 201910003123A CN 109743062 A CN109743062 A CN 109743062A
- Authority
- CN
- China
- Prior art keywords
- decoding
- admm
- spc
- code word
- shutting down
- 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.)
- Withdrawn
Links
Abstract
The invention discloses a kind of early method of shutting down of low density parity check code ADMM-LP decoding, include the following steps: S1, setting initiation parameter；S2, code word bits hard-decision values are obtained；S3, the verification constraint number that code word is met is calculated；S4, judge l >=IminIt is whether true；S5, judge l > IminIt is whether true；S6, judge l >=IdecAnd NcounterWhether < T sets up simultaneously；S7, judge whether to meet the early condition for stopping decoding；S8, ADMM-LP decoding are early to be stopped.The verification that code word is met when the present invention decodes current iteration according to ADMM-LP constrains number Nspc (l)And then continuous iteration several times when code word Nspc (l)Variation tendency can detect wrong code word in advance and stop decoding, to reduce unnecessary decoding iteration number, reduce decoding complexity, improve decoding convergence rate.
Description
Technical field
The invention belongs to fields of communication technology, in particular to low density parity check code (Low-Density Parity-
Check Codes, LDPC code) it is based on alternating direction multipliers method (Alternating Direction Method of
Multipliers, ADMM) linear programming (Linear Programming, LP) decoding (referred to as ADMM-LP decoding) one
The early method of shutting down of kind low density parity check code ADMM-LP decoding.
Background technique
LDPC code is a kind of packeting error-correcting code, and related calculation amount is little when decoding, and can obtain and approach Shannon
Limit superperformance, hardware realization is simple and convenient, therefore modern communications in real life, in terms of compared
Wide application.ADMM-LP interpretation method with maximum likelihood authentication feature can be improved the existing LP decoding of LDPC code and calculate
The decoding speed of method, but its decoding complexity is still higher.The early method of shutting down of LDPC code iterative decoding passed through in early stage
Stop iteration, decoding speed can be effectively improved.
Currently, only there are two types of decode method of shutting down: the epsilon-move and Hx of standard for the ADMM-LP decoding of LDPC codeT=0 early stops
Only method.In paper " Decomposition methods for large scale LP decoding " (IEEE
Transactions on Information Theory, 2013,59 (12): 7870-7886) in propose standard epsilon-move will
ADMM decoding in every time iteration main residual error (Primal Residual) and antithesis residual error (Dual Residual) respectively with ε (ε
> 0) compare, if the two residual values are both less than ε, decoding iteration stops；And in paper " Efficient iterative LP
decoding of LDPC codes with alternating direction method of multipliers”(IEEE
International Symposium on Information Theory Proceedings (ISIT), Istanbul,
Hx used in 2013:1501-1505)T=0 early method of shutting down is mainly according to all check equations HxTWhether=0 meet to come really
Definite decoding stops.The shortcoming of both method of shutting down be it is higher in the mean iterative number of time decoded compared with low signal-to-noise ratio region,
It is slower to decode convergence rate.
Patented technology " the linear programming interpretation method of high-speed low density parity check code " (authorized announcement date: 2017
November 17, Authorization Notice No.: CN 104682968B) in disclose a kind of linear gauge of high-speed low density parity check code
Draw interpretation method.Interpretation method in the patent application by only update in each iteration a check-node message and
The message of the variable node adjacent with the check-node owns to overcome in the existing each iteration of linear programming interpretation method
The message of check-node and the message of variable node are required to the shortcomings that updating, to accelerate decoding convergence, improve decoding
Speed.But its shortcoming is that the mean iterative number of time in middle low signal-to-noise ratio region is still higher.
Summary of the invention
The purpose of the present invention is to provide a kind of early method of shutting down of low density parity check code ADMM-LP decoding, pass through
The number for calculating the verification constraint that code word in ADMM-LP decoding iteration process is met, judges wrong code word in early stage
And stop decoding, so as to reduce the mean iterative number of time of decoding, decoding convergence rate is improved, to solve above-mentioned background technique
Middle proposition the problems of the prior art.
There is parameter according to the present invention during decoding:
Nspc (l): indicate verification constraint number (the Satisfied parity-check that code word is met
Constraints, abbreviation SPC)；
Ncounter: indicate Nspc (l)The counter of variation tendency；
T: the comparison threshold value of wrong code word is detected；
Imin: statistical counter NcounterInitial iteration number；
Idec: detect the initial iteration number of wrong code word；
D: N is calculatedcounterWhen twice in succession between iteration code word Nspc (l)Difference is non-positive number；
dmax: the N that the present invention allowsspc (l)Difference is non-positive number maximum value.
To achieve the above object, present invention employs following technical solutions:
A kind of early method of shutting down of low density parity check code ADMM-LP decoding, includes the following steps:
S1, setting initiation parameter:
10 are set by the tolerance in LDPC code ADMM-LP decoding-5, overrelaxation parameter is set as 1.9, and code word is met
Verification constraint number difference is non-positive number maximum value dmaxIt is set as 2；
S2, code word bits hard-decision values are obtained:
The vector obtained when decoding the l times iteration to ADMM-LPHard decision operation is executed,
Obtain vectorWhereinThe hard-decision values of each code word bits as after current iteration；
S3, the verification constraint number that code word is met is calculated:
The verification that the code word is met, which is calculated, according to the hard-decision values of code word bits each in step S2 constrains number；
S4, judge l >=IminIt is whether true:
Whether judgement is greater than or equal to statistical counter N as previous the number of iterations lcounterInitial iteration number IminIf
l≥Imin, then willValue be assigned to Ncounter；
S5, judge l > IminIt is whether true:
Whether judgement is greater than statistical counter N as previous the number of iterations lcounterInitial iteration number IminIf l >
Imin, then the N of vector between iteration twice in succession is calculatedspc (l)Difference Nspc (l)-Nspc (l-1)If at this moment difference Nspc (l)-Nspc (l -1)> 0, then by Ncounter+(Nspc (l)-Nspc (l-1)) value be assigned to Ncounter, otherwise, keep NcounterIt is constant；
S6, judge l >=IdecAnd NcounterWhether < T sets up simultaneously:
Judgement detects the initial iteration number I of wrong code word when whether previous the number of iterations l is greater than or equal todec, Yi Jitong
Count counter NcounterWhether the comparison threshold value T for detecting wrong code word is less than, if l >=IdecAnd Ncounter< T, then statistical decoding from
IminAll difference N of secondary to the l times iterationspc (l)-Nspc (l-1)For non-positive number d；
S7, judge whether to meet the early condition for stopping decoding, if so, S8 is thened follow the steps, otherwise, by current iteration number
Step S2 is executed after adding 1；
S8, ADMM-LP decoding are early to be stopped.
Preferably, in the step S2 according toExecute hard decision operation.
Preferably, basis in the step S3Current code word is met when calculating the l times iteration
Verification constrain number Nspc (l), wherein m is the information bit length 288,1 of (576,288) irregular LDPC codesTExpression length is m
Complete 1 row vector, H be (576,288) irregular LDPC codes check matrix.
Preferably, the early stop condition of decoding in the step S7 refers to the situation for meeting one of following three conditions:
Condition 1: when the verification constraint number that previous iterative code word is met is equal to the letter of (576,288) irregular LDPC codes
Cease bit length, i.e. Nspc (l)=m；
Condition 2: difference Nspc (l)-Nspc (l-1)It is more than the non-positive difference number that the method for shutting down allows for non-positive number d
Maximum value dmax, i.e. d > dmax；
Condition 3:ADMM-LP decoding iteration number reaches 40.
The early method of shutting down of a kind of low density parity check code ADMM-LP decoding proposed by the present invention, with prior art phase
Than having the advantage that the verification that code word is met when the present invention decodes current iteration according to ADMM-LP constrains number Nspc (l)
And then continuous iteration several times when code word Nspc (l)Variation tendency can detect wrong code word in advance and stop decoding,
To reduce unnecessary decoding iteration number, decoding complexity is reduced, decoding convergence rate is improved.
Detailed description of the invention
Fig. 1 is implementation flow chart of the invention；
Decoding mean iterative number of time comparison diagram Fig. 2 of the invention when being parameter T variation；
Frame error rate comparison diagram Fig. 3 of the invention when being parameter T variation；
Fig. 4 is IminAnd IdecDecoding mean iterative number of time comparison diagram of the invention when variation；
Fig. 5 is IminAnd IdecFrame error rate comparison diagram of the invention when variation；
Fig. 6 is to decode mean iterative number of time comparison diagram with existing method when the present invention takes optimal parameter to combine；
Fig. 7 is to decode frame error rate comparison diagram with existing method when the present invention takes optimal parameter to combine.
Specific embodiment
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 description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Herein
Described specific embodiment is only used to explain the present invention, is not intended to limit the present invention.Based on the embodiments of the present invention,
Every other embodiment obtained by those of ordinary skill in the art without making creative efforts, belongs to this hair
The range of bright protection.
It is decoded for (576, the 288) irregular LDPC codes for being 0.5 with code rate in IEEE802.16E standard, this hair
It is bright to provide a kind of early method of shutting down of low density parity check code ADMM-LP decoding as shown in Figure 1, include the following steps:
Step 1: setting initiation parameter:
10 are set by the tolerance in LDPC code ADMM-LP decoding-5, overrelaxation parameter is set as 1.9, and the present invention allows
Nspc (l)Difference is non-positive number maximum value dmaxIt is set as 2.
Step 2: obtaining code word bits hard-decision values:
If the vector obtained when ADMM-LP punishment the l times iteration of decoding isAccording toHard decision operation is executed, vector is obtainedWhereinIt is as current
The hard-decision values of each code word bits after iteration.
Step 3: calculating the verification constraint number that code word is met:
According to the hard-decision values of code word bits each in step S2, according toCalculate the l times iteration
When current code word met verification constraint number Nspc (l), wherein m is the information bit length of (576,288) irregular LDPC codes
288,1TIndicate that length is complete 1 row vector of m, H is the check matrix of (576,288) irregular LDPC codes.
Step 4: judging l >=IminIt is whether true:
Whether judgement is greater than or equal to statistical counter N as previous the number of iterations lcounterInitial iteration number IminIf
l≥Imin, then willValue be assigned to Ncounter。
Step 5: judging l > IminIt is whether true:
Whether judgement is greater than statistical counter N as previous the number of iterations lcounterInitial iteration number IminIf l >
Imin, then the N of vector between iteration twice in succession is calculatedspc (l)Difference Nspc (l)-Nspc (l-1)If at this moment difference Nspc (l)-Nspc (l -1)> 0, then by Ncounter+(Nspc (l)-Nspc (l-1)) value be assigned to Ncounter, otherwise, keep NcounterIt is constant；
Step 6: judging l >=IdecAnd NcounterWhether < T sets up simultaneously:
Judgement detects the initial iteration number I of wrong code word when whether previous the number of iterations l is greater than or equal todec, Yi Jitong
Count counter NcounterWhether the comparison threshold value T for detecting wrong code word is less than, if l >=IdecAnd Ncounter< T, then statistical decoding from
IminAll difference N of secondary to the l times iterationspc (l)-Nspc (l-1)For non-positive number d.
Step 7: judging whether to meet the early condition for stopping decoding, if so, S8 is thened follow the steps, otherwise, by current iteration
Number executes step S2 after adding 1.
The early stop condition of the decoding refers to the situation for meeting one of following three conditions:
Condition 1: when the verification constraint number that previous iterative code word is met is equal to the letter of (576,288) irregular LDPC codes
Cease bit length, i.e. Nspc (l)=m.
Condition 2: difference Nspc (l)-Nspc (l-1)It is more than the non-positive difference number that the method for shutting down allows for non-positive number d
Maximum value dmax, i.e. d > dmax。
Condition 3:ADMM-LP decoding iteration number reaches 40.
Stop Step 8: ADMM-LP decoding is early.
Decoding effect of the invention is further described with reference to the accompanying drawing:
Emulation content: under additive white Gaussian noise awgn channel, using ADMM punishment interpretation method to IEEE802.16E
(576,288) irregular LDPC codes that code rate is 0.5 in standard are emulated, and decoding stop condition is then respectively morning of the invention
Two kinds of method of shutting down of method of shutting down and existing ADMM-LP decoding.During decoding, using in " The ADMM penalized
decoder for LDPC codes”(IEEE Transactions on Information Theory,2016,62(6):
L 2966-2984) mentioned in paper1Penalty function, and by l1The punishment parameter α of penalty function is set as 1.5, augmentation Lagrange letter
Several punishment parameter μ are set as 4.1.
Currently, the ADMM-LP decoding of LDPC code is there are two types of method of shutting down is decoded, one is in paper " Decomposition
methods for large scale LP decoding”(IEEE Transactions on Information Theory,
2013,59 (12): 7870-7886) in propose standard epsilon-move, the method for shutting down by ADMM decode in every time iteration master
Residual error (Primal Residual) and antithesis residual error (Dual Residual) are respectively compared with ε (ε > 0), if the two residual errors
Value is both less than ε, then decoding iteration stops；Another kind is in paper " Efficient iterative LP decoding of
LDPC codes with alternating direction method of multipliers”(IEEE
International Symposium on Information Theory Proceedings (ISIT), Istanbul,
Hx used in 2013:1501-1505)T=0 early method of shutting down, this method finally verify check equations Hx in each iterationT=0
It is whether true, stopping is decoded if equation is set up, wherein H is the check matrix of LDPC code, and x is the hard-decision values of code word bits.
Decoding mean iterative number of time and frame error rate comparison diagram Fig. 2 and Fig. 3 of the invention when being parameter T variation.Fig. 2 is two dimension
Plan view, abscissa are signal-to-noise ratio, and ordinate is mean iterative number of time, of the invention when taking different value it gives parameter T
Decode mean iterative number of time.Fig. 3 is 2 d plane picture, and abscissa is signal-to-noise ratio, and ordinate is frame error rate, and it gives parameters
Frame error rate T of the invention when taking different value.Wherein:
The early method of shutting down of ADMM-LP decoding proposed by the present invention is respectively indicated in Fig. 2 and Fig. 3 with the curve that circle identifies
Mean iterative number of time and frame error rate when taking T=260；
The early method of shutting down of ADMM-LP decoding proposed by the present invention is respectively indicated in Fig. 2 and Fig. 3 with the curve that cross identifies
Mean iterative number of time and frame error rate when taking T=270；
The early method of shutting down of ADMM-LP decoding proposed by the present invention is respectively indicated in Fig. 2 and Fig. 3 with the curve that asterisk identifies
Mean iterative number of time and frame error rate when taking T=280.
Using decoding mean iterative number of time when three kinds of difference T values in high s/n ratio region it can be seen from Fig. 2 and Fig. 3
It is roughly the same, and the mean iterative number of time decoded in low signal-to-noise ratio region T=280 is minimum, however it decodes frame error rate instead
It is worst.Decoding mean iterative number of time as T=260 and T=270 is essentially identical, and in low signal-to-noise ratio region T=270
Frame error rate is preferable.Therefore, we choose parameter value of the T=270 as early method of shutting down of the invention.
Fig. 4 and Fig. 5 is IminAnd IdecDecoding mean iterative number of time and frame error rate comparison diagram of the invention when variation.Fig. 4 is
2 d plane picture, abscissa are signal-to-noise ratio, and ordinate is mean iterative number of time, and it gives parameter IminAnd IdecTake different value
When decoding mean iterative number of time of the invention.Fig. 5 is 2 d plane picture, and abscissa is signal-to-noise ratio, and ordinate is frame error rate, the figure
Give parameter IminAnd IdecFrame error rate of the invention when taking different value.Wherein:
The early method of shutting down of ADMM-LP decoding proposed by the present invention is respectively indicated in Fig. 4 and Fig. 5 with the curve that circle identifies
Take (Imin=10, Idec=15) mean iterative number of time and frame error rate when；
The early method of shutting down of ADMM-LP decoding proposed by the present invention is respectively indicated in Fig. 4 and Fig. 5 with the curve that cross identifies
Take (Imin=15, Idec=20) mean iterative number of time and frame error rate when；
The early method of shutting down of ADMM-LP decoding proposed by the present invention is respectively indicated in Fig. 4 and Fig. 5 with the curve that asterisk identifies
Take (Imin=20, Idec=25) mean iterative number of time and frame error rate when.
The parameter I it can be seen from Fig. 4 and Fig. 5minAnd IdecThree groups of difference values decoding mean iterative number of time height believe
It makes an uproar more roughly the same than region, and it is different in low signal-to-noise ratio region, wherein taking parameter value (Imin=20, Idec=25) it is decoded when
Mean iterative number of time highest, take parameter value (Imin=10, Idec=15) mean iterative number of time decoded when is minimum；Three groups different
The frame error rate of value is almost the same in high s/n ratio region, and chooses (I in low signal-to-noise ratio regionmin=15, Idec=20) mistake
Therefore, we choose (I to the preferable of frame per secondmin=15, Idec=20) parameter value as early method of shutting down of the invention.
By emulation, the optimal combination parameter that we obtain code (576,288) irregular LDPC codes is T=270, Imin=
15 and Idec=20.
Fig. 6 and Fig. 7 is to decode mean iterative number of time and frame error rate pair with existing method when the present invention takes optimal parameter to combine
Than figure.Fig. 6 is 2 d plane picture, and abscissa is signal-to-noise ratio, and ordinate is mean iterative number of time, and it gives the present invention to take most
When good parameter combination compared with the decoding mean iterative number of time of existing method.Fig. 7 is 2 d plane picture, and abscissa is signal-to-noise ratio,
Ordinate is frame error rate, when taking optimal parameter to combine it gives the present invention compared with the frame error rate of existing method.Wherein:
The Average Iteration of the epsilon-move method of shutting down of existing standard is respectively indicated in Fig. 6 and Fig. 7 with the curve that circle identifies
Several and frame error rate；
Existing Hx is respectively indicated with the curve that asterisk identifies in Fig. 6 and Fig. 7TThe mean iterative number of time of=0 early method of shutting down
And frame error rate；
The early method of shutting down of ADMM-LP decoding proposed by the present invention is respectively indicated in Fig. 6 and Fig. 7 with the curve that asterisk identifies
Mean iterative number of time and frame error rate when optimal parameter being taken to combine.
Frame error rate it can be seen from Fig. 6 and Fig. 7 when decoding using three kinds of method of shutting down is roughly the same, and it is average repeatedly
Generation number is then significantly different.Compared with low signal-to-noise ratio region HxTThe ε that the mean iterative number of time of=0 early method of shutting down is higher than standard is advised
Then, this is because it is usually more compared with low signal-to-noise ratio zone errors code word, and HxT=0 early method of shutting down needs more iteration time
Number could decision error code word.And early method of shutting down of the invention overcomes HxTThis disadvantage of=0 early method of shutting down, compared with
Low signal-to-noise ratio region can obviously reduce the mean iterative number of time of LDPC code decoding, and decoding performance is almost without loss.
Finally, it should be noted that the foregoing is only a preferred embodiment of the present invention, it is not intended to restrict the invention,
Although the present invention is described in detail referring to the foregoing embodiments, for those skilled in the art, still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features,
All within the spirits and principles of the present invention, any modification, equivalent replacement, improvement and so on should be included in of the invention
Within protection scope.
Claims (10)
1. a kind of early method of shutting down of low density parity check code ADMM-LP decoding, which comprises the steps of:
S1, setting initiation parameter；
S2, code word bits hard-decision values are obtained；
S3, the verification constraint number that code word is met is calculated；
S4, judge l >=IminIt is whether true；
S5, judge l > IminIt is whether true；
S6, judge l >=IdecAnd NcounterWhether < T sets up simultaneously；
S7, judge whether to meet the early condition for stopping decoding, otherwise be held after current iteration number is added 1 if so, executing S8
Row S2；
S8, ADMM-LP decoding are early to be stopped.
2. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 1, feature exist
In: the specific steps of initiation parameter are set in the S1 are as follows: set 10 for the tolerance in LDPC code ADMM-LP decoding-5,
Overrelaxation parameter is set as 1.9, and it is non-positive number maximum value d that code word, which meets verification constraint number difference,maxIt is set as 2.
3. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 1, feature exist
In: in the S2 obtain code word bits hard-decision values specific steps are as follows: to ADMM-LP decode the l times iteration when obtain to
AmountHard decision operation is executed, vector is obtainedWhereinAs work as
The hard-decision values of each code word bits after preceding iteration.
4. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 3, feature exist
In: the vector that the ADMM-LP is obtained when decoding the l times iterationAccording toExecute hard decision operation.
5. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 1, feature exist
In: the specific steps for the verification constraint number that code word is met are calculated in the S3 are as follows: according to the hard of code word bits each in S2
Decision value calculates the verification constraint number that the code word is met.
6. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 5, feature exist
In: the verification constraint number that described calculating code word is met is basisWork as when calculating the l times iteration
The verification that preceding code word is met constrains number Nspc (l), m therein is the information bit length of (576,288) irregular LDPC codes
288,1TIndicate that length is complete 1 row vector of m, H is the check matrix of (576,288) irregular LDPC codes.
7. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 1, feature exist
In: l >=I is judged in the S4minWhether true specific steps are as follows: whether judgement is greater than or equal to system as previous the number of iterations l
Count counter NcounterInitial iteration number IminIf l >=Imin, then by Nspc(Imin) value is assigned to Ncounter。
8. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 1, feature exist
In: l > I is judged in the S5minWhether true specific steps are as follows: whether judgement is greater than statistical counting as previous the number of iterations l
Device NcounterInitial iteration number IminIf l > Imin, then the N of vector between iteration twice in succession is calculatedspc (l)Difference Nspc (l)-Nspc (l-1)If at this moment difference Nspc (l)-Nspc (l-1)> 0, then by Ncounter+(Nspc (l)-Nspc (l-1)) value be assigned to
Ncounter, otherwise, keep NcounterIt is constant.
9. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 1, feature exist
In: l >=I is judged in the S6decAnd NcounterThe specific steps whether < T sets up simultaneously are as follows: previous the number of iterations l is worked as in judgement
Whether the initial iteration number I for detecting wrong code word is greater than or equal todecAnd statistical counter NcounterWhether detection is less than
The comparison threshold value T of mistake code word, if l >=IdecAnd Ncounter< T, then statistical decoding is from IminSecondary to the l times iteration it is all
Difference Nspc (l)-Nspc (l-1)For non-positive number d.
10. a kind of early method of shutting down of low density parity check code ADMM-LP decoding according to claim 1, feature
Be: the early stop condition of decoding in the S7 refers to the situation for meeting one of following three conditions:
Condition 1: when the verification constraint number that previous iterative code word is met is equal to the information bit of (576,288) irregular LDPC codes
Length, i.e. Nspc (l)=m；
Condition 2: difference Nspc (l)-Nspc (l-1)It is more than the maximum for the non-positive difference number that the method for shutting down allows for non-positive number d
Value dmax, i.e. d > dmax；
Condition 3:ADMM-LP decoding iteration number reaches 40.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910003123.3A CN109743062A (en) | 2019-01-03 | 2019-01-03 | A kind of early method of shutting down of low density parity check code ADMM-LP decoding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910003123.3A CN109743062A (en) | 2019-01-03 | 2019-01-03 | A kind of early method of shutting down of low density parity check code ADMM-LP decoding |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109743062A true CN109743062A (en) | 2019-05-10 |
Family
ID=66363084
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910003123.3A Withdrawn CN109743062A (en) | 2019-01-03 | 2019-01-03 | A kind of early method of shutting down of low density parity check code ADMM-LP decoding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109743062A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120054576A1 (en) * | 2010-08-25 | 2012-03-01 | The Royal Institution For The Advancement Of Learning / Mcgill University | Method and system for decoding |
| US20140089759A1 (en) * | 2012-09-26 | 2014-03-27 | Stark Draper | Error Correction Circuit for Data Communication Providing Parallelizable Linear Programming Decoding |
| CN107689801A (en) * | 2017-09-07 | 2018-02-13 | 西安电子科技大学 | The early method of shutting down of LDPC code ADMM iterative decodings |
| CN108199721A (en) * | 2017-12-22 | 2018-06-22 | 西安电子科技大学 | Low density parity check code linear programming interpretation method based on BADMM |
| CN108964669A (en) * | 2018-07-06 | 2018-12-07 | 西安电子科技大学 | LDPC code quadratic programming interpretation method based on degree decomposition and alternating multiplier method |
-
2019
- 2019-01-03 CN CN201910003123.3A patent/CN109743062A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120054576A1 (en) * | 2010-08-25 | 2012-03-01 | The Royal Institution For The Advancement Of Learning / Mcgill University | Method and system for decoding |
| US20140089759A1 (en) * | 2012-09-26 | 2014-03-27 | Stark Draper | Error Correction Circuit for Data Communication Providing Parallelizable Linear Programming Decoding |
| CN107689801A (en) * | 2017-09-07 | 2018-02-13 | 西安电子科技大学 | The early method of shutting down of LDPC code ADMM iterative decodings |
| CN108199721A (en) * | 2017-12-22 | 2018-06-22 | 西安电子科技大学 | Low density parity check code linear programming interpretation method based on BADMM |
| CN108964669A (en) * | 2018-07-06 | 2018-12-07 | 西安电子科技大学 | LDPC code quadratic programming interpretation method based on degree decomposition and alternating multiplier method |
Non-Patent Citations (1)
| Title |
|---|
| 王彪等: "LDPC码ADMM惩罚译码的早停止方法", 《西安电子科技大学学报》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8209579B2 (en) | Generalized multi-threshold decoder for low-density parity check codes | |
| CN110995278B (en) | Improved polarity code serial elimination list bit overturning decoding method and system | |
| CN107528597B (en) | LDPC code post-processing decoding method based on CRC code | |
| CN100578944C (en) | Alternative decoding method and device for LDPC code | |
| US7454684B2 (en) | Apparatus and method for turbo decoder termination | |
| US11601138B2 (en) | Decoding method of LDPC codes based on partial average residual belief propagation | |
| CN107689801B (en) | The early method of shutting down of LDPC code ADMM iterative decoding | |
| CN105577193A (en) | Loop-break based mixed weighted bit-flipping LDPC decoding method | |
| CN104009763A (en) | Ahead-of-time stopping method of low complexity LDPC code weighting bit flipping decoding algorithm | |
| CN105530014A (en) | LDPC code alternating direction multiplier decoding method based on simplified projection operator | |
| Pillet et al. | On list decoding of 5G-NR polar codes | |
| Sun et al. | Error patterns in belief propagation decoding of polar codes and their mitigation methods | |
| EP3540949A1 (en) | Bit flipping decoding for ldpc codes | |
| CN109743062A (en) | A kind of early method of shutting down of low density parity check code ADMM-LP decoding | |
| US8019020B1 (en) | Binary decoding for correlated input information | |
| CN111726202B (en) | Early termination iteration method for polarization code belief propagation decoding | |
| Lin et al. | A fast iterative check polytope projection algorithm for ADMM decoding of LDPC codes by bisection method | |
| Li et al. | LDPC design for 5G NR URLLC & mMTC | |
| CN101707486A (en) | LDPC decryption method of multi-state belief propagation (BP) iteration with unidirectional rectification | |
| Dai et al. | Noisy gradient descent bit-flipping decoder based on adjustment factor for LDPC codes | |
| CN115664432A (en) | Adaptive iteration number method and system of decoder | |
| CN114759930A (en) | LP early-stop decoding method based on LDPC (Low Density parity check) code of alternating direction multiplier method | |
| Geldmacher et al. | Hard decision based low SNR early termination for LTE Turbo decoding | |
| CN113285722A (en) | Multi-deviation segmented redundancy check auxiliary statistical decoding method for short polarization code | |
| Gendron et al. | Revisiting augmented decoding techniques for LTE Turbo Codes |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190510 |