CN101964690A - HARQ merged decoding method, device and system - Google Patents
HARQ merged decoding method, device and system Download PDFInfo
- Publication number
- CN101964690A CN101964690A CN2009100554020A CN200910055402A CN101964690A CN 101964690 A CN101964690 A CN 101964690A CN 2009100554020 A CN2009100554020 A CN 2009100554020A CN 200910055402 A CN200910055402 A CN 200910055402A CN 101964690 A CN101964690 A CN 101964690A
- Authority
- CN
- China
- Prior art keywords
- data
- harq
- cut position
- time
- merges
- 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
Images
Abstract
The invention discloses an HARQ (hybrid automatic repeat request) merged decoding method, which comprises the following steps of: sampling data subjected to first quantified interception and HARQ merging processing, calculating the statistical property of the sampled data, and determining a scaling factor according to the statistical property; and performing second quantified interception processing on the data after the HARQ merging processing according to the scaling factor, and outputting the processing result for subsequent decoding processing. Meanwhile, the invention also discloses an HARQ merged decoding device and an HARQ merged decoding system, which can ensure the performance of a decoder under a multi-carrier system environment.
Description
Technical field
The present invention relates to communication technical field, relate in particular to a kind of HARQ and merge interpretation method, Apparatus and system.
Background technology
At existing single-carrier system, as Time Division-Synchronous Code Division Multiple Access (TD-SCDMA, Time-DivisionSynchronization Code Division-Multiple-Access) in the system, because the characteristic of channel is constant in the same subframe of hypothesis, so it is less to input to the data fluctuations amplitude of decoder, automatic repeat requests (HARQ can mixed, Hybrid Automatic Repeat Request) merges the back and use single zoom factor, make that fixed point Turbo decoder does not all have significantly loss with respect to the floating-point decoder performance under all reference channels.
As shown in Figure 1, for HARQ in the prior art merges and the quantification treatment flow process, process is: at first according to the modulation system of fixing zoom factor or demodulation module the input data are made to quantize cut position; Then HARQ is merged data based its process number of module output and the spatial cache of control module distribution and in buffer memory (Buffer) module, store soft Bit data; Soft Bit data after will merging is again decoded as the input of decoder; At last Cyclical Redundancy Check (CRC, the CyclicalRedundancy Check) assay after the Turbo decoding is fed back to the Buffer module, as successfully decoded, the buffer memory that then this code block is taken discharges; As decoding failure, then in the Buffer module, keep the soft Bit data after the merging of this code block, and according to the requirement of HARQ control module soft Bit data is exported to HARQ merges module and data retransmission merges.
The Turbo decoder is an error correction coding commonly used in the present communication system, because itself data volume and decoding complexity, its hardware is realized directly influence time-delay and internal memory cost.The bit resolution of Turbo decoder input influences its performance, and the bit number of decoder intercepting is the balance of doing between decoding performance and overhead and trades off.When the significant bit length of input data reduces, cause the entrained soft Bit data partial loss of input data, decoding performance also can produce corresponding decline.Similarly,, represent, can cause losing of soft Bit data equally with the bound of fixed-point number because the restriction of number of bits makes the soft Bit data that originally is distributed in outside the represented scope of fixed-point number by saturated.The general quantification still can keep the performance close with floating-point under back reservation 4～6 bit data conditions.
For multicarrier system, as the LTE system, it need provide the peak rate of 100M/s, physical down link sharing channel (PDSCH, Physical Downlink Shared Channel) use Turbo code to encode as forward error correction (FEC, Forward Error Correction).This case inventor is through discovering, if directly adopting above-mentioned prior art in the LTE system handles, because under the multicarrier condition, each subcarrier place channel response differs greatly, and there is frequency selective fading, cause the data amplitude of demodulation module output to alter a great deal, when HARQ merged module and FEC decoding module work fixed point quantification cut position, single zoom factor can not satisfy performance requirement under all reference channels.That is to say that the prior art scheme can't guarantee the Turbo decoder performance under the various reference channels when demodulator output work 4 bit cut positions are directly inputted into the Turbo decoder.In addition, in the LTE system, use the prior art scheme to mean that also needs increase the number of bits of Turbo decoder input data, because the peak value downstream rate of LTE system is designed to 100Mb/s, throughput has increased by one more than the order of magnitude than the TD-SCDMA system, decoder is imported one of every increase all needs to distribute a large amount of extra internal memories, obviously, prior art can't satisfy the demand of LTE system.
Summary of the invention
The embodiment of the invention provides a kind of HARQ to merge interpretation method, Apparatus and system, can guarantee the decoder performance under the multicarrier system environment.
The embodiment of the invention provides following technical scheme:
A kind of HARQ merges interpretation method, comprising:
To through quantizing for the first time the statistical property that cut position and HARQ merge the data do sampling after handling and calculate data from the sample survey, determine zoom factor according to described statistical property;
According to described zoom factor the data that described HARQ merges after handling are carried out quantizing the second time cut position processing, result output is handled for follow-up decoding.
Preferably, described to merging the statistical property that data after handling are done sampling and calculated data from the sample survey through quantizing for the first time cut position and HARQ, determine that according to described statistical property the step of zoom factor comprises:
Calculate the statistical property of data from the sample survey;
Probability distribution according to described statistical property estimation sampling data;
Search the emulation form that presets according to pre-set criteria and obtain zoom factor.
Preferably, described statistical property comprises variance or probability density distribution.
Preferably, quantize the bit number of intercepting in the cut position processing the described first time greater than the bit number that quantizes intercepting in the cut position processing the described second time.
Preferably, the described step that result output is handled for follow-up decoding comprises:
If successfully decoded, then abandon the Bit data before quantize cut position the second time;
If decoding failure then keeps the Bit data before quantize cut position the second time, be used for once more the data that HARQ merges when handling and data retransmission merges;
If number of retransmissions has surpassed the number of times of regulation, then no matter decode results is success or failure, all abandons to quantize the preceding Bit data of cut position the described second time.
A kind of HARQ merges code translator, comprising:
First processing module is used for merging the statistical property that data after handling are done sampling and calculated data from the sample survey through quantizing for the first time cut position and HARQ, and determines zoom factor according to described statistical property;
Second processing module is used for according to described zoom factor the data that described HARQ merges after handling being carried out quantizing the second time cut position processing, and result output is handled for follow-up decoding.
Preferably, described first processing module comprises:
Sampling unit is used for sampling through the data that quantize for the first time after cut position and HARQ merging is handled;
Computing unit is used to calculate the statistical property of data from the sample survey;
Estimation unit is used for the probability distribution according to described statistical property estimation sampling data;
Acquiring unit is used for searching the emulation form that presets according to pre-set criteria and obtains zoom factor.
Preferably, described HARQ merging code translator also comprises:
The 3rd processing module, be used for if successfully decoded, the Bit data that quantizes the described second time before the cut position abandoned, if decoding failure, the Bit data that quantizes the described second time before the cut position is kept, be used for carrying out once more HARQ and merge when handling data with the data retransmission merging; And determine that number of retransmissions has surpassed the number of times of regulation, and no matter decode results is success or failure, the Bit data that is all quantized the described second time before the cut position abandons.
A kind of HARQ merges decoding system, comprises that aforesaid HARQ merges code translator.
The HARQ that the embodiment of the invention provides merges interpretation method, Apparatus and system, can guarantee the decoder performance under the multicarrier system environment.To different input data by the choice of dynamical zoom factor, so that data after demodulating is lost soft Bit data when doing to quantize less as far as possible.Simultaneously, technical scheme provided by the invention can adapt to all reference channels, and the loss of floating-point decoder performance is less relatively.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the existing techniques in realizing block diagram;
Fig. 2 is that the HARQ that the embodiment of the invention provides merges the interpretation method overview flow chart;
Fig. 3 is that the HARQ that the embodiment of the invention provides merges the interpretation method particular flow sheet;
Fig. 4 is the realization block diagram of concrete application example provided by the invention;
Fig. 5 is that the HARQ that the embodiment of the invention provides merges the code translator structural representation;
Fig. 6 is the first processing module structural representation that the embodiment of the invention provides.
Embodiment
The embodiment of the invention provides a kind of HARQ to merge interpretation method, Apparatus and system, can guarantee the decoder performance under the multicarrier system environment.For making purpose of the present invention, technical scheme and advantage clearer, the embodiment that develops simultaneously with reference to the accompanying drawings, the present invention is described in more detail.
As shown in Figure 2, the HARQ that provides for the embodiment of the invention merges the interpretation method overview flow chart.Described method comprises:
As shown in Figure 3, the HARQ that provides for the embodiment of the invention merges the interpretation method particular flow sheet.Described method specifically comprises:
Particularly, to carrying out quantizing the first time cut position, for example quantize to keep in the cut position quantitative information of M bit according to fixing zoom factor in the first time through the data after the demodulation.Need to prove that the specific implementation of this step can be with reference to prior art, promptly all adopt identical zoom factor under different data inputs and channel, detailed process is not done too much explanation herein.
Particularly, will merge with the data in the buffer memory through quantizing for the first time the data that cut position handles, the data after the merging are stored in the buffer memory when exporting, for the usefulness of merging next time.Need to prove herein,, that is to say that also do not have data in the buffer memory this moment if the data after this process demodulation are to send for the first time then do not need to do to merge to handle.
The statistical property of step 304, calculating data from the sample survey;
Particularly, the statistical property of calculating data from the sample survey can be to calculate variance, probability density distribution (PDF) or the cumulative distribution (CDF) of data from the sample survey.
Particularly, described pre-set criteria can be set according to actual needs, better performances when 25% of the data after for example will merging place the saturation region.
The described emulation form that presets can be by distributing at different pieces of information, and the emulation testing of Turbo fixed point decoder performance obtains, and comprises limited index value and the zoom factor corresponding with this index value.Wherein, definite process of this index value can also be according to the quantity of different increase in demand index values, to improve the accuracy that zoom factor is chosen except the data decision by analogue simulation; Perhaps reduce the quantity of index value, to reduce the expense of the memory space that this form causes.
Search the process that this form obtains zoom factor and be with reference to calculating the statistical property that data from the sample survey obtains, find from form and the immediate index value of statistical property, and choose corresponding zoom factor, for example zoom factor is 1/4.
Particularly, quantize to keep in the cut position processing quantitative information of N bit the described second time, wherein, the quantitative information of the N bit of this intercepting must be less than the quantitative information of the M bit that quantizes for the first time to keep in the cut position processing.Why have such restriction to be because after quantizing for the first time, may the data that repeatedly retransmit be merged, thus input for the first time quantize the bit number N that keeps must be less than M, otherwise may when merging, cause decoding performance to descend by drop-out.
Particularly, the CRC assay after the decoding is fed back to buffer memory,, then abandon the Bit data before quantize cut position the second time, just the shared space in buffer memory of the data after the described merging is discharged if successfully decoded; As decoding failure, the temporary transient Bit data (data after promptly merging) that quantizes for the second time before the cut position that keeps in buffer memory then merges when handling and data retransmission merges for HARQ once more.If number of retransmissions has surpassed the number of times of regulation, then no matter decode results is success or failure, and the Bit data that all can be quantized the described second time before the cut position abandons.
Technical scheme for a better understanding of the present invention is described further below by an instantiation.
As shown in Figure 4, be the realization block diagram of concrete application example provided by the invention.Detailed process is as follows:
The soft Bit data of input after the demodulation is designated as A
0, A
1..., A
k, to each A
iDo to quantize cut position, keep 6 Bit datas, for example with A
iRound after enlarging several times, will be saturated greater than 31 value be 31, the value less than-32 is saturated to be-32.The soft Bit data that quantize behind the cut position first time is designated as a
1, a
2... a
k, export HARQ to and merge module.
The soft Bit data that quantizes the first time behind the cut position merges in the data that HARQ merges in module and the buffer memory, and the data of establishing in the buffer memory are b
1, b
2... b
k, with a
1, a
2... a
kObtain c after merging with it
1, c
2... c
k, for example can be directly with a
iWith b
iAddition makes c
i=a
i+ b
i, i=1,2 ..k is with c
1, c
2... c
kStore in the buffer memory, and the module after outputing to.Need to prove, if the soft Bit data A of the input after this demodulation
0, A
1..., A
kBe to send for the first time, then do not need to do to merge to handle herein.For the data with a collection of transmission, if last decoding module decode results mistake, then the soft Bit data before this time decoding will be retained in the buffer memory and can not be dropped, and the data of this batch transmission afterwards can be retransmitted.The data that re-transmission receives HARQ merge in the module with before the soft Bit data of decoding failure reattempt decoding after doing to merge, the number of times that the process of this re-transmission repeats is according to regulation.
To the data c after the HARQ merging
1, c
2... c
kP data from the sample survey extracted in sampling
, calculate statistical properties such as its variances sigma, estimate c according to σ or other statistical properties then
1, c
2... c
kProbability distribution, according to pre-determined criterion, better performances when 25% of the data after for example will merging place the saturation region with reference to the form that emulated data obtains, is made index by σ, searches c
1, c
2... c
kDo to quantize the zoom factor that cut position adopts for the second time, the zoom factor of for example finding is 1/4.
The 1/4 couple of c of zoom factor that obtains according to lookup table
1, c
2... c
kDeal with, then keep all c
iHigh four, be about to all c
1, c
2... c
kMove to right two and obtain d
1, d
2... d
k, just all data all be multiply by 1/4.With d
1, d
2... d
kDecoding module trial and error decoding after inputing to.If decoding module is successfully decoded, then this transmission block receives, and discharges c
1, c
2... c
kShared space in buffer memory; If the decoding module decoding failure then keeps c in buffer memory
1, c
2... c
k, when handling this transmission block next time, the c that keeps in the buffer memory
1, c
2... c
kPromptly be equivalent to be used in the abovementioned steps the data cached b that merges with new data
1, b
2... b
kIf number of retransmissions has surpassed the number of times of regulation, so no matter whether the decoding module decode results is correct, all will discard data from the sample survey c in buffer memory
1, c
2... c
kNeed to prove that described decoding module adopts the Turbo decoder usually.
As shown in Figure 5, the HARQ that provides for the embodiment of the invention merges the code translator structural representation.This device comprises: first processing module 510, second processing module 520; Wherein,
Described first processing module 510 is used for merging the statistical property that data after handling are done sampling and calculated data from the sample survey through quantizing for the first time cut position and HARQ, and determines zoom factor according to described statistical property;
Described second processing module 520 is used for according to described zoom factor the data that described HARQ merges after handling being carried out quantizing the second time cut position processing, and result output is handled for follow-up decoding.
Particularly, as shown in Figure 6, described first processing module 510 comprises: sampling unit 610, computing unit 620, estimation unit 630, acquiring unit 640; Wherein,
The data that 610 pairs of processes of described sampling unit quantize after cut position and the HARQ merging processing are for the first time sampled; Described computing unit 620 calculates the statistical property of data from the sample survey; Particularly, the statistical property of calculating data from the sample survey can be to calculate variance, probability density distribution (PDF) or the cumulative distribution (CDF) of data from the sample survey.Described estimation unit 630 is according to the probability distribution of described statistical property estimation sampling data; Described acquiring unit 640 is searched the emulation form that presets according to pre-set criteria and is obtained zoom factor.
In addition, described HARQ merges decoding and can further include: the 3rd processing module 530, be used for if successfully decoded, the Bit data that quantizes the described second time before the cut position is abandoned, if decoding failure, the Bit data that quantizes the described second time before the cut position is kept, be used for carrying out once more HARQ and merge when handling data with the data retransmission merging; And determine that number of retransmissions has surpassed the number of times of regulation, and no matter decode results is success or failure, the Bit data that is all quantized the described second time before the cut position abandons.
Need to prove that the bit number that quantizes intercepting in the cut position processing the described first time is greater than the bit number that quantizes intercepting in the cut position processing the described second time.
The embodiment of the invention also further provides a kind of HARQ to merge decoding system, comprises that the described HARQ of the foregoing description merges code translator, specifies and sees the foregoing description for details, repeats no more herein.
One of ordinary skill in the art will appreciate that and realize that all or part of step that the foregoing description method is carried is to instruct relevant hardware to finish by program, described program can be stored in a kind of computer-readable recording medium, this program comprises one of step or its combination of method embodiment when carrying out.
In addition, each functional unit in each embodiment of the present invention can be integrated in the processing module, also can be that the independent physics in each unit exists, and also can be integrated in the module two or more unit.Above-mentioned integrated module both can adopt the form of hardware to realize, also can adopt the form of software function module to realize.If described integrated module realizes with the form of software function module and during as independently production marketing or use, also can be stored in the computer read/write memory medium.
The above-mentioned storage medium of mentioning can be a read-only memory, disk or CD etc.
In sum, this paper provides a kind of HARQ to merge interpretation method, Apparatus and system, can guarantee the decoder performance under the multicarrier system environment.To different input data by the choice of dynamical zoom factor, so that data after demodulating is lost soft Bit data when doing to quantize less as far as possible.Simultaneously, technical scheme provided by the invention can adapt to all reference channels, and the loss of floating-point decoder performance is less relatively.
More than to HARQ provided by the present invention merge interpretation method, Apparatus and system is described in detail, used specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment just is used for helping to understand the solution of the present invention; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (10)
1. a HARQ merges interpretation method, it is characterized in that, comprising:
To through quantizing for the first time the statistical property that cut position and HARQ merge the data do sampling after handling and calculate data from the sample survey, determine zoom factor according to described statistical property;
According to described zoom factor the data that described HARQ merges after handling are carried out quantizing the second time cut position processing, result output is handled for follow-up decoding.
2. HARQ according to claim 1 merges interpretation method, it is characterized in that, described to merging the statistical property that data after handling are done sampling and calculated data from the sample survey through quantizing for the first time cut position and HARQ, determine that according to described statistical property the step of zoom factor comprises:
Calculate the statistical property of data from the sample survey;
Probability distribution according to described statistical property estimation sampling data;
Search the emulation form that presets according to pre-set criteria and obtain zoom factor.
3. HARQ according to claim 2 merges interpretation method, it is characterized in that described statistical property comprises variance or probability density distribution.
4. HARQ according to claim 1 merges interpretation method, it is characterized in that, the bit number that quantizes intercepting in the cut position processing the described first time is greater than the bit number that quantizes intercepting in the cut position processing the described second time.
5. HARQ according to claim 1 merges interpretation method, it is characterized in that, the described step that result output is handled for follow-up decoding comprises:
If successfully decoded, then abandon the Bit data before quantize cut position the second time;
If decoding failure then keeps the Bit data before quantize cut position the second time, be used for once more the data that HARQ merges when handling and data retransmission merges;
If number of retransmissions has surpassed the number of times of regulation, then no matter decode results is success or failure, all abandons to quantize the preceding Bit data of cut position the described second time.
6. a HARQ merges code translator, it is characterized in that, comprising:
First processing module is used for merging the statistical property that data after handling are done sampling and calculated data from the sample survey through quantizing for the first time cut position and HARQ, and determines zoom factor according to described statistical property;
Second processing module is used for according to described zoom factor the data that described HARQ merges after handling being carried out quantizing the second time cut position processing, and result output is handled for follow-up decoding.
7. HARQ according to claim 6 merges code translator, it is characterized in that described first processing module comprises:
Sampling unit is used for sampling through the data that quantize for the first time after cut position and HARQ merging is handled;
Computing unit is used to calculate the statistical property of data from the sample survey;
Estimation unit is used for the probability distribution according to described statistical property estimation sampling data;
Acquiring unit is used for searching the emulation form that presets according to pre-set criteria and obtains zoom factor.
8. HARQ according to claim 6 merges code translator, it is characterized in that, the bit number that quantizes intercepting in the cut position processing the described first time is greater than the bit number that quantizes intercepting in the cut position processing the described second time.
9. HARQ according to claim 6 merges code translator, it is characterized in that, also comprises:
The 3rd processing module, be used for if successfully decoded, the Bit data that quantizes the described second time before the cut position abandoned, if decoding failure, the Bit data that quantizes the described second time before the cut position is kept, be used for carrying out once more HARQ and merge when handling data with the data retransmission merging; And determine that number of retransmissions has surpassed the number of times of regulation, and no matter decode results is success or failure, the Bit data that is all quantized the described second time before the cut position abandons.
10. a HARQ merges decoding system, it is characterized in that, comprises as any described HARQ of claim 6 to 9 merging code translator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100554020A CN101964690B (en) | 2009-07-22 | 2009-07-22 | HARQ merged decoding method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100554020A CN101964690B (en) | 2009-07-22 | 2009-07-22 | HARQ merged decoding method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101964690A true CN101964690A (en) | 2011-02-02 |
| CN101964690B CN101964690B (en) | 2012-07-04 |
Family
ID=43517443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100554020A Active CN101964690B (en) | 2009-07-22 | 2009-07-22 | HARQ merged decoding method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101964690B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102761509A (en) * | 2011-04-27 | 2012-10-31 | 联芯科技有限公司 | Receiving system of OFDM system and method for reducing system memory |
| CN104485933A (en) * | 2014-11-19 | 2015-04-01 | 中国科学院微电子研究所 | Digital pulse width modulation device used for switching power supply digital controller |
| CN108259400A (en) * | 2017-12-22 | 2018-07-06 | 北京松果电子有限公司 | Soft bit processing method, device and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ES2350494T3 (en) * | 2005-04-01 | 2011-01-24 | Qualcomm Incorporated | PROCEDURE AND APPLIANCES FOR CODING AND DECODING A HIGH BAND PART OF A SPEAKING SIGNAL. |
| CN101192410B (en) * | 2006-12-01 | 2010-05-19 | 华为技术有限公司 | Method and device for regulating quantization quality in decoding and encoding |
| CN101471749A (en) * | 2007-12-28 | 2009-07-01 | 三星电子株式会社 | Method for generating log-likelihood ratio for QAM-OFDM modulating signal |
-
2009
- 2009-07-22 CN CN2009100554020A patent/CN101964690B/en active Active
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102761509A (en) * | 2011-04-27 | 2012-10-31 | 联芯科技有限公司 | Receiving system of OFDM system and method for reducing system memory |
| CN102761509B (en) * | 2011-04-27 | 2016-01-06 | 联芯科技有限公司 | The receiving system of ofdm system and the method for reduction receiving system internal memory |
| CN104485933A (en) * | 2014-11-19 | 2015-04-01 | 中国科学院微电子研究所 | Digital pulse width modulation device used for switching power supply digital controller |
| CN108259400A (en) * | 2017-12-22 | 2018-07-06 | 北京松果电子有限公司 | Soft bit processing method, device and storage medium |
| CN108259400B (en) * | 2017-12-22 | 2021-05-18 | 北京小米松果电子有限公司 | Soft bit processing method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101964690B (en) | 2012-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8681819B2 (en) | Programmable multifield parser packet | |
| Jaldén et al. | Worst-and average-case complexity of LLL lattice reduction in MIMO wireless systems | |
| CN107370560A (en) | Coding and speed matching method, the device and equipment of a kind of polarization code | |
| CN103905149B (en) | Codebook design method, communication equipment and codebook training method | |
| CN1656714A (en) | Adaptive algorithm for a cholesky approximation | |
| US11652573B2 (en) | Error correction in network packets using soft information | |
| CN101964690B (en) | HARQ merged decoding method, device and system | |
| CN109582640B (en) | Sliding window-based data deduplication storage method and device and storage medium | |
| WO2019056941A1 (en) | Decoding method and device, and decoder | |
| US20160124955A1 (en) | Dual overlay query processing | |
| CN112148644B (en) | Method, apparatus and computer program product for processing input/output requests | |
| CN108259400B (en) | Soft bit processing method, device and storage medium | |
| US11106514B2 (en) | Message stream processor microbatching | |
| US11012822B2 (en) | Multi-antenna system, and transmitting apparatus and method based on index coding and beamforming thereof | |
| EP3836706A1 (en) | Transport block (tb) scheduling method and device, storage medium and electronic device | |
| US9917789B2 (en) | Computing element allocation in data receiving link | |
| CN114676629A (en) | Multi-means composite light weight processing method for modulation type recognition model | |
| KR102144732B1 (en) | A method and apparatus for fast decoding a linear code based on soft decision | |
| CN110752891B (en) | Polar code decoding method and device, storage medium and electronic device | |
| CN115514686A (en) | Flow acquisition method and device, electronic equipment and storage medium | |
| CN112187684A (en) | Soft demodulation method, device, receiver and storage medium | |
| CN101065921A (en) | TFCI decoding apparatus and method | |
| CN117574976B (en) | Large language model software and hardware collaborative quantization acceleration calculation method and system | |
| CN116909755B (en) | Access method, processor, electronic device and readable storage medium | |
| CN115037313B (en) | LDPC decoding quantization method, device and equipment, and LDPC decoding method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: DATANG TELECOM TECHNOLOGY INDUSTRY HOLDING CO., LT Free format text: FORMER OWNER: LEADCORE TECHNOLOGY CO., LTD. Effective date: 20141222 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 200233 XUHUI, SHANGHAI TO: 100191 HAIDIAN, BEIJING |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20141222 Address after: 100191 Haidian District, Xueyuan Road, No. 1, No. 40, Patentee after: Datang Telecom Technology Industry Holding Co., Ltd. Address before: 200233 Shanghai, Jiang Road, No. 41, building 4, building 333 Patentee before: Leadcore Technology Co., Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201110 Address after: 100083 Haidian District, Xueyuan Road, No. 29, Patentee after: DATANG MOBILE COMMUNICATIONS EQUIPMENT Co.,Ltd. Address before: 100191 District 1, Beijing, Haidian District, Xueyuan Road Patentee before: Datang Telecom Technology Industry Holding Co.,Ltd. |