CN104253679B - Coding/decoding method - Google Patents
Coding/decoding method Download PDFInfo
- Publication number
- CN104253679B CN104253679B CN201410424271.XA CN201410424271A CN104253679B CN 104253679 B CN104253679 B CN 104253679B CN 201410424271 A CN201410424271 A CN 201410424271A CN 104253679 B CN104253679 B CN 104253679B
- Authority
- CN
- China
- Prior art keywords
- redundancy versions
- self
- packet
- rate
- rsn
- 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.)
- Active
Links
- 230000005540 biological transmission Effects 0.000 claims abstract description 88
- 230000001702 transmitter Effects 0.000 claims abstract description 19
- 238000000034 method Methods 0.000 claims description 6
- 238000004080 punching Methods 0.000 claims description 4
- 230000014509 gene expression Effects 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims 2
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 2
- 235000005035 ginseng Nutrition 0.000 claims 2
- 235000008434 ginseng Nutrition 0.000 claims 2
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 230000011664 signaling Effects 0.000 description 36
- 230000002708 enhancing Effects 0.000 description 9
- UIIMBOGNXHQVGW-UHFFFAOYSA-M buffer Substances [Na+].OC([O-])=O UIIMBOGNXHQVGW-UHFFFAOYSA-M 0.000 description 8
- 230000004044 response Effects 0.000 description 8
- 238000004140 cleaning Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 230000000875 corresponding Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000001965 increased Effects 0.000 description 3
- 201000000194 ITM2B-related cerebral amyloid angiopathy 2 Diseases 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- LCTONWCANYUPML-UHFFFAOYSA-N pyruvic acid Chemical compound CC(=O)C(O)=O LCTONWCANYUPML-UHFFFAOYSA-N 0.000 description 2
- 230000036633 rest Effects 0.000 description 2
- 230000001360 synchronised Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 235000004418 Durio kutejensis Nutrition 0.000 description 1
- 101700081364 PRAG1 Proteins 0.000 description 1
- 241000196435 Prunus domestica subsp. insititia Species 0.000 description 1
- 230000018199 S phase Effects 0.000 description 1
- 230000003044 adaptive Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000003111 delayed Effects 0.000 description 1
- 230000001419 dependent Effects 0.000 description 1
- 238000005562 fading Methods 0.000 description 1
- 230000002452 interceptive Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006011 modification reaction Methods 0.000 description 1
- 230000000051 modifying Effects 0.000 description 1
- 230000002035 prolonged Effects 0.000 description 1
- 101700044500 ran1 Proteins 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Abstract
This is related to a kind of for the decoded coding/decoding method of the information content included at least one packet clearly, at least one packet is transferred to receiver (BS) by data link (E DCH) from transmitter (UE), wherein described information is indicated by bit sequence, the bit sequence is converted into the redundancy versions that can be transmitted and wherein a) gives information to receiver (BS) from transmitter (UE) initial transmission in the first packet for the first time, wherein information is indicated by the first redundancy versions, the redundancy versions can self-demarking code；B) confirm incorrect reception by sending confirmation from receiver (BS) to transmitter (UE)；C) one the confirmation in step b) is received, at least the information is retransmitted from transmitter (UE) to receiver (BS) in the second packet for the second time, wherein it is to indicate that the information uses the second redundancy versions, the selection of the redundancy versions is executed according to coding parameter, and whether the coding parameter describes redundancy versions can self-demarking code.
Description
The application be the applying date be September in 2005 15, application No. is 200580031120.7 (international application no PCT/
) and the divisional application of the application for a patent for invention of entitled " coding/decoding method " EP2005/054578.
Technical field
The present invention relates to a kind of method being used for transmission the information content and a kind of telecommunication apparatus for executing the method.
Background technology
Often in order to obtaining better decoding result and combining being transmitted several times for same packet.Data splitting grouping pair
In using grouping transmission communication system be well known.One example of this Transmission system is so-called E-DCH schemes
(E-DCH:Enhancement type special channel, the substantially improvement to existing UMTS uplink channels), the E-DCH schemes are just
It is being standardized as the enhancing (UMTS to UMTS system immediately：Universal Mobile Communication System).
The summary of E-DCH schemes can be in 3GPP RAN1 technical report TR 25.896v2.0.0 " Feasibility
Study for Enhanced Uplink for UTRA FDD (Release 6) ", R1-040392, February 2004,
It is found in Malaga, Spain.
This scheme is designed to utilize HARQ (HARQ：Mix ARQ, mixed automatic repeat request) scheme.In this side
In case, transmission packe, and if these groupings are not received correctly, once the negative for receiving receiver is true
Recognize, i.e. so-called " non-response (not acknowledge) " (NACK), just transmission retransmits.If reception is correct, just send out
Send positive acknowledgment, i.e. so-called " response (acknowledge) (ACK) ".In the feelings for transmitting same packet more than once
Under condition, at receiver initial transmission and retransmit all be used for packet decoding.Wherein utilize the information of two packets
" soft bit adjudicates (soft bit decision) " is used.It means that the every of the packet of transmission or re-transmission
One bit all distributes an amount, and the amount joins with the probability correlation that the bit is 1 or 0.Then in the feelings for considering two amounts
It is decoded under condition.
This provides a kind of relatively good performance, just looks like not consider the previous transmission (mixing with selection combination
ARQ it is only used only and retransmits in the case of).
In order to make above-mentioned this method correctly work, it should be ensured that the transmission received by two actually relates to identical
(higher) grouping transmitted, that is, two transmission are all obtained from the identical information content, identical (higher
) grouping (but being transmitted possibly also with the different groupings on first layer).Layer refers to OSI (open system interconnections at this
(Open System Interconnection)) model.By the transmission characteristic considered in first layer, a certain application
Information expression, the application itself considered in higher may be different for transmission and re-transmission.
There are several approach to ensure this point：One is so-called sync retransmission agreements.In this agreement, if being more than one
Secondary re-transmission is foreseen, then sending weight after in initial transmission or also retransmitting previous at regular intervals
It passes.In this way, receiver knows that it is expected to retransmit given grouping at what time.
However, receiver does not know whether two transmission in compatible (compatible) time actually relate to still
Same grouping, or whether had begun the transmission of new grouping.If receiver cannot receive it is all grouping but for example
It is missed due to interference, then this especially such case.If transmission utilizes soft handover (SHO), that means that more
In one receiver attempt receive these grouping, then it is possible that be a receiver fail receive grouping, but another
Receiver can receive this grouping and response this grouping.In this case, new grouping can be transmitted next time.
The receiver for being not received by the first grouping do not know by the response (acknowledgement) transmitted by another receiver, because
This now has to recognize that this new grouping cannot be with previously received any packet assembling in some way.
About E-DCH, if the terminal of data connection is established with first base station close to the second base station, then just sending out
Raw soft handover.In order to ensure when from a cell by first BTS management to the second community by second BTS management
Smooth or soft transition exist to the connection of two base stations in the stage of transition.
In order to easily identify, it is also possible to introduce so-called re-transmission sequence number (Retransmission Sequence
Number) (RSN) or retransmission counter：
If new grouping is transmitted, which is reset (be, for example, 0), and the counter with weighing each time
It passes and is incremented by.If receiver compares RSN differences and time difference (known time between considering sync retransmission agreement and retransmitting is poor),
Then if the time difference matches with RSN differences, receiver can combine these receptions, if these differences mismatch, not combine this
It is a little to receive.
According to the prior art, problem is, RSN values range is at least big as the maximum times of possible re-transmission：It is typical
Ground, if cannot leave this grouping behind with maximum number of retransmissions transmission packe and transmitted next grouping.It is possible heavy
The maximum times of biography may be quite big.This this may result in excessive signaling information amount, because RSN must be passed with each grouping
Defeated and re-transmission is transmitted together.
Invention content
Based on the prior art, therefore it is an object of the present invention to provide the possibilities for the transmission for improving packet, wherein
The re-transmission of packet is foreseen.
Another task of the present invention is to provide a kind of improved repeating method.This target can pass through institute in the present invention
Disclosure is realized.It is advantageously improved, describes in a preferred embodiment of the invention.
According to the present invention, receiver is grouped into from transmitter transmission data.Include to pass through bit sequence in the packet
Arrange the information indicated.In order to transmit, this bit sequence is converted into redundancy versions (redundancy version), this redundancy version
Originally one of more possible expressions of transmission are suitable for.In the case where packet is not correctly received in receiver, this connects
It receives device and sends and confirm that (such as NACK) arrives transmitter.As soon as receiving, transmitter retransmits the information in the second packet, institute
Stating the second packet may be with first difference.For the re-transmission, using the second redundancy versions, which may be with
One difference.It is based on coding parameter for the second transmission selection redundancy versions, whether the coding parameter instruction redundancy versions
Can self-demarking code, also this means that only own ability decodable code.
The advantages of the selection is, the selections of redundancy versions can consider transmission characteristic, whether such as receiver has can
The case where being used for the previous version of decoded packet gets off completion.Therefore preferable decoding result can be obtained.Meanwhile
Seldom additional effort is not paid or only paid for additional signaling.
Other aspects and standard to the selection of redundancy versions are described in detail in the description of embodiment and attached drawing.
The present invention also provides " re-transmission sequence number (RSN) " enhancing of scheme, this allows the value range of reduction RSN and still permits
It is allowable it come determine combination which grouping, at least for majority of case, especially the number of transmission be no more than RSN maximum
For those of value situation.It can be designed as according to previous description according to the first and/or second number of retransmissions of this application
" re-transmission sequence number ".
May there are some cases, wherein receiver that can no longer conclude whether packet must be combined now now, but
It is these situations rareness enough and will not seriously damages performance.It is more valuable, it reduces and required bit is encoded to RSN
Number, for example it is reduced to dibit from three bits, this causes to transmit " expense (overhead) " information institute together in company with packet
The resource needed greatly reduces.
Description of the drawings
Next other advantages and embodiment utilizes attached drawing to be described in detail, wherein：
Fig. 1 shows the connection between terminal and base station.
Specific implementation mode
The terminal U with the connection E-DCH to base station BS is shown in FIG. 1.
This terminal can be any communication equipment, such as mobile phone, PDA (personal digital assistant), mobile computer etc.
Deng.
Base station can be any central location in communication system CS.This communication system can for example pass through mobile radio communication
(such as UMTS (Universal Mobile Communication System)), LAN (such as WLAN (WLAN)) or broadcast system indicate.
Connection E-DCH can be any data link between terminal UE and base station BS, and wherein transmission data is grouped.Except increasing
Except strong type up-link, the connection can be HS-DSCH (high-speed downlink shared channel (High-
Speed Downlink Shared Channel)), either ... or such as broadcast other any data link of connection etc.
Terminal UE forwards packets to base station BS.With ACK, either NACK connects come response is correct or wrong for base station
It receives, the ACK or NACK are sent back to terminal UE.
In case of a nack, terminal UE retransmitted data packet.As showing above, packet is from the information content
Be identical, but be available with different codings, for example, the bit being encoded different subsets.
In following description, number of retransmissions according to the present invention is abbreviated as RSN.It is general using RSN as follows according to the present invention
It reads：If transmitting new grouping (namely in first transmission of new grouping), RSN is reset to 0.
If the re-transmission of grouping is transmitted, and RSN is still under its maximum value, then this RSN is incremented by.If grouping
Re-transmission be transmitted, and RSN has had reached its maximum value, then with previous suggestion on the contrary, RSN is not incremented by but protects
It holds in maximum value.Such advantage is that the required space of signaling can be limited to value set in advance.
The maximum value of RSN is named as RSNMAX, and to make further discussion, (RSNMAX can be the largest re-transmission time
Number).
If such as one of following condition 1 or 2 set up, receiver can by received grouping with previously
Received packet assembling.For this purpose, the grouping (upper one grouping received) that receiver had also previously been received consideration
RSN and since receive that grouping (such as according to the present invention, the time difference may be by multiple re-transmission times time difference
The formed delay of step).It is to be noted that, it is assumed that not new grouping is transmitted and RSN is limited to RSNMAX and sends out not yet
It is raw, the time difference with from described upper one be grouped since the incremental quantity of RSN it is identical.
If：
1) RSN is less than RSNMAX and the difference of RSN is identical as the difference of transmission time (both excessive without RSN in this case
Go out and without sending new initial packet) or
2) RSN is equal to RSNMAX, has received upper one grouping received the time difference of RSNMAX to be at most, and upper one
The RSN of grouping is at least RSNMAX plus the time difference, then the combination of packet is possible.It in this case may be
There are the spillings of RSN, but the received grouping corresponds to the grouping received by upper one, and otherwise RSN is described upper one
It will be reset to 0 after grouping, and be not increased to RSNMAX, because the time difference is too small since then.
Such as grouping cannot be combined in following situation 3 or 4：
3) RSN is less than RSNMAX and the difference of RSN (is overflowed without RSN in this case not equal to the difference of transmission time
And new initial packet is had sent, therefore data related with old grouping can be safely deleted) or
4) RSN is equal to RSNMAX and to have received the grouping received by upper one more than the time difference of RSNMAX.At this
Receiver cannot determine whether to combine grouping current and received by upper one in the case of kind, that is, not can determine that it is only
The re-transmission for only having sent this grouping has still had sent the so many re-transmission of new grouping and this new grouping so that
RSN has reached RSNMAX.Therefore, in order to avoid the data of different groupings may be obscured, it is necessary to which deletion has with old grouping
The information of pass.
In the case where RSN is equal to RSNMAX, it becomes virtually impossible to be, to receive upper one less than the time difference of RSNMAX
The RSN of the grouping and a upper grouping that are received is less than RSNMAX plus this time difference.Therefore, strictly speaking, it is not necessary to count
The RSN for counting a grouping in adds the time difference, and by it compared with RSNMAX.However this be easy to do and in this case
It may detect that some rare error conditions correctly not decoded to RSN.This is unlikely that because RSN no matter
How must be suitably encoded/protect, but be easy to because calculating, done so and use the additional consistency
Verification is advantageous.
In order to which the present invention is explained further, some more detailed examples will be provided with reference to following table 0：
Table 0 is shown：
Time T is (by re-transmission time as unit of transmission time, that is, in sync retransmission agreement in the first row
Difference).Time T can be indicated with the time slot of fixed quantity or so-called TTI (Time Transmission Interval).
The RSN being actually transmitted in a second row.
Received RSN, is represented as RX in the third line, if being not received by RSN (such as due to dry
Disturb), then the reception lost is indicated with dash (-).
In addition, " f " shows that buffer has been cleared by, that is, receiver is it is assumed that the grouping that actual received arrives cannot
It is combined with previous one.
In fourth line, for illustrative purposes, another possible sequence of transmitted RSN is provided so that indicate whether can
Identical RSN can have been received but had sent new grouping.According to this example, at the time 0, RSN be set as 0 and
Grouping is transmitted.At the time 1, new grouping is transmitted and RSN is reset to 0 again.However, receiver is not received by this
RSN(-)。
At the time 2, the re-transmission for the grouping transmitted at the time 1 is transmitted.Receiver is recognized should not will be described
Grouping and previously received grouping (namely the grouping transmitted at the time 0) are combined, because actual RSN is less than
The difference that RSNMAX and actual RSN subtract the RSN (its at the time 0 be 0) received by one is 1-0=1, and equal to 2
The time difference of (time 2 subtracts the time 0) mismatches.According to circumstances 3 is as described above for this.It is to be noted that：It is come from if also had sent
The re-transmission of the grouping of time 0, then the RSN will be incremented to 2 till now, as shown in Hyp rows.Therefore receiver can
To exclude such case, which means that the receiver does not execute combination.
At the time 3, re-transmission is transmitted, but is not received by.
At the time 4, re-transmission is transmitted, and RSN increases and reaches RSNMAX.Receiver detects this point, from a upper institute
Time difference since the RSN received is 2, and the RSN (1) received by upper one obtains 3 plus the time difference, this is at least
RSNMAX.Therefore, according to circumstances 2, the grouping and the upper packet assembling at the time 2.
In the time 5 to 6, re-transmission is sent, and RSN rests on RSNMAX.
It retransmits and is sent at the time 7, RSN remains in RSNMAX.According to rule 2：RSN is RSNMAX, and the time difference is
3, namely be at most RSNMAX and previous RSN plus the time difference is 6, namely at least RSNMAX, receiver is by the grouping
With packet assembling previously received at the time 4.If having had sent new grouping since the time 4, RSN is
It through being reset, and may at most increase to 2 from that time, such as be shown in Hyp rows.
In the time 8 to 10, re-transmission is sent, and RSN rests on RSNMAX.
At the time 11, re-transmission is sent, and RSN remains in RSNMAX.According to rule 4：RSN is RSNMAX, time difference
It is 11-7=4, namely is more than RSNMAX, receiver is not by the grouping and packet assembling previously received at the time 7.Newly
Grouping may be sent at the time 8, RSN has been reset and may have been added to 3 from that time, such as exists
Hyp rows are shown.Therefore, receiver does not know whether the grouping received at the time 11 can be with point at the time 7 really
Group combination, it is therefore necessary to make one it is conservative it is assumed that namely not data splitting.This is by the word in the time 11 in the 3rd row
Female f is shown.
At the time 12, new grouping is sent and is received, and RSN is reset to 0.
At the time 13, re-transmission is sent but is not received, and RSN increases to 1.
At the time 14, further retransmit is sent and is received, and RSN increases to 2.According to rule 1：RSN 2
Still less than the poor 2-0=2 of RSNMAX and RSN be equal to transmission time poor 14-12=2, receiver by this grouping with when
Between packet assembling previously received at 12.Hyp rows show new if having had sent (but missing) at the time 13
Grouping, then RSN will not be 2 but 1.
Embodiment is detailed till now, wherein if same packet has been retransmitted, the embodiment
It has been applicatory.
Further, it is possible that in HARQ system for identical content and different groupings (this is so-called " chases after
Catch up with merging (chase combining) ") but the grouping that for example encodes in different ways sent, such as known as different superfluous
Remaining version (also referred to as " steadily increase redundancy ").
In this case, the instruction being transmitted together to which redundancy versions and given grouping can also be used.This
Overhead can extraly be increased.Therefore, it can implicitly be held together with the time for sending grouping together or implicitly with RSN
The signaling of row redundancy versions.Transmitter and receiver (or can be equivalent to transmission time from either RSN or transmission time
It is transmission frame number, connection frame number (also referred to as CFN), the latter will be used for the hereinafter intention without limiting other aspects) it falls into a trap
Calculate applied redundancy versions (RV).
For the improved RSN as being provided in the present invention, the determination of RV can be carried out as follows：For being less than
The RSN values of RSNMAX, can calculate RV from RSN.However, when RSN reaches RSNMAX, for all remaining heavy of grouping
RSN is maintained at this value for biography.Therefore in this case, not relatively good from RV is calculated in RSN but from CFN.In short,
According to this aspect of the invention, if RSN<RSNMAX, if then calculate RV and RSN=RSNMAX from RSN, from
RV is calculated in CFN.
In another embodiment of the invention, especially for RSN<RSNMAX, can be before being transmitted by RV to RSN's
Relationship is signaled to transmitter from receiver.
In another embodiment, it is understood that there may be the predefined default association between transmitter and receiver, the default pass
Connection can be associated with by signal and be substituted.Such as it can be signaled for some values of RSN, particularly with the most frequently used value
The association, and default value is then used for other values, the value being for example less commonly used.Alternatively, if system can make
With different grouping dimension or encoding scheme (this is particularly suitable for AMC (adaptive modulation and coding) scheme), then default
Value can be used for certain size/schemes, and the RV values of explicit signaling can be used for others.
In addition, in two class coded-bits, i.e. systematic bits (systematic bit) and check bit (parity
Bit the coding method distinguished between).Substantially systematic bits correspond to information to be transmitted, and check bit provides redundancy letter
Breath.In these cases, it may be possible to emphasize systematic bits advantageously for initial transmission but be emphasized at least some re-transmissions
Check bit.In this case, it is possible to not signal the fact to emphasize systematic bits in initial transmission.This can be with
Some transmission bandwidths are saved to the relationship of RSN for signalling RV.In addition, in some instances it may even be possible to be enough to use initial transmission always
Predefined RV and for (some) retransmit be provided solely for explicit signalings of the RV to RSN relationships.In addition, for RSN=RSNMAX
The case where, emphasize that systematic bits may also be also advantageous.This is particularly suitable for low encoding rate (such as has weight in the grouping transmitted
The redundancy wanted is available).Then for RSN=RSNMAX the case where, it is possible to also not having to explicitly signalling to emphasize systematic bits.
Following Examples are related to the re-transmission of redundancy versions.One important aspect is how to carry out the selection of redundancy versions, because
The decoded validity of combined type of packet is often influenced when being initially transmitted together with re-transmission for this, the re-transmission is at this
It is redundancy versions in the case of kind.
Redundancy versions are accurately selected to be described in detail in next part dependent on RSN.For the technology of this field
For personnel obviously, these aspects can be combined advantageously with the either side of previous the disclosed embodiments.Equally
Ground, these aspects can also be applied independently.
It has been proposed that selecting different redundancy versions based on so-called RV indexes (redundancy versions index).RV indexes are basic
On define to defining all parameters needed for specific redundancy versions.This index allows easily to refer to specific redundancy
Version, because only a single parameter, i.e. RV indexes need to be designated.This is specified and is especially transmitted from terminal UE or " uses signal
Notice " gives base station BS.About E-DCH it is diagrammatic in terms of (such as generality Signaling) in Rl-0414083GPP TR
25.808VO.2.3 (2004-10) technical report third generation partner program；FDD enhanced uplinks；In terms of physical layer
In be discussed.
One of relationship between RV indexes and the parameter obtained is proposed in Rl-041354, Editor (Siemens),
" in its chapter in CR 25.212-Introduction of E-DCH " (November 2004, Shin Yokohama, Japan)
It is provided in section " 4.8.4.3HARQ Rate Matching Stage ".
Especially parameter s and r is relevant to the relationship of E-DC-RV indexes：
The parameter of Rate Matching stage depends on the value of RV parameters s and r.S in view of E-DCH corresponding to each RV and
R combinations are listed in table 15.First transmission attempt of transmission block should use the RV (RV with s=1) for emphasizing systematic bits.
Higher level signaling be used to control the redundancy versions number that UE should be used.
Table 1：The RV of E-DCH
| E-DCH RV indexes | S | r |
| 0 | 1 | 0 |
| 1 | 0 | 0 |
| 2 | 1 | 1 |
| 3 | 0 | 1 |
Use following parameters：
-Nsys=Np1=Np2=NE, j/3
-Ndata=NE, data, j
-rmax=2
How using these parameters rate matching pattern used in reality (rate-matching pattern) is determined
Exact method in R1-041354, Editor (Siemens), " CR 25.212-Introduction of E-DCH "
It is described in chapters and sections cited in the upper surface of (November 2004, Shin Yokohama, Japan).This method can also
Applied to various embodiments provided herein.
Particularly, next two chapters and sections of this document show how to obtain rate matching pattern.First reference, the
4.2.7.5 chapter " Rate matching Pattern determination (rate matching pattern determination) " is related to being based on parameter
eplus, eminusAnd einiRate matching pattern accurate determination.E in these parametersiniCurrent or actual and expectation is described
Rate of punchinging (puncturing ratio) between initial error, eplusDescription changes the step-length (stepwidth) of the error,
eminusAnother step-length used in speed matching algorithm is described.4.5.4.3 chapters and sections " (HARQ Second Rate
Matching Stage (the second Rate Matching stages of HARQ)) " how description is from input parameter Nsys NdataAnd rmaxIt is middle to calculate this
A little parameters, wherein NsysIndicate systematic bits number, the systematic parameter is to compare carrying information with the check bit for coding
Bit, NdataIndicate the sum of bit, namely check bit and systematic bits, and rmaxIndicate that r adds 1 maximum value.
Come illustrate how these parameters are arranged in the 4.5.4.3 chapters of file since then, in particular for the algorithm in 4.2.7.5 chapters
eplus, eminusAnd eini：
Whether parameter s specifies RV can self-demarking code, it means that is only decodable when considering the RV.Decodable code is in this side
Face means to find out the information content indicated by redundancy versions, packet.If s=1, when in rate-matched
When period punchinges, check bit of the so-called systematic bits prior to turbo codes.Such redundancy versions can typically self solve
Code, it means that can be decoded by itself, be disturbed very much except non-reception certainly.This (was verified when the case where s=0
Bit priority), it may occur in this case or even RV can not be decoded by itself in the case of no noise, and be
It is decoded with together with another RV.Therefore, grouping first time transmission should always can with self-demarking code, namely utilize s=1.
It may occur during soft handover (SHO)：One base station receives grouping until sometime from mobile station, and
Another base station receives grouping after the time.If becoming to compare first base station to the instantaneous path loss of the second base station
Path loss it is good, then this occur, this may be easy to happen due to fast-fading.Path loss is via such as terminal and base
The decaying for the signal that a certain connection between standing is transmitted.Signal uses certain route, i.e. path.Due to reflecting, interfering
Deng signal is easy to decay.
If becoming more preferable to the path loss of the second base station, if that all redundancy versions be can self-demarking code when,
It is advantageous.Otherwise such conversion (switch) afterwards the second base station may not be able to immediately to these packet decodings because
Second base station be likely not to have received packet can self-demarking code initial transmission and only alone it is non-can self-demarking code it is later
Transmission.
The present invention also includes in 4.2.7.5 and 4.5.4.3 chapters and sections described above.
Inventors have realised that surprisingly this means that being unnecessary, if connection or data transmission are worked as
It is preceding to be in SHO, then should be that each RV selects s=1.The reason is that application is punchinged if as discussed above, then s=0 is only
Only create it is non-can self-demarking code RV.Leave out each bit by punchinging and be described.It punchinges to reduce the sum of bit, such as
Sum is set to adapt to the fixed capacity of transmission.Punchinging for bit is carried out in this wise so that not or only information as few as possible
It loses.Therefore, generally it is often that check bit is punctured.
If executing repetition now, all coded-bits will be transmitted anyway, or even repeat therein one
A bit.Therefore s=0 can safely be selected.S=0 only selects that another redundancy versions different from s=1 in this case.
Using different RV enhancing performances (so-called steadily increase redundancy (incremental redundancy) (IR)) and
It is better that more how different RV can be selected.Therefore the general rule for using s=1 in soft-handoff as provided in the reference
It is not actually advantageous in this case, and the case where repeating should be directed to and changed.
Therefore according to advantageous embodiment, the selection of redundancy versions is not merely based on RSN but also is based on being that will punching also
It is the fact that will be recycled and reused for rate-matched.
Additionally or alternatively considered another standard is encoding rate.Encoding rate is that bit number before the coding removes
With bit number after coding and after rate matching.Rate-matched is to obtain institute's phase at a certain time interval
The final data number of prestige either correspondingly desired data rate and punching to bit or repeat.
Typically, for E-DCH, payload data (payload data) is compiled using so-called " turbo codes "
Code.It is compareed with the data for signaling etc., payload data is the data of actual bearer information.These turbo codes have
(about) 1/3 encoding rate is also that each bit to be encoded generates a 3 coded-bits, i.e. systematic bits and two
Check bit.If encoding rate (after rate matching) is less than 1/3, using repetition during rate-matched, if compiled
Code check is more than 1/3, then using punchinging.Accordingly, with respect to using which RV and be especially whether to use the decision of s=0 can have
It is based on (coding) rate sharply.
About this point, it should be noted that not the encoding rate of the turbo codes of rate-matched be not just 1/3 but
It is slightly smaller, because extraly so-called " terminating bit (termination bit) " is attached to the end of the data of coding
End.However, for the purpose of described embodiment in the context of that application, the difference is small enough to be incoherent.
It therefore, may the either calculation code rate or not in the case where considering to terminate bit in the embodiment of detailed description.Normally result
It is equal.
According to other advantageous embodiments, the lower column selection based on RSN and encoding rate to RV is realized：
Table 2：Relationship between RSN values and E-DCH RV indexes
In table 2, show that the redundancy versions of E-DCH index for different encoding rates and different re-transmission sequence number RSN.
From RSN values, used encoding rate and if RSN=3 also from TTIN (TTI (Time Transmission Interval (time
Transmission interval)) number) calculate E-DCH RV indexes.For UMTS, 10ms is selected if it is TTI, then TTI
Number it is equal to Connection Frame Number CFN, for the TTI of 2ms, defines TTIN=5*CFN+ subframe numbers, wherein subframe numbers count 5 TTI, institute
TTI is stated in given CFN, from 0 for the first TTI to 4 for last TTI.In other words, for UMTS, subframe
Length is 5 TTI.
NARQ is to mix ARQ into number of passes, refers to the how much transmission for executing individual packet parallel.Such as there are data
It is grouped the transmission and re-transmission of A.The transmission and re-transmission of packet B or other packets are had begun simultaneously.If CFN
Including more than one TTI so using TTIN rather than CFN is necessary, to distinguish the TTI within a CFN.This takes
Certainly setting up in individual data connection and system, wherein establishing data connection in the system.Therefore, these conditions can be with
It is generalized other numbers to contain the TTI in frame.Division (division) by NARQ numbers, ARQ into number of passes is necessary
Because previous transmit after will not be ranked immediately re-transmission but only receive corresponding response (ACK or
NACK after).As above explained, different HARQ process transmission packets are utilized therebetween.
As when comparing last row and secondary series of table x above it is noted that E-DCH RV indexes are always chosen
It is fixed so as to use s=1 (namely E-DCH RV indexes are 0 or 2, compare table 2 above) in secondary series.It is less than 1/ in rate
3 namely for repeat in the case of used in last row in, also use s=0 (E-DCH RV indexes 1 and/or 3).
Therefore, according to another embodiment, using for 1/3<Rate<1/2 s=1, just as shown in Table 2.According to
Another embodiment does not use s=0, even if data link or data transmission are not in SHO.In this way, must not
Establish to whether use SHO public understanding, this for example pushes signalling due to following：
Because the beginning of SHO is since the delay of relevant signaling is always somewhat delayed by, and because the signaling is prolonged
It is unnecessary equal for from RNC (radio network controller) to the signaling of base station and terminal or mobile station late, so
The perfect synchronization of public understanding is troublesome, wherein typically switching since RNC.It avoids determining to work as by above embodiment
Preceding necessity for whether executing SHO.
Only select two RV unfavorable unlike as considering in the past in this case, because for these rates two
A RV is enough to transmit：By retransmitting twice, all bits caused by turbo codes can be transmitted：For rate 1/2,
After rate-matched, all systematic bits and half check bit are transmitted in a RV, in another RV, all systems
Bit and the other half check bit are transmitted again.Therefore, it is possible to transmit even more check bits it is non-can self-demarking code RV
Use be unnecessary for this encoding rate, but only for higher encoding rate necessity.Inventors have realised that
The method of the prior art is unfavorable in terms of systematic bits, because usually systematic bits should be more emphasised than check bit,
Also in this way when multiple TV are used together.
Therefore, only for the encoding rate higher than 1/2, it is proposed that use s=0, namely can self-demarking code for the non-of re-transmission
The RV of (non self-decodable).Particularly, if 1/2≤rate<5/6, then suggest using one it is non-can self-demarking code
RV, see the middle column in table 9.In encoding rate 2/3, can self-demarking code the first RV include all systematic bits and 1/4 school
Test bit, it is non-can the 2nd RV of self-demarking code include purely check bit namely 3/4 check bit.Therefore, until rate
(2/3), it is possible to transmit all bits with two redundancy versions.In higher rate, it is necessary to using more than one non-
Can self-demarking code RV：If encoding rate is 1, the Transmission system bit in initial transmission, first and second retransmit include respectively
50% check bit.Therefore for rate 2/3, one it is non-can the RV of self-demarking code be ideal, and for rate 1, two
It is non-can the RV of self-demarking code be ideal.Because 5/6 is 2/3 and 1 arithmetic mean of instantaneous value, it is recommended that for one and two it is non-can
Transition between the transmission of self-demarking code uses this value.Obviously others thresholdings are also possible, down toward 2/3 because 2/3 be can
With with one can self-demarking code and one it is non-can the transmission of self-demarking code transmit the maximum rate of all check bits.Similarly,
The thresholding of rate 1/2 can also be arranged slightly higher, such as the arithmetic mean of instantaneous value between 1/2 and 2/3, be 7/12 or
Any value between person 1/2 and 7/12.In addition undoubtedly when UE (user equipment (User Equipment), the movement in soft handover
The synonym of platform) given maximum data rate should be only used.The reason is that the mobile station in soft handover can be in two cells
In cause to interfere, so limiting their data rate advantageously influences the situation in two cells.In this maximum data
A certain encoding rate is applicable in when rate.Therefore in an embodiment of the present invention, transition point can be selected namely also self solved using non-
The minimum code rate of the transmission of code corresponds to the permissible maximum data rate in SHO.
In row on the right, for the encoding rate higher than 5/6, as illustrated above, it is proposed that two it is non-can self-demarking code
RV is shown in Table x.
As described above, RV having to explicitly to be distributed to three initial transmission, wherein RSN<RSNMAX.If RSN reaches
RSNMAX is 3 in the RSNMAX example given above, then being based on TTIN computing redundancy versions.This is with following excellent
Point, i.e., to continuously transmitting using different RV, even if always using identical RSN (namely RSNMAX).Only in RSN
The first transmission in the case of=RSNMAX, selects best RV to be advantageous, however then will must after RSN reaches RSNMAX
This RV must be always used, this is clearly undesirable.For different rates, need to take into account when that RV is selected in selection
The formula selected, this is described in detail below：
If 1/3<Rate<1/2, then be only used only can self-demarking code RV, therefore for RSN>RSNMAX also should be only
Only select these can self-demarking code RV.This passes through in table x aboveTo realize.
If 1/2≤rate<5/6 (or usually in middle column), need one it is non-can self-demarking code transmission.Still it answers
This keeps systematic bits preferential, thus using two can self-demarking code RV.For RSN=1, E-DCH RV indexes 3 are selected, and for
RSN=3 cyclically selects RV for 0,1,2 by formula.A certain other gains can be reached in this way, because for
RSN=1 and RSN=3 using it is different it is non-can self-demarking code RV, this causes a certain additional IR gains.This passes through above
In table xIt realizes.
If rate >=5/6 (or 2/3 as described above), for comprising all check bits, two it is non-can self-demarking code
RV be necessary.In order to also emphasize systematic bits in this case, it is proposed that for RSN=3 use 4 RV, two it is non-can
RV and two (not only minimum value be 1) of self-demarking code can self-demarking code RV.This passes throughCome real
It is existing.
At thisExpression takes maximum integer downwards, which is not more than x, is also called floor (x) or int sometimes
(x)。
In following embodiment, using another different enhancing to RSN schemes, the enhancing again can or with
The mode or used independently of their mode that other enhancings provided in the present invention combine.
In Rl-041339, Panasonic, Downlink signaling related issues for Enhanced
It suggested in Uplink, November 2004, Shin Yokohama, Japan, if cannot be decoded to E-DCCH data,
Terminal UE should retransmit initial redundancy versions.E-DCCH is carried and the relevant Overhead of payload data, such as is made
Transformat, namely payload bit number, and for being necessary to payload data decoding.So if
Base station cannot decode E-DCCH, then the base station will cannot utilize the data transmitted on E-DCDCH.For example, if to E-
DCCH data add CRC (cyclic redundancy check (cyclic redundancy check)), then base station can determine E-DCCH not
Correctly it is detected.If base station cannot utilize the first RV, in Rl-041339, Panasonic, Dwonlink
Signaling related issues for Enhanced Upl ink, November 2004, Shin Yokohama,
Suggested in Japan again this retransmit the first RV, so as to determine for the first actually available RV of base station be can self-demarking code biography
It is defeated.Multiple proposals are provided in this document：How this is signaled to mobile station, for example, by introduction except ACK and
Third situation except NACK.In the context of the embodiment described, the accurate method of signalling is not critical, therefore is appointed
Where the signaling used in HARQ processes can be applied.
In the frame of this application, which is referred to as CNAK (control NACK).However, this file do not disclose about
How any information of first RV should be reselected in conjunction with RSN schemes.
In this document, especially chapters and sections " Node B reception scenarios (Node-B receiver scene) " are
Relevant, especially following reference, the reference is since the third section of this chapters and sections：
Following table shows 3 kinds of scenes, when this is likely to occur in the initial transmission of UE transmission datas grouping.In table
Secondary series describe the reception state for each scene at node B, third row show preferred UE in this case
Behavior.
In the case where E-DPCCH cannot be correctly decoded, as listed by the first scene in this table, node B cannot locate
Manage data received on the e-dpdch.Received data can not possibly be decoded and therefore node B abandons the number
According to.In our view, in this case again transmission can the RV of self-demarking code be advantageous.When UE is sent out according to the RV sequences of regulation
When sending re-transmission, only transmit it is next can be possible to data packet decoding after the RV of self-demarking code, this may result in increase
Delay.Therefore would rather retransmit in such a scenario can self-demarking code RV (initial transmission).
Scene 2 describes such case：Node B correctly decodes E-DPCCH, but CRC failures on the e-dpdch.
Here when combining energy received on the e-dpdch with other re-transmissions, node B can use described in E-
The upper received energy of DPDCH.Therefore UE should be retransmitted according to the RV sequences of regulation in this case.It is recommended that UE behavior roots
Whether had occurred and that according to the first or second scene and different.
Substantially there are two types of possibilities：A kind of possibility is that the first RV is not only assigned to RSN=0 in this case
It is also allocated to RSN=1, to which the RV of next is also moved into corresponding next RSN values.Although this approach is feasible,
There are better approach used in embodiment：
If mobile station receive CNAK, again utilize RSN=0 and accordingly with first can self-demarking code RV send
Next transmission.At first sight this may mean that RSN agreements described above cannot be used for such case again.However, it can be with
It is enhanced to equally support such case：Base station and mobile station all know this situation by CNAK signalings.The two therefore can be with
The fact is considered for RSN agreements.In addition, when it cannot utilize the first transmission, base station can determine to clean anyway
(flush) soft buffer, this is consistent with the processing of RSN agreements.If this occurs in soft-handoff, other base stations allusion quotation
Do not know that first base station has transmitted CNAK to type.However, due to RSN agreements, the others base station is able to detect that newly
RSN=0 sent and cleaned their soft buffer.Although cleaning buffer is not strictly necessary in this case
, but its effect be other base stations learn that RV sequences are reset the fact that, and will therefore use correct RV.Otherwise,
The others base station will use the RV of mistake, and to which they do not have any chance to packet decoding, this will be resource
Waste.Even if one in other base stations in soft-handoff, which can be the first transmission, has been correctly detected grouping, RSN associations
View still works well with：Then, while first base station transmits CNAK, which sends ACK.Then mobile station will shine
Often next grouping is transmitted using RSN=0.First base station will be unaware that the second base station has had been acknowledged this thing of the first grouping
It is real, and therefore do not know that new grouping is transmitted.However.Because just cleaning buffer from first grouping anyway,
No matter present first or it is next grouping be transmitted it is all unimportant.
Even if CNAK mistranslations are mistakenly ACK by mobile station or terminal UE, base station remains able to be correctly received next
Grouping.First packet loss in this case, but be not as the enhancing of RSN agreements and merely because the garbled-reception of ACK.
When NAK mistranslated for ACK when, this error condition has been possible in the case of no RSN protocol enhancements, namely enhancing
Type RSN agreements do not lead to any further decaying.
The enhancing to RSN has been provided in the present invention.It will be noted that there is also by other devices or slightly not
Same signaling provides other possibilities of similar function.Propose to be NDI (new data indicator (New Data as one
Indicator)).NDI is identical for the re-transmission of grouping for the incremental mould that each new grouping is to maximum value.
NDI schemes may be more steady in the case where there is many retransmit, because NDI is then only seldom incremented by and exists
(such as because backrush (wrap around)) becomes no longer specific less risk.
According to another aspect of the present invention, RSN NDI schemes can be used according to the characteristic of connection.This means that very
To the RSN and NDI can be used according to characteristic sometimes sometimes in single connection.Such characteristic can include connection whether
The fact that in soft-handoff either used grouping dimension or encoding rate.
In a preferred embodiment, in the case of no additional explicit signaling, can by transmitter and receiver Lai
Determine selected characteristic.Additional signaling overheads is not introduced in this case.
It is obvious for those skilled in the art, it can be by combining the aspect that is provided in this application and another
A or other known flow combines the present invention.Therefore, it is considered also fitting by the present invention described in above example
For those situations.Particularly, these examples and description are not construed as limiting the scope of the invention.
Next other examples, explanation, embodiment and the modification of the present invention are provided：
1. introducing
It achieves in upper primary meeting and improves substantially about what the HARQ to E-DCH was defined.Important agreement is that have
Synchronous protocol with sync retransmission and with IR and pursue combined HARQ.In addition, agreeing to when operating IR suitable with what is given
Sequence always emphasizes systematic bits (s=1) using redundancy versions and the first transmission.For the re-transmission in non-SHO, emphasize and not
Emphasize that systematic bits should all be possible (s=0 or s=1).
Another agreement for SHO is always to emphasize that the utilization of the transmission of systematic bits (s=1) is typically considered advantageous
's.
It is true and some E-TFC can be contacted RV and CFN and for other E-TFC having to explicitly signalings from these
The agreement of change starts, this document is that the signalling of HARQ relevant informations suggested solution.
2.HARQ related signaling informations
HARQ related signaling informations are transmitted using each E-DCH grouping transmission together with E-TFC information.Its function is
Notify node B about the used redundancy versions (X for needed for solution rate-matched (de-rate matching)rvValue), again
Trigger the cleaning to the soft buffers of node B.
The sequence of redundancy versions and application redundancy versions should be by network for example by higher level signaling control or can be with
It is specified for each TFC using fixed matching rule.The rule is all known for UE and network and can be " with hardwired fashion
(hard-wired) " it is performed.In our view, for each TFC, three redundancy versions are designated, and (one is used for initially
Transmission, one retransmitted for first and one for the second transmission) or from higher level signaling.
Next it is required due to the difference of the reliability indicated about new data and RV is selected in SHO and non-SHO
The different link effect gains selected, separately discuss two kinds of situations.
Signalling in the case of non-SHO
In the case of non-SHO, it is proposed that dibit RSN, as in Rl-040958, Ericsson, " E-DCH
Described in Physical Layer Hybrid ARQ Processing ", Prague, Czech Republic, August 2004
As, it is disclosed that comprising RSN and obtaining redundancy versions based on RSN in uplink signalling.Described is believing
It has also been described in the context of the introduction of RSN above comprising RSN in orderization.
Therefore, the redundancy versions applied are signaled to node B from UE.Additionally, RSN provides new data instruction
Function (in the case of RSN=0, indicating that soft buffer is to be cleaned).RSN is incremented by with each re-transmission and with each
Initial transmission and be set to 0.In the case where the number of re-transmission is more than 2, for all other re-transmission, which is set to 3.This
The advantages of sample is can be less than log for the bit number of RSN2(Nmax) without losing IR gains, because of the number allusion quotation of different RV
It is less than maximum the number of transmissions N to typemax.For RSN=3, RV selects to be to obtain further gain based on CFN.
Signalling in the case of SHO
As in Rl-040906, Motorola, " Synchronous HARQ and reliable signaling
During SHO (Enhanced Uplink) ", Prague, described in Czech Republic, August 2004,
In the case of SHO, specific TFC is selected by UE.Usual SHO TFC provide lower rate.For lower rate, learning
It has been shown that it is about 0.3dB to be likened to the combined IR link effect gains of pursuit during the clause stage.However, in difference
IR schemes between and also the performance difference between the IR schemes with different RV sequences be it is negligiblely small, as
In Rl-040719, Qualcomm, " Link Performance with different RV for Low Data Rates ",
In Cannes, France, June 2004 shown in as.
Therefore it is that best performance is also obtained in SHO, it is proposed that be used together specific TFC with IR.Due to IR schemes
Negligible difference, it is unreasonable that dominant rule is specified in the selection for being RV.Therefore, it is suggested that implicitly and based on CFN classes
Be similar to described in several other documents [2,3] as carry out RV selections.
In SHO, then no longer needs and can reuse for dibit RSN described in non-SHO.It keeps it in mind,
For in SHO HARQ related signalings sixty-four dollar question be avoid Node B buffer damage (corruption), it is proposed that make again
Use two RSN bits as dibit NDI.It is compareed with the RSN for retransmitting incremental every time, NDI is at UE with initial transmission each time
And it is incremented by.Compared with RSN, NDI provides higher reliability, especially will after the first retransmission remaining BLER with 1%
In the case of target.
3. summarizing
In this document, it is proposed that dibit is only used only and is used to be used to new data instruction and RV selection functions
HARQ correlation UL E-DCH signaling informations.
According to selected TFC, counter is based on each re-transmission (=RSN) or each initial transmission (=NDI) and passs
Increase, to support NDI RV selection functions as well as possible.
The background of the solution is the given fact, i.e., in the case of SHO the reliability ratio of NDI in non-SHO
It is even more critical, and the selection of RV is more much bigger to the influence of performance than in SHO in non-SHO.
Further it is proposed that the RV selections being also contained in SHO, this implicitly utilizes those for example such as described in [2,3] of CFN
Sample carries out.In the case of non-SHO, for initial transmission and the RV of the first and second re-transmissions by network signal, and actually
Use by UE RSN indicate.
Abbreviation used in explanation：
E-DCH：Enhancement type special channel, the substantially improvement to existing UMTS uplink channels.
E-DCDH：Enhanced dedicated data channel carries the payload data of E-DCH.
E-DCCH：Enhanced dedicated control channel, overhead data (L1 signaling) of the carrying for E-DCDCH.
IR：Steadily increase redundancy
SHO：Soft handover
CFN：Connection Frame Number
E-TFC：Enhanced transport format combination
UE：User equipment, mobile station
Log2：Logarithm bottom 2
UL：Uplink
RSN：Re-transmission sequence number
RSNMAX：The maximum value of re-transmission sequence number
TTIN：No. TTI
TTI：Transmission Time Interval
CFN：Connection Frame Number
ARQ：Automatic repeat request
HARQ：Mix ARQ
ACK：Response
NACK：Non- response
Floor (x), namely the maximum integer no more than x
RV：Redundancy versions
NARQ：ARQ is mixed into number of passes
UE：The synonym of user equipment, mobile station or terminal
Claims (24)
1. a kind of method that transmission is included in the information content at least one packet, at least one packet are logical
It crosses data link and is transferred to receiver from transmitter, wherein described information content is indicated by bit sequence, the bit sequence
It is encapsulated in the redundancy versions that can be transmitted, the method includes：
For the first time the described information content in the first packet, wherein described information content are transmitted from sender to receiver
By can self-demarking code the first redundancy versions and first rate match pattern indicate, the first rate match pattern be selected from by table
The set that at least two rate matching patterns that the redundancy version parameters of bright rate matching pattern determine are constituted；
The confirmation confirmed to incorrect reception is received from receiver at transmitter；
One receives the confirmation confirmed to incorrect reception, is retransmitted from sender to receiver at least one first time
Described information content in second packet, thus using can the second redundancy version of self-demarking code indicated described at least one originally
The information content in the second packet for the first time, according to encoding rate selection can self-demarking code the second redundancy versions, if compiled
Code check is less than predefined upper encoding rate, then selection can self-demarking code redundancy versions, utilize redundancy version parameters to select another speed
Rate match pattern, wherein another rate matching pattern is different from the first rate used in the first packet matching mould
Formula；And
Another confirmation confirmed to incorrect reception is received from receiver at transmitter；
One receives another confirmation confirmed to incorrect reception, is retransmitted at least second from sender to receiver
Described information content in third packet, thus according to another encoding rate using can self-demarking code third redundancy versions, such as
Another encoding rate described in fruit be higher than predefined upper encoding rate, then selection can self-demarking code redundancy versions, joined using redundancy versions
Number selects other rate matching patterns, wherein other described rate matching patterns be different from using in the first packet the
One rate matching pattern and identical as another rate matching pattern used in the second packet.
2. according to the method described in claim 1, wherein, predefined upper encoding rate is equal to 1/2.
3. according to the method described in claim 1, wherein, selection can the second redundancy versions of self-demarking code be based on parameter of punchinging
, it is described to punching parameter description whether in the case where punchinging to bit sequence or in the case where not punchinging to bit sequence
Produced can self-demarking code the second redundancy versions.
4. according to the method described in claim 1, wherein, selection can the second redundancy versions of self-demarking code be based on the number of transmissions ginseng
Number, the number of retransmissions n of information of the number of transmissions parameter description in nth data grouping.
5. according to the method described in claim 4, wherein, the set of redundancy versions can be obtained from bit sequence, wherein described
First subset of the set of redundancy versions can be used for the number of transmissions of the first range, and the second of the set of the redundancy versions
Subset can be used for the number of transmissions of another range.
6. according to the method described in claim 5, wherein, part is different each other for the first subset and second subset.
7. according to the method described in claim 5, wherein, the first subset and second subset are completely different from each other.
8. according to the method described in claim 5, wherein, the second subset of the set of the redundancy versions is the super of the first subset
Collection.
9. according to the method described in claim 1, further including：
Send at least one re-transmission of described information content；
Determine the number of the information content of transmitted in parallel；And
The number of described information content is used as to the process number parameter of the redundancy versions for selecting to transmit.
10. according to the method described in claim 9, further including：
The redundancy version parameters and the process number parameter are combined, to obtain another parameter.
11. according to the method described in claim 1, further including：
As soon as receive control confirm, number of retransmissions is reset to initial value, wherein the control confirm show to transmit it is superfluous
The incorrect decoding of remaining version is foreseen.
12. according to the method described in claim 1, further including：
The redundancy versions for describing to transmit by parameter s, wherein the redundancy versions that can be transmitted described in parameter s instructions whether be
It can self-demarking code.
13. a kind of terminal for the information content being used for transmission included at least one packet, at least one data
Grouping is transferred to base station by data link from the terminal, and wherein described information content is indicated by bit sequence, the ratio
Special sequence is encapsulated in the redundancy versions that can be transmitted, and the terminal includes：
Transmitter, the transmitter are configured to：
For the first time the described information content in the first packet, wherein described information are transmitted from the terminal to the base station
Content by can self-demarking code the first redundancy versions and first rate match pattern indicate that the first rate match pattern is selected from
The set constituted by showing at least two rate matching patterns that the redundancy version parameters of rate matching pattern determine；
Receiver, the receiver are configured to：
The confirmation confirmed to incorrect reception is received from the base station at the terminal；
The transmitter is further configured to：
One receives the confirmation confirmed to incorrect reception, heavy from the terminal to the base station at least one first time
Pass described information content in the second packet, thus utilize can self-demarking code the second redundancy versions for expression at least one
The information content of a first time in the second packet, according to encoding rate selection can self-demarking code the second redundancy versions, if
Encoding rate be less than predefined upper encoding rate, then selection can self-demarking code redundancy versions, using redundancy version parameters select it is another
Rate matching pattern, wherein another rate matching pattern is different from the first rate used in the first packet matching
Pattern；
The receiver is further configured to：
Another confirmation confirmed to incorrect reception is received from the base station at the terminal；And
The transmitter is further configured to：
One receives another confirmation confirmed to incorrect reception, from the terminal to the base station at least second
Secondary described information content of the re-transmission in third packet, thus according to another encoding rate using can self-demarking code third redundancy
Version, if another encoding rate be higher than predefined upper encoding rate, selection can self-demarking code redundancy versions, utilize redundancy
Release parameter selects other rate matching patterns, wherein other described rate matching patterns are different from making in the first packet
First rate match pattern and identical as another rate matching pattern used in the second packet.
14. terminal according to claim 13, wherein predefined upper encoding rate is equal to 1/2.
15. terminal according to claim 13, wherein selection can the second redundancy versions of self-demarking code be based on punchinging parameter,
It is described punching parameter description whether in the case where punchinging to bit sequence or in the case where not punchinging to bit sequence
Produce can self-demarking code the second redundancy versions.
16. terminal according to claim 13, wherein selection can self-demarking code the second redundancy versions be based on the number of transmissions ginseng
Number, the number of retransmissions n of information of the number of transmissions parameter description in nth data grouping.
17. terminal according to claim 16, wherein the set of redundancy versions can be obtained from bit sequence, wherein described
First subset of the set of redundancy versions can be used for the number of transmissions of the first range, and the second of the set of the redundancy versions
Subset can be used for the number of transmissions of another range.
18. terminal according to claim 17, wherein part is different each other for the first subset and second subset.
19. terminal according to claim 17, wherein the first subset and second subset are completely different from each other.
20. terminal according to claim 17, wherein the second subset of the set of the redundancy versions is the first subset
Superset.
21. terminal according to claim 13, wherein the transmitter is further configured to send described information content extremely
A few re-transmission；And
The terminal is configured to：
Determine the number of the information content of transmitted in parallel;And
The number of described information content is used as to the process number parameter of the redundancy versions for selecting to transmit.
22. terminal according to claim 21, wherein the terminal is further configured to the redundancy version parameters and institute
The conjunction of process array is stated, to obtain another parameter.
23. terminal according to claim 13, wherein the terminal is further configured to, and is confirmed as soon as receiving control, will
Number of retransmissions is reset to initial value, wherein the control confirms that the incorrect decoding for the redundancy versions for showing to transmit is foreseen.
24. terminal according to claim 13, wherein what the terminal was further configured to can to transmit by parameter s description
Redundancy versions, wherein whether the redundancy versions that can be transmitted described in parameter s instructions can self-demarking codes.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP04021970A EP1638238A1 (en) | 2004-09-15 | 2004-09-15 | Method for combining data packets by a telecommunication device |
| EP04021970.1 | 2004-09-15 | ||
| EP04027144.7 | 2004-11-15 | ||
| EP04027144A EP1657844A1 (en) | 2004-11-15 | 2004-11-15 | Decoding method |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2005800311207A Division CN101023618A (en) | 2004-09-15 | 2005-09-15 | Decoding method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104253679A CN104253679A (en) | 2014-12-31 |
| CN104253679B true CN104253679B (en) | 2018-08-31 |
Family
ID=
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5657325A (en) * | 1995-03-31 | 1997-08-12 | Lucent Technologies Inc. | Transmitter and method for transmitting information packets with incremental redundancy |
| CN1351438A (en) * | 2000-10-31 | 2002-05-29 | Lg电子株式会社 | packet data transmission with mixed automatic request to resend |
| CN1462516A (en) * | 2001-04-25 | 2003-12-17 | 皇家菲利浦电子有限公司 | Radio communication system |
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5657325A (en) * | 1995-03-31 | 1997-08-12 | Lucent Technologies Inc. | Transmitter and method for transmitting information packets with incremental redundancy |
| CN1351438A (en) * | 2000-10-31 | 2002-05-29 | Lg电子株式会社 | packet data transmission with mixed automatic request to resend |
| CN1462516A (en) * | 2001-04-25 | 2003-12-17 | 皇家菲利浦电子有限公司 | Radio communication system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8935588B2 (en) | Decoding method | |
| EP1584155B1 (en) | Operation of a forward link acknowledgement channel for the reverse link data | |
| US7346035B2 (en) | System and method for retransmitting uplink data from a mobile terminal in a soft handover region in an asynchronous CDMA mobile communication system servicing an enhanced uplink dedicated transport channel | |
| JP4528830B2 (en) | Data packet combining method by communication equipment | |
| US20100146354A1 (en) | Method and apparatus for h-arq in a wireless communication system | |
| EP2249505A1 (en) | Physical Layer Packet Retransmission Handling in WCDMA in Soft Handover | |
| KR101087117B1 (en) | Power for uplink acknowledgment transmission | |
| US8151154B2 (en) | Method and a transmitter/receiver for reduced signaling in a retransmission system using hybrid automatic repeat | |
| CN1954533A (en) | Method and system for providing autonomous retransmissions in a wireless communication system | |
| TW201006165A (en) | Exploiting known rate matching information in blind decoding of downlink wireless data transmissions | |
| US20090238148A1 (en) | Arrangement And Method In A Mobile Communication Network | |
| EP1624606A1 (en) | Apparatus and method for radio transmission | |
| CN104253679B (en) | Coding/decoding method | |
| US8121555B2 (en) | Transmission power control in HSPDA | |
| CN101208897A (en) | Apparatus and method for mixing automatic request for repetition in wireless communication system | |
| EP1657844A1 (en) | Decoding method | |
| CN1735004A (en) | The equipment of transmitting redundancy edition signaling and method in the Wideband Code Division Multiple Access (WCDMA) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |