Dissociation rate matching method
Technical field
The present invention relates to the mobile communication technology field, be specifically related to the dissociation rate matching method in a kind of 3G (Third Generation) Moblie long evolving system (hereinafter to be referred as 3G LTE).
Background technology
In the up link of GSM, the code multiplexing of transmission channel is handled and to be comprised that mainly CRC adds (Cyclic Redundancy Code, CRC), chnnel coding, rate-matched, interweaves and step such as transmission channel is multiplexing.Can consult among the international standard protocol 3GPP36.212 of 3G LTE the code multiplexing to transmission channel and handle the regulation of carrying out, the part relevant with the present invention carried out brief description here, the general multiplexing process process of transmission channel coding generally comprises following steps:
Step a: to the additional CRC of transmission block;
Step b: transmission block is carried out code block cut apart, to the additional CRC of each code block after cutting apart;
The data of transmission channel are unit with a transmission block; Because transmission block is big or small changeable; And the Turbo encoder has the restriction of a scope to the coding code block length; Therefore to carry out code block to transmission block and cut apart, carry out code block for transmission block and cut apart greater than the restriction of Turbo coding maximum length, and because the Turbo coding of LTE uses QPP (quadratic permutation polynomials; Arrange multinomial 2 times) interleaver; The size of the code block length after cutting apart must satisfy in the 3GPP36.212 agreement value among the table 5.1.3-3, and the dummy bit that therefore also need add right quantity in the front of first code block for the code block after cutting apart satisfy the requirement that code block is cut apart, and does not cut apart and need not pass through code block for the transmission block of convolutional encoding;
Step c: each code block is carried out chnnel coding, and chnnel coding is divided into convolutional encoding and Turbo coding, is divided into the output of 3 sub-block through the data behind the coding;
Steps d: the data after the chnnel coding are carried out rate-matched.
The purpose of rate-matched is the amount of bits that can carry for the physical channel that makes amount of bits behind the coding satisfy the special use that distributes.3GPP36.212 the rate-matched of stipulating in the agreement comprises: sub-block interleaving, bit collection and bit are selected and are pruned.In sub-block interleaving, be to interweave with block interleaver; When the group block length does not satisfy the total length of block interleaver also needs add suitable dummy bit in the front of every sub-block; And then interweave, sub-block interleaving is to carry out respectively to 3 sub-block after the chnnel coding, so need carry out bit collection to 3 sub-block after accomplishing sub-block interleaving; Carry out merging into 1 data block behind the bit collection; Then this data block is carried out bit selection and pruning, in selecting and pruning, will judge at first whether this bit is dummy bit, if then will remove this dummy bit; Carry out repetition (Repeated) or punching (Punctured) operation according to the amount of bits of distributing to the physical channel of this sub-block then; For the transmission block of Turbo coding, the dummy bit here comprises dummy bit two parts that code block adds when cutting apart dummy bit and the sub-block interleaving of interpolation, and for the transmission block of convolutional encoding; Owing to need not pass through code block and cut apart, the dummy bit that adds when therefore the dummy bit here includes only sub-block interleaving.The flow process of rate-matched is as shown in Figure 1, among the figure
With
Be 3 sub-block of exporting after the chnnel coding,
With
Be the dateout behind the sub-block interleaving, w
kFor inciting somebody to action
With
Carry out the data behind the bit collection, e
kBe w
kThrough the data after bit selection and the pruning.The implementation of concrete rate-matched is seen the 3GPP36.2125.1.4 joint.
In the down link of GSM; Receiving terminal then need be separated rate-matched to the data that receive; It is corresponding with the rate-matched process to separate the rate-matched process; Traditional method of separating rate-matched is the position that calculates the dummy bit that when code block is cut apart with sub-block interleaving, adds earlier according to the method for rate-matched; The dummy bit that adds when cutting apart with sub-block interleaving at code block then adds in the data that receive, and at last the data of having added dummy bit is separated rate-matched according to the inverse process of rate-matched.Because the dummy bit that adds during sub-block interleaving only is used for sub-block interleaving or deinterleaving; Need not be used for the decoding of back; And need be used for the decoding of back at the dummy bit that code block is cut apart interpolation, the dummy bit that adds when therefore traditional method of separating rate-matched also need be removed sub-block interleaving at last.
The method of separating rate-matched of traditional T urbo coding is following:
Step 1: according to the size of transmission block, the number that the mode of cutting apart according to code block calculates the size of each code block and when code block is cut apart, adds dummy bit;
Step 2: the number that calculates the dummy bit that adds in the sub-block interleaving;
Step 3: the position that calculates dummy bit according to the size of the top every sub-block that calculates and the summation of when code block is cut apart with sub-block interleaving, adding dummy bit according to the mode of sub-block interleaving and bit collection;
Step 4: the dummy bit that adds when cutting apart with sub-block interleaving at code block according to the position of dummy bit is inserted in the data that receive and goes;
Step 5: separate repetition or separate the punching process according to the amount of bits of the physical channel of each code block and the physical length of code block;
Separate repetition or separate punch operation corresponding to repetition in the rate-matched or punch operation, separate and repeat to abandon the data of repetition or the mode that the data that repeat are merged for the The data that repeats exactly, separating punching is exactly that the data of punching are replenished up.
Step 6: the mode according to bit collection is separated bit collection, and code block is divided into 3 sub-block;
It is corresponding with bit collection to separate bit collection, separates bit collection and is the data after separating repetition or separating punching are decomposed into 3 sub-block according to the inverse process of bit collection.
Step 7: 3 sub-block are carried out sub-piece deinterleaving respectively, remove the dummy bit that adds in the sub-block interleaving then;
The method of separating rate-matched of traditional convolution code is following:
Step 1: the size according to transmission block calculates the number of adding dummy bit in the sub-block interleaving;
Step 2: the position of adding dummy bit when calculating sub-block interleaving according to the mode of sub-block interleaving and bit collection;
Step 3: the dummy bit that adds during with sub-block interleaving according to the position of dummy bit is inserted in the data that receive and goes;
Step 4: separate repetition or separate the punching process according to the amount of bits of the physical channel of every sub-block and the physical length of sub-piece;
Step 5: sub-piece is separated bit collection according to the mode of bit collection, be divided into 3 data flow;
Step 6: 3 data flow points are not carried out sub-piece deinterleaving, remove the dummy bit that adds in the sub-block interleaving then;
Existing dissociation rate matching method implementation procedure is loaded down with trivial details; Need calculate the position of the dummy bit of destroying according to the mode of rate-matched; Then dummy bit is inserted in the reception data, this need carry out lot of data moves, and therefore needs bigger memory space and long processing time; The treatment circuit area can not further be dwindled, and power consumption is bigger simultaneously.
Summary of the invention
In view of this, in order to address the above problem, the present invention provides a kind of dissociation rate matching method, and the handling process of simplified solution rate-matched has reduced the data volume that needs processing, has improved treatment effeciency.
The objective of the invention is to realize like this, dissociation rate matching method may further comprise the steps:
The code element of 1) needs that receive being separated rate-matched is separated repetition or is separated punch operation;
2) separate the bit collection operation;
3) data of separating behind the bit collection are carried out the deinterleaving operation.
Further, before the said step 1), also comprise the steps:
A) mode of cutting apart according to the size and the code block of transmission block, the size and the code block that obtain each code block are cut apart the middle number F that adds dummy bit;
B), obtain the address of sub-block interleaving and the number N of the dummy bit that in sub-block interleaving, adds according to the size of interleaver formula and code block
d
C) according to step a), b) the result, calculate code block cut apart in sub-block interleaving in the dummy bit sum that adds, again according to the interleaver address, obtain the point of addition of each dummy bit;
D) N before obtaining in the rate-matched in the data behind the bit collection according to the position of dummy bit
CbThe total length L en1 of non-dummy bit in the individual data, said N
CbQuantity for soft buffer;
E) k before obtaining in the rate-matched in the data behind the bit collection according to the position of dummy bit
0The total length L en2 of non-dummy bit in the individual data, wherein
Be the line number of sub-block interleaver, rv
IdxBe redundancy versions;
Said step 2) also comprises the steps: before
F) will separate repetition or separate punching after code element in the data montage of back Len2 length anterior to said code element;
Further, in the said step a), cut apart middle number of adding dummy bit according to the computes code block:
F=C
+·K
++C
-·K
--B′
C wherein
+And C
-The size of expression code block, K
+And K
-Expression is divided into C respectively
+And C
-The number of code block, B ' expression code block adds the length overall behind the check bit;
Further, in the said step b), according to the number of the address of the sub-block interleaving of computes and the dummy bit that in sub-block interleaving, adds:
Wherein
and
representes the line number and the columns of sub-block interleaver respectively, and D representes to import the data length of interleaver;
Further; Step 1) specifically comprises the steps: the amount of bits E of the physical channel of comparison Len1 and each code block; Greater than Len1, then the data that receive are separated reprocessing like E, like Len1 more than or equal to E; Then the data that receive are separated punch operation, the data length after separating repetition or separating punching is Len1;
Further, said step 2) specifically comprise the steps: to calculate the interleaver address according to the interleaver formula, obtain the address of deinterleaving according to the sum of the dummy bit that when code block is cut apart with sub-block interleaving, adds that comprises in every sub-block;
Further, the number of the dummy bit that also comprises the steps: after the said step 3) to add when cutting apart according to code block is added corresponding dummy bit before the data of first code block, and the value of the dummy bit of interpolation is 0;
Further; Said step 1) specifically comprises the steps: the size of the amount of bits E that comparison is transmitted through the transport block length 3T after the convolutional encoding and each transmission block on physical channel, if 3T=E, then data are separated repetitive operation; If 3T>E, then separate punch operation to data;
Further, said step 2) in, the data of separating repetition or separating punching is equally divided into 3 sub-block, accomplishes and separate the bit collection operation;
Further, in the said step 3), 3 sub-block of separating behind the bit collection are carried out deinterleaving according to the number of the dummy bit that adds when the sub-block interleaving that comprises in every sub-block:
301) calculate the interleaver address according to the interleaver formula;
302), calculate the address of sub-block interleaving and the number of the dummy bit that in sub-block interleaving, adds according to the interleaver formula;
The number of the dummy bit that 303) adds when being included in sub-block interleaving in every sub-block calculates the address of deinterleaving.
The present invention proposes a kind of method of simply separating rate-matched; The dummy bit that adds in the time of before separating repetition or separating punch operation, need code block not being cut apart with sub-block interleaving is inserted in receiving data; Directly separate repetition or separate punch operation, separate bit collection and deinterleaving then receiving data.Wherein, The rate-matched of separating for Turbo code; The present invention compares with traditional rate-matched of separating, and has cancelled the dummy bit that adds when cutting apart with sub-block interleaving at code block and has been inserted into the process in receiving data, only need calculate in the rate-matched in the data behind the bit collection before N
CbIndividual and preceding k
0The length of non-dummy bit can be separated repetition or separated punch operation in the individual data; For the rate-matched of separating of convolution code, the present invention then more simplifies, and directly can separate repetition or separate punch operation receiving data according to the size of transmission block; Compared with prior art; The present invention simplifies the process of understanding rate-matched; The dummy bit that adds in the time of before separating repetition or separating punch operation, need code block not being cut apart with sub-block interleaving is inserted in receiving data, has reduced the data handling process, greatly reduces the processing time; And reduced the data volume that needs processing, improved treatment effeciency.Because it is simple that the more traditional processing procedure of processing procedure is wanted, therefore be beneficial to software and realize, practiced thrift code space, reduce the treatment circuit area and the time power consumption.
Other advantages of the present invention, target; To in specification subsequently, set forth to a certain extent with characteristic; And to a certain extent,, perhaps can from practice of the present invention, obtain instruction based on being conspicuous to those skilled in the art to investigating of hereinafter.Target of the present invention and other advantages can be passed through following specification, claims, and the structure that is particularly pointed out in the accompanying drawing realizes and obtains.
Description of drawings
In order to make the object of the invention, technical scheme and advantage clearer, will combine accompanying drawing that the present invention is made further detailed description below:
Fig. 1 shows the flow process of rate-matched;
Fig. 2 shows the flow process of separating rate-matched that is used for convolution code;
Fig. 3 shows the flow process of separating rate-matched that is used for the Turbo coding.
Embodiment
Below will carry out detailed description to the preferred embodiments of the present invention with reference to accompanying drawing.
Referring to Fig. 3, be used for the dissociation rate matching method of Turbo coding, may further comprise the steps:
101) mode of cutting apart according to the size and the code block of transmission block obtains the size of every sub-block, adds the number F of dummy bit in cutting apart according to the computes code block:
F=C
+·K
++C
-·K
--B′
C wherein
+And C
-The length of expression code block, K
+And K
-Expression is divided into C respectively
+And C
-The number of code block, B ' expression code block adds the length overall behind the check bit;
102) according to the interleaver formula, according to the number N of the address of the sub-block interleaving of computes and the dummy bit that in sub-block interleaving, adds
d:
Wherein
and
representes the line number and the columns of sub-block interleaver respectively, and D representes to import the data length of interleaver; The calculating of the address of interleaver can save with reference to 3GPP36.212 agreement 5.1.2.
103) according to step 101), 102) the result, meter obtain code block cut apart in sub-block interleaving in the dummy bit sum Q=F+N that adds
d,, obtain the position Addr_null [] of each dummy bit again according to the address Addr [] of interleaver.Here defining initial address is 0, and the computing formula of the address of dummy bit is expressed as so: when Addr [i] during Q, Addr_null [j]=Addr [i], i=0 wherein, 1, L, N-1; J=0,1, L, Q-1; N representes the length of interleaver.
104) N before obtaining in the rate-matched in the data behind the bit collection according to the position of dummy bit
CbThe total length L en1 of non-dummy bit in the individual data, said N
CbQuantity for soft buffer;
105) k before obtaining in the rate-matched in the data behind the bit collection according to the position of dummy bit
0The total length L en2 of non-dummy bit in the individual data, wherein
Be the line number of sub-block interleaver, rv
IdxBe redundancy versions;
106) code element that receives is separated repetition or separated punch operation; The amount of bits E of the physical channel of Len1 and each code block at first relatively, if E greater than Len1, then the data that are transmitted of expression some repeat to send; Therefore the data that receive are separated reprocessing; The data that are repeated to send can merge, and perhaps directly abandon the data of repetition, if Len1 is more than or equal to E; Some has been struck off then to represent the actual data that will send; Therefore will the data of being struck off be added in the data flow that receives, the value of interpolation is 0, and the data length after separating repetition or separating punching is Len1;
107) will separate repetition or separate punching after the data montage of code element back Len2 length before the said code element;
108) data are separated bit collection operation: the data after the position of the dummy bit that adds when cutting apart with sub-block interleaving at code block and the mode of bit collection will be separated repetition or separated punching have been divided into 3 sub-block;
109) data of separating behind the bit collection are carried out the deinterleaving operation: calculate the interleaver address according to the interleaver formula; The address computation formula of interleaver can obtain the address of deinterleaving referring to 3GPP36.212 agreement 5.1.4 joint according to the sum of the dummy bit that when code block is cut apart with sub-block interleaving, adds that comprises in every sub-block; Computing formula is following:
As Addr [i] >=during Q, De_addr [Addr [i]-Q]=j
I=0 wherein, 1, L, N-1; J=0,1, L, N-Q; The address sequence of Addr [] expression interleaver, the address sequence of De_addr [] expression deinterleaving, N representes the length of interleaver, the sum of the dummy bit that Q representes when code block is cut apart with sub-block interleaving, to add in every sub-block;
110) add dummy bit according to the dummy bit number of cutting apart interpolation at code block.
Referring to Fig. 2, be used for the speed method of separating of convolution code, comprise the steps:
201) data of separating rate-matched are separated repetition or separated punch operation with the amount of bits E that each transmission block transmits on physical channel according to the size of transmission block: the size of establishing transmission block comprises that the length of CRC is T; Be 3T then through length after the convolutional encoding; If 3T=E, some repeats the data that then are transmitted to send, and therefore the data of separating rate-matched is separated reprocessing; The data that are repeated to send can merge; Perhaps directly abandon the data of repetition, if 3T>E, some has been struck off then to represent the actual data that will send; Therefore will the data of being struck off be added in the data flow of separating rate-matched, the value of interpolation is 0;
202) separate bit collection: the data that will separate repetition or separate punching are equally divided into 3 sub-block;
203) 3 sub-block of separating behind the bit collection are carried out deinterleaving according to the number of the dummy bit that adds when the sub-block interleaving that comprises in every sub-block:
2031) calculate the interleaver address according to the interleaver formula;
2032), calculate the address of sub-block interleaving and the number of the dummy bit that in sub-block interleaving, adds according to the interleaver formula;
The number of the dummy bit that 2033) adds when being included in sub-block interleaving in every sub-block calculates the address of deinterleaving, with the aforementioned dissociation rate matching method step 109 that is used for the Turbo coding) method that adopts is identical, repeats no more at this.
The above is merely the preferred embodiments of the present invention, is not limited to the present invention, and obviously, those skilled in the art can carry out various changes and modification and not break away from the spirit and scope of the present invention the present invention.Like this, belong within the scope of claim of the present invention and equivalent technologies thereof if of the present invention these are revised with modification, then the present invention also is intended to comprise these changes and modification interior.