CN101835184A - Method of LTE (Long Term Evolution) downlink auxiliary synchronizing channel detection with low complexity - Google Patents

Abstract

The invention relates to a method of LTE (Long Term Evolution) downlink auxiliary synchronizing signal detection with low complexity, belonging to the technical field of mobile communication. The method comprises the following steps of: separating odd bits and even bits of an auxiliary synchronizing sequence of the received frequency-region signals to acquire an odd sequence and an even sequence; respectively carrying out sequence rearrangement, fast transformation of a Hadamard matrix and sequence rearrangement again on the odd sequence and the even sequence after descrambling to realize serial correlation operation; screening the identification number m0 corresponding to the maximal correlation value from the odd sequence; reducing the serial number range of the even sequence according to the corresponding relation between m0 and m1; screening the identification number m1 corresponding to the maximal correlation value from the even sequence of which the serial number range is reduced; and acquiring the corresponding cell identification mark according to the size relation between m0 and m1, wherein the subframe is a subframe 0 or a subframe 5. The invention simplifies the algorithm complexity of S-SCH signal detection from multiplication of 31*31 times to addition of 5*32 times, thereby greatly reducing the computation complexity in the detection process.

Description

The method that a kind of LTE downlink subsidiary synchronizing channel of low complex degree detects

Technical field

The invention belongs to wireless communication technology field, be specifically related to a kind of method of LTE downlink subsidiary synchronizing channel detection of low complex degree.

Background technology

(3rd Generation Partnership Project Long Term Evolution, 3GPP-LTE) mobile communication system is the third generation (3rd generation, 3G) evolution of mobile communication system to third generation partner program Long Term Evolution.

Down-going synchronous is the important content of Cell searching, and travelling carriage need be obtained the synchronous of time, frequency from the descending synchronous signal channel that cell base station sends, obtain cell ID.In the LTE system, travelling carriage can pass through primary synchronization channel (primarysynchronization channel, P-SCH) and auxiliary synchronization channel (secondary synchronization channel S-SCH) finishes down-going synchronous.Particularly, travelling carriage is finished symbol timing and frequency offset correction, is carried in the cell ID group and number by P-SCH, and carries out channel estimating; Carry out the 10ms frame regularly synchronously by S-SCH, and detected cells sign group number.

Utilize the correlation properties of auxiliary synchronous signals when S-SCH assists synchronous detecting at present, do coherent detection at frequency domain to received signal, this detection method will take out the secondary synchronization sequences position from receive subframe OFDM symbol and all local alternative auxiliary synchronous signals carry out related calculation in twos, obtain a plurality of correlations, and detect maximum in all correlations, determine actual length type of cyclic prefix and the auxiliary synchronous signals that adopts according to maximum.After detecting auxiliary synchronous signals, can determine cell ID group number (0～167).

Be the sequence of N if signal is a length, utilize present coherent detection algorithm to detect a signal and just need carry out N * N multiplying, have certain complexity.The rudimentary algorithm that traditional signal correction detects is to carry out coherent detection respectively for strange/even number part of 31 of the S-SCH signal of the subframe that receives, and needs to calculate 31 * 31 (O (N at every turn ²)) inferior computing.

Summary of the invention

The objective of the invention is to propose a kind of LTE downlink subsidiary synchronizing channel detection method that can significantly reduce computation complexity.

The detection method that the present invention proposes is the detection mode of the sharp m of employing sequence Fast transforms, replaces the coherent detection in traditional auxiliary synchronization channel detection.Its amount of calculation is 5 * 32 sub-additions, reduces the complexity of auxilliary synchronous detecting greatly.

The m sequence is the abbreviation that the maximum cycle linear displacement is deposited sequence, and its length is n=2 ^M-1Generate by m level linear feedback shift register.This sequence is that its cycle of periodic sequence is n, and each cycle comprises 2 ^M-1Individual logical one and 2 ^M-1-1 logical zero.In the practical application logical zero and 1 is become bipolarity level 1 and-1, the bipolar code that is of the present invention.Because the m sequence has good autocorrelation performance, can utilize it to carry out the detection of m sequence, concrete grammar is relevant with reception m sequence work for the different cyclic shifts with local m sequence, determines to receive the displacement of m sequence by the correlation under the more different cyclic shifts.The circular matrix that whole related operation can be expressed as being made up of local cyclic vector carries out multiplication mutually with the reception vector.Note m sequence is I=0,1 ..., n-1; The circular matrix of its formation is

The capable j column element of i of this matrix is defined as M _IjBecause each row of this circular matrix is formed through after the different cyclic shifts by same m sequence, can know

I wherein, j=0,1 ..., n-1.The column vector that the definition receiving sequence is formed is expressed as

The different cyclic shifts of then local m sequence are made associative operation with reception m sequence and can be expressed as

The vector of forming for the correlation of related operation gained.

Be the matrix of n * n, so

Computation complexity be O (n ²).

With the hadamard battle array following relation is arranged:

$\tilde{M}=P_L{\mathrm{HP}}_S$

P wherein _SWith P _LBe respectively row and rank transformation matrix, H is a hadamard matrix.Above-mentioned formula shows, rearranges by row and row to a hadamard matrix, can obtain correlation matrix Because the hadamard matrix related operation can be finished with plus-minus iteration fast algorithm, so detection method of the present invention is utilized row-column transform matrix P exactly _LAnd P _SReceiving sequence is carried out order reset, utilize the coherent detection step of hadamard matrix fast algorithm implementation then, specifically comprise the steps: the S-SCH signal

Step 1: according to " 3rd Generation Partnership Project (3GPP) Technical Specification GroupRadio Access network; Evolved Universal Terrestrial Radio Access (E-UTRA); Physicalchannels and modulation, Release 8,3GPP TS 36.211 V8.5.0.2008 " form of the local correlation matrix of agreement regulation, obtain P _LAnd P _SThe concrete form of battle array, and the form of calculation of equal value of the correlation matrix after the acquisition augmentation:

For subsequent detection is prepared.

Wherein Being 32 * 32 circular matrix, is according to the augmented matrix behind the lastrow of the local correlation matrix of " 3GPP TS 36.211 V8.5.0.2008 " agreement defined and the first from left row cover; H is a hadamard matrix; P _L, P _SIt is respectively the ranks permutation matrix.Concrete

Production process comprises:

" 3rd Generation Partnership Project (3GPP) Technical Specification Group Radio Accessnetwork; Evolved Universal Terrestrial Radio Access (E-UTRA); Physical channels andmodulation, Release 8,3GPP TS 36.211 V8.5.0.2008 " the agreement defined the descending auxilliary 31 long sequences that generate by shift LD in synchronously.31 basic sequence cyclic shifts form the cyclic shift matrices that 31 row 31 are listed as thus.Element in this matrix is converted to logical zero and 1 by bipolarity level+1 and-1, and the circular matrix that comprises logical zero and 1 element after the note conversion is designated as M.M can be decomposed into the product of one 31 * 5 matrix L and one 's 5 * 31 matrix S, promptly be expressed as

M＝LS

Wherein S constitutes L=SA by preceding 5 row elements of M ^-1Wherein but A is the inverse square matrix that preceding 5 column elements by matrix S constitute, and the computing in the regulation matrix is the mould binary arithmetic operation.And can find every row of S battle array and every row of L battle array are considered as the binary system ordered series of numbers, can form natural sequence 1,2,3 just ... 31 out of order arrangement.

Remember that 32 rank hadamard square formations are H, it can be decomposed into H=BB ', and wherein B is 32 * 5 matrix

$B={\left(\begin{array}{ccccc}0& 0& 0& 0& 0\\ 0& 0& 0& 0& 1\\ 0& 0& 0& 1& 0\\ 0& 0& 0& 1& 1\\ & & ...& & \\ 1& 1& 1& 1& 1\end{array}\right)}_{32*5}$

According to the character of hadamard matrix as can be known, if every row of B is considered as binary sequence, then be the arrangement according to the order of sequence of natural sequence 0～31 just.

Go up most a row in the M battle array and add a row 0, the first from left row add row 0, this new matrix are designated as the augmented matrix of M

Then can be right

Be decomposed into following form:

$\tilde{M}=\tilde{L}\tilde{S}$

Wherein

For adding a row 0 on, L forms 32 * 5 matrix;

Adding row 0 for S is the most left and form, is 5 * 32 matrix.Like this,

It is consistent with the dimension of B, Consistent with the dimension of B '.Only needing that the B matrix is carried out line translation can obtain

B ' is carried out rank transformation can be obtained

That is:

$\tilde{L}=P_{L}B;$ $\tilde{S}=B^{\′}{P}_S;$

By

With the line translation of B relation and With Rank transformation relation, can obtain corresponding P _LAnd P _SConcrete form.P _LAnd P _SIt is corresponding 32 * 32 ranks permutation matrix.

Comprehensively obtain correlation matrix after the augmentation by above step Can become following form by hadamard matrix and ranks displacement matrix representation:

Step 2: S-SCH burst carries out parity bit and separates to received signal, obtains the sequence of 31 bit lengths that the sequence of 31 bit lengths of being made up of the odd bits signal and is made up of the even bit signal.

Step 3: the first place of 31 odd-numbered bit sequences obtaining in the step 2 is mended 0, make it become 32 sequences to be measured, make it consistent with the columns of correlation matrix after the augmentation.The auxiliary synchronization channel coherent detection can be converted into like this:

$\tilde{h}=\tilde{M}\·\tilde{y}=P_L{\mathrm{HP}}_S\·\tilde{y}$

Wherein

Be the burst of auxilliary synchronous odd-numbered bit sequence after first place benefit 0 that receives, here except P _LAnd P _SThe expression rearrangement process is consistent with unipolarity, all the other H,

Be actual bipolarity level signal and matrix.Above-mentioned formula is represented: at first utilize P _SThe vector that 32 odd number sequences of mending after 0 constitute is reset, then the sequence after resetting is carried out the hadamard Fast transforms, again to the sequence P after the conversion _LReset.From 32 bit sequences that obtain, filter out the call number m of maximum related value correspondence ₀The P that draws by step 1 _LAnd P _SConcrete form can be finished concrete rearrangement process.

Step 4: the 31 digit pair number sequences that obtain in the step 2 are listed in first mend 0, make it become 32 sequences to be measured equally, the m that utilizes step 3 to obtain ₀Information form relevant scrambler, and with carrying out descrambling by relevant scrambler dual numbers bit sequence.Utilize matrix P _SThe even number sequence of mending after 0 is reset, the sequence after resetting is carried out the hadamard Fast transforms, the matrix P of the sequence after the conversion _LReset, from 32 bit sequences that obtain, according to the m of 3GPP TS 36.211 V8.5.0.2008 agreement defineds ₀With m ₁Corresponding relation, the m that obtains in conjunction with previous step ₀, select and m the ultimate sequence after resetting ₀Possible corresponding whole correlations filter out maximum related value in these correlations, then the call number of this correlation correspondence is m ₁

Step 5: according to m ₀And m ₁Magnitude relationship, determine that current subframe is that subframe 0 still is a subframe 5, wherein corresponding relation is: if m ₀＜m ₁, then current subframe is 5 work song frames, if m ₀＞m ₁, then current subframe is 5 work song frames, exchange m ₀And m ₁Value.And obtain corresponding district sign group number.

Present invention is directed at the S-SCH coherent detection, provided the optimized Algorithm and the device that utilize m sequence Fast transforms to propose, reduced the complexity of original related algorithm widely, and be easy to Project Realization.And the present invention utilizes m sequence Fast transforms and good correlation, multiplicative complexity all can be converted into addition by the hadamard Fast transforms, is log ₂32 * 32 times, i.e. 5 * 32 sub-addition computings.

Description of drawings

Fig. 1 is converted into hadamard matrix and the flow chart of resetting the battle array product for Metzler matrix.

Fig. 2 is the flow chart of the low complex degree S-SCH channel detection method of patent of the present invention.

Embodiment

In order to make purpose of the present invention, technical scheme and advantage are clearer, below in conjunction with accompanying drawing the specific embodiment of the present invention are described further.

Method of the present invention be to the S-SCH sequence of the frequency-region signal that receives carry out parity bit separate obtain odd number sequence and the first cover of even number sequence after, it is carried out " rearrangement-hadamard matrix Fast transforms-rearrangement " operation respectively, the related operation that replaces traditional direct and local correlation matrix, thereby the algorithm complex of S-SCH input by 31 * 31 multiplication, is reduced to 5 * 32 sub-additions.The cell ID group number by the secondary synchronization sequences call number to (m ₀, m ₁) unique definite.M wherein ₀, m ₁Detection respectively by odd bits coherent detection and even bit obtains.

Fig. 1 is the flow chart of fast detecting process of the present invention, as shown in Figure 1:

The present invention includes following steps:

Step 110～190: by shown in Figure 1 will be by " 3rd Generation Partnership Project (3GPP) TechnicalSpecification Group Radio Access network; Evolved Universal Terrestrial Radio Access (E-UTRA); Physical channels and modulation, Release 8,3GPP TS 36.211V8.5.0.2008 " the original local correlation matrix that produces in the agreement is converted to the relevant battle array of corresponding unipolarity, and the matrix that is designated as after the conversion is M.M is decomposed, be expressed as the product representation of two matrixes, i.e. M=LS, wherein S constitutes L=SA by preceding 5 row elements of M ^-1Wherein but A is the inverse square matrix that preceding 5 column elements by matrix S constitute, and matrix operation wherein is the mould binary arithmetic operation.

If with 2 ^mThe rank hadamard matrix is designated as

Its generation method is as follows

1 rank H ₁=[0]

2 rank $H_2=\left[\begin{array}{cc}0& 0\\ 0& 1\end{array}\right];$

2 ^mRank $H_{2^m}=\left[\begin{array}{cc}{H}_{2^{m-1}}& H_{2^{m-1}}\\ H_{2^{m-1}}& -H_{2^{m-1}}\end{array}\right].$

32 rank hadamard matrix H are decomposed, be expressed as H=BB ', wherein B is 32 * 5 matrix, and every row of B can be considered as by 0 to 31 binary sequence.

In step 120～130, add delegation's 0, one row 0 respectively at the lastrow of M battle array and the first from left row, constitute a new matrix, be designated as the augmented matrix of consideration M

Will

Be decomposed into

In the step 150, carry out line translation and can obtain decomposing B matrix that hadamard matrix obtains

B ' is carried out rank transformation can be obtained

That is:

P _LAnd P _SBe the ranks permutation matrixes, be 32 * 32 Sparse Array, for P _LAnd P _SNon-0 element particular location is as follows in the battle array, wherein P _L(i, j), P _S(i, j) difference representing matrix P _LWith matrix P _SThe capable j column position of i

P _L(1，1) P _L(6，2) P _L(5，3) P _L(23，4) P _L(4，5)

P _L(9，6) P _L(22，7) P _L(15，8) P _L(3，9) P _L(32，10)

P _L(8，11) P _L(11，12) P _L(21，13) P _L(30，14) P _L(14，15)

P _L(26，16) P _L(2，17) P _L(12，18) P _L(31，19) P _L(27，20)

P _L(7，21) P _L(24，22) P _L(10，23) P _L(16，24) P _L(20，25)

P _L(19，26) P _L(29，27) P _L(18，28) P _L(13，29) P _L(28，30)

P _L(25，31) P _L(17，32)

P _S(1，1) P _S(2，2) P _S(3，3) P _S(5，4) P _S(10，5)

P _S(19，6) P _S(6，7) P _S(12，8) P _S(23，9) P _S(13，10)

P _S(26，11) P _S(20，12) P _S(8，13) P _S(16，14) P _S(32，15)

P _S(31，16) P _S(29，17) P _S(25，18) P _S(18，19) P _S(4，20)

P _S(7，21) P _S(14，22) P _S(28，23) P _S(24，24) P _S(15，25)

P _S(30，26) P _S(27，27) P _S(22，28) P _S(11，29) P _S(21，30)

P _S(9，31) P _S(17，32)

Can in step 160, obtain the transformation relation formula thus:

The matrix that aforesaid H=BB ' obtains is that element only is 0 or 1 unipolarity matrix.In step 170～180,, need H entry of a matrix element is carried out following conversion for the bipolarity hadamard matrix: 0 →+1,1 →-1, obtain bipolarity 32 rank hadamard matrixs, because the equivalence of single bipolarity conversion, coherent detection can be expressed as:

With aforementioned:

For the sequence to be detected after the first place benefit 0, except P _LAnd P _SDo not participate in concrete calculating for resetting battle array, its complementary submatrix and vector are bipolarity in the formula.

In step 200 to received signal the S-SCH burst carry out parity bit and separate, obtain 31 odd number sequences and 31 digit pair number sequences row.And the scrambling code information of utilizing main synchronous detecting to obtain is carried out descrambling to sequence of parity respectively.

Step 230, the 280th is to through P _SSequence after the rearrangement is carried out the hadamard Fast transforms, and is as follows for 32 rank bipolarity hadamard Fast transforms processes: H=G ⁵, wherein G is 32 rank sparse matrixes, and its expression-form has multiple, and the present invention uses following form when carrying out emulation:

There are two 1 in the every row of preceding 16 row, and the same position of 16 row, back exists 1 and-1, and all the other elements all are 0.Each like this and vector product is converted into the poor of two element sums or two elements, and promptly this Fast transforms process is converted into the interative computation of 5 sub-additions and subtraction.

In step 210～250, get 31 odd number sequence first places and mend 0 and become 32 sequences to be measured, utilize P _SThe odd number sequence of mending after 0 is reset, the sequence after resetting is carried out the hadamard Fast transforms, the sequence P after the conversion _LReset, from 32 bit sequences that obtain, filter out the call number m of maximum related value correspondence ₀Wherein in step 210, obtain 31 odd number sequences that need to detect, and this sequence is mended 0 obtain 32 sequence to be detected.Because Metzler matrix obtains after conversion: M=P _LHP _S, transformation matrix P wherein _LWith P _SAnd hadamard matrix all is 32 * 32 matrix, sequence figure place to be detected is become 32 ability realize Fast transforms, therefore to mend this be necessary, so 31 strange/even number sequences to be detected should be carried out corresponding dimension expansion, way among the present invention is: at arbitrary value of benefit foremost of strange/even number sequences of 31 that separate from the S-SCH signal, because cover is but does not influence calculating simultaneously.So select in the emulation to mend 0.Obtain 32 sequence to be detected.

Utilize transformation results

With the testing process of signal by with the process that changes " rearrangement-hadamard Fast transforms-rearrangement " into of the equivalence of directly multiplying each other of Metzler matrix:

Wherein, Metzler matrix is converted into hadamard matrix and the process of resetting the battle array product as shown in Figure 1.

Step 220 is that the matrix that back obtains is reset.In step 250,, obtain the call number m of maximum correspondence from resetting the maximum that the back selects the sequence ₀

In step 260～290, get 31 digit pair number sequences row and mend 0 and become 32 sequences to be measured, utilize m ₀The even number sequence descrambling of information after to cover.Utilize P _SSequence behind the descrambling is reset, the sequence after resetting is carried out the hadamard Fast transforms, the sequence P after the conversion _LReset, this Fast transforms substitute related operation way and principle and previous step suddenly in way and principle to the odd number matrix identical.From 32 bit sequences that obtain, according to m ₀With m ₁Corresponding relation, the m that obtains in the integrating step 250 ₀, select and m the ultimate sequence after resetting ₀Possible corresponding whole correlations filter out maximum related value in these correlations, the call number of maximum related value correspondence is m ₁

Realization detects m from the even number sequence ₁, to m ₁Detection can utilize and detect m ₀Rearrangement and the fast computing of identical fast algorithm implementation to mending the even number sequence after 0, the complexity of this algorithm is 5 * 32 sub-additions equally, from 32 bit sequences that obtain, according to the m of 3GPP TS 36.211 V8.5.0.2008 agreement defineds ₀With m ₁Corresponding relation, the m that obtains in conjunction with previous step ₀, select and m the ultimate sequence after resetting ₀Possible corresponding whole correlations filter out maximum related value in these correlations, then the call number of this correlation correspondence is m ₁

At m ₁Detection is chosen in the process, adopts fast algorithm of the present invention can play the effect of simplifying well.Utilize fast algorithm that the even number sequence of mending after 0 is carried out Fast transforms earlier, from possible correlation serial number range, select maximum related value according to the correlation corresponding relation again, and obtain corresponding call number m ₁Detect m ₁With detection m ₀Difference be in the step 300 about corresponding m ₁Choosing of value will obtain by the discussion of correlation.The corresponding relation of correlation sequence number and m ₀And m ₁Corresponding relation be consistent.M wherein ₀With m ₁Concrete corresponding relation is as follows:

Work as m ₀During ∈ [0,2], m ₁∈ [0, m ₀+ 7], and m ₁≠ m ₀

Work as m ₀During ∈ [3,7], m ₁∈ [0, m ₀+ 6], and m ₁≠ m ₀

Work as m ₀During ∈ [8,9], m ₁∈ [m ₀-7, m ₀+ 6], and m ₁≠ m ₀

Work as m ₀During ∈ [10,24], m ₁∈ [m ₀-6, m ₀+ 6], and m ₁≠ m ₀

Work as m ₀During ∈ [25,30], m ₁∈ [m ₀-6,30], and m ₁≠ m ₀

Step 310: according to m ₀And m ₁Magnitude relationship, determine that this subframe is subframe 0 or subframe 5, concrete decision method is: if m ₀＞m ₁Be subframe 5; If m ₀＜m ₁Be subframe 0.And by detected m ₀And m ₁Obtain corresponding district sign group number.

Claims (5)

1. the LTE downlink subsidiary synchronizing channel detection method of a low complex degree is characterized in that concrete steps are as follows:

Step 1: according to " 3rd Generation Partnership Project (3GPP) Technical Specification GroupRadio Access network; Evolved Universal Terrestrial Radio Access (E-UTRA); Physicalchannels and modulation, Release 8,3GPP TS 36.211 V8.5.0.2008 " form of the local correlation matrix of agreement regulation, obtain P _LAnd P _SThe concrete form of battle array, and the form of calculation of equal value of the correlation matrix after the acquisition augmentation:

For subsequent detection is prepared;

Be 32 * 32 matrix, be circular matrix.It is lastrow of local correlation matrix of " 3GPP TS 36.211 V8.5.0.2008 " agreement defined and the augmented matrix behind the first from left row cover; H is a hadamard matrix; P _L, P _SBe the ranks permutation matrixes.

Step 2: auxiliary synchronous signals sequence is to received signal carried out the parity bit separation, obtains 31 odd number sequences and 31 digit pair number sequences row; The scrambling code information of utilizing main synchronous detecting acquisition is to sequence of parity difference descrambling;

Step 3: odd column is carried out coherent detection, get first of 31 odd number sequence and mend 0 and become 32 sequences to be measured, utilize P _SThe odd number sequence of mending after 0 is reset, the sequence after resetting is carried out the hadamard Fast transforms, the sequence P after the conversion _LReset, from 32 bit sequences that obtain, filter out the index m of maximum related value correspondence ₀

Step 4: use m ₀The further descrambling of relevant scrambler dual numbers sequence that forms, the antithesis ordered series of numbers carries out coherent detection again, gets 31 digit pair number sequences and is listed in first and mends 0 and become 32 sequences to be measured, utilizes P _SThe even number sequence of mending after 0 is reset, the sequence after resetting is carried out the hadamard Fast transforms, the sequence P after the conversion _LReset, from 32 bit sequences that obtain, according to m ₀With m ₁Corresponding relation, the m that obtains in conjunction with previous step ₀, dwindle and detect m ₁Required sequence number range is selected maximum related value corresponding identification m in the scope after dwindle ₁

Step 5: according to m ₀And m ₁Magnitude relationship, determine that this subframe is subframe 0 or subframe 5, obtain corresponding district sign group number.

2. method according to claim 1 is characterized in that: in the step 3, the hadamard matrix of 32 dimensions is finished by the interative computation of 5 sub-additions and subtraction with the multiplying of vector.

3. method according to claim 1 is characterized in that: in the step 4, detect m from the even number sequence ₁Step be, to m ₁Detection at first utilize and detect m ₀Rearrangement and the fast computing of identical fast algorithm implementation to mending the even number sequence after 0 is again according to corresponding m ₀, m ₁Corresponding relation from less than filtering out m 32 the probable value ₁

4. method according to claim 1 is characterized in that: in the step 2, setting the contained call number of odd-numbered bit sequence is m ₀, the contained call number of even bit is m ₁, odd number sequence and 31 local standard secondary synchronization sequences are carried out relevant, the parameter m that comprises in the odd number sequence is proposed ₀Information, to be detected finishing again according to m ₀, m ₁Size redefine subframe numbers, and when current subframe is judged to be 5 work song frames with m ₀, m ₁Value exchange, make it satisfied " 3rd Generation PartnershipProject (3GPP) Technical Specification Group Radio Access network; Evolved UniversalTerrestrial Radio Access (E-UTRA); Physical channels and modulation, Release 8,3GPP TS36.211 V8.5.0.2008 " prescribed form of agreement.

5. method according to claim 1 is characterized in that: in the step 5, described according to m ₀And m ₁Magnitude relationship, determine that this subframe is subframe 0 or subframe 5, wherein corresponding relation is:

If m ₀＜m ₁, then current subframe is 5 work song frames,

If m ₀＞m ₁, then current subframe is 5 work song frames, exchange m ₀And m ₁Value.