CN102790738B - Frequency offset correcting method and device - Google Patents
Frequency offset correcting method and device Download PDFInfo
- Publication number
- CN102790738B CN102790738B CN201110129370.1A CN201110129370A CN102790738B CN 102790738 B CN102790738 B CN 102790738B CN 201110129370 A CN201110129370 A CN 201110129370A CN 102790738 B CN102790738 B CN 102790738B
- Authority
- CN
- China
- Prior art keywords
- matrix
- meansigma methodss
- construction
- column vector
- normalization
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2647—Arrangements specific to the receiver only
- H04L27/2655—Synchronisation arrangements
- H04L27/2657—Carrier synchronisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2647—Arrangements specific to the receiver only
- H04L27/2655—Synchronisation arrangements
- H04L27/2668—Details of algorithms
- H04L27/2673—Details of algorithms characterised by synchronisation parameters
- H04L27/2675—Pilot or known symbols
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L5/00—Arrangements affording multiple use of the transmission path
- H04L5/0001—Arrangements for dividing the transmission path
- H04L5/0014—Three-dimensional division
- H04L5/0023—Time-frequency-space
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L5/00—Arrangements affording multiple use of the transmission path
- H04L5/003—Arrangements for allocating sub-channels of the transmission path
- H04L5/0048—Allocation of pilot signals, i.e. of signals known to the receiver
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a frequency offset correcting method and device. The method comprises the steps of constructing a corresponding matrix through responses of all pilot sub-carriers of each Tilestrip in each sub-channel, carrying out normalizing treatment on the constructed matrix, determining the mean value of the normalized matrix of each antenna so as to determine the mean value of the normalized matrix of each receiving antenna of a base station; carrying out fast Fourier transform according to the mean value of the normalized matrix of the base station so as to carry out frequency offset estimation, and carrying out frequency offset correcting according to a frequency offset estimation result, thereby achieving frequency offset estimation and correction in the OFDM system.
Description
Technical field
The present invention relates to wireless communication technology field, more particularly, to a kind of method and device of frequency offset correction.
Background technology
In OFDM (Orthogonal Frequency Division Multiplexing, OFDM) technology,
Requirement subcarrier spectrum is overlapped, and therefore the orthogonality between sub-carrier proposes and compares strict requirements.Once subcarrier
Between orthogonality can not strictly meet, then will produce interference between each subcarrier, thus deteriorating the performance of system.Therefore right
In ofdm system and multiple-input and multiple-output (Multiple-Input Multiple-Out-put, MIMO)-ofdm system, receive
The stability of the synchronicity of carrier frequency of making a start and carrier phase proposes higher requirement.
For the frequency departure between transmitter and receiver local oscillator (LO), referred to as carrier wave frequency deviation
(CFO), under mobile channel environment, Doppler frequency shift is also one of CFO source.CFO not only causes to demodulate the phase place rotation of constellation point
Turn thus affecting symbol judgement, and also the orthogonality between ofdm system subcarrier can be made to be destroyed, thus leading to son
The degradation interfering (ICI), causing systematic function of intercarrier.In addition, the verified less CFO of getting up early will lead
Cause the very big performance loss of OFDM, and also show in additive white Gaussian noise (Additive White Gaussion
Noise, AWGN) under channel more than the 4% of subcarrier spacing, be more than the 2% of subcarrier spacing CFO under a multipath fading channel
The loss that the performance of system is caused just cannot have ignored it is necessary to take CFO to estimate and indemnifying measure.MIMO-OFDM system pair
CFO is similarly very sensitive.
Due to ofdm system and MIMO-OFDM system to the sensitivity of CFO so that actual system design must take CFO
Estimate to ensure systematic function with corrective action.
Content of the invention
In view of this, the embodiment of the present invention provides a kind of method and device of frequency offset correction, in order to solve to ofdm system
And the carrier wave frequency deviation in MIMO-OFDM system estimated and corrected..
Embodiments provide a kind of method of frequency offset correction, including：
For the data receiving on an antenna of base station, according in every sub-channels each Tilestrip all
The response of pilot sub-carrier, obtains the meansigma methodss of the response of different pilot sub-carriers, according to the different pilot sub-carriers obtaining
The meansigma methodss construction column vector of response, and the column vector structural matrix according to construction, carry out scale process to the matrix of construction,
According to the matrix after scale process, and the mark of the matrix after this process is normalized to the matrix after this process；
According to the matrix after each Tilestrip normalized, determine the average of the matrix after every antenna normalization
Value, and the meansigma methodss according to the matrix after the every antenna normalization determining, determine the average of the matrix after this base station normalization
Value；
The element in meansigma methodss according to the matrix after the normalization of base station, carries out fast Fourier transform, and according to quick
Result after Fourier transformation carries out frequency offset correction.
Embodiments provide a kind of device of frequency offset correction, described device includes：
First normalized module, for the data receiving on an antenna for base station, believes according to every height
The response of all pilot sub-carriers of each Tilestrip in road, obtains the meansigma methodss of the response of different pilot sub-carriers, according to
The meansigma methodss construction column vector of the response of different pilot sub-carriers obtaining, and the column vector structural matrix according to construction, to structure
The matrix made carries out scale process, the matrix after being processed according to scale, and the mark of this matrix is normalized to this matrix
Process；
Second normalized module, for according to the matrix after each Tilestrip normalized, determining every sky
The meansigma methodss of the matrix after line normalization, and the meansigma methodss according to the matrix after the every antenna normalization determining, determine this base
Stand the meansigma methodss of the matrix after normalization；
Frequency offset correction module, for the element in the meansigma methodss according to the matrix after the normalization of base station, is carried out in quick Fu
Leaf transformation, and frequency offset correction is carried out according to the result after fast Fourier transform.
Embodiments provide a kind of method and device of frequency offset correction, the method passes through each in every sub-channels
The response of all pilot sub-carriers of Tilestrip, constructs corresponding matrix, and the matrix of construction is normalized,
Determine the meansigma methodss of the matrix after every antenna normalization, so that it is determined that the matrix after the reception antenna normalization of this base station every
Meansigma methodss, according to the meansigma methodss of the matrix determining after the normalization of base station, carry out fast Fourier transform, thus carry out frequency deviation estimating
Meter, the result according to offset estimation carries out frequency offset correction.It is achieved thereby that the estimation to ofdm system frequency deviation and correction.
Brief description
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the present invention, this
Bright schematic description and description is used for explaining the present invention, does not constitute inappropriate limitation of the present invention.In the accompanying drawings：
Fig. 1 is the process of frequency offset correction provided in an embodiment of the present invention；
Fig. 2 is the basic resources structural representation of uplink data channels in IEEE802.16e system；
Fig. 3 is the apparatus structure schematic diagram of frequency offset correction provided in an embodiment of the present invention.
Specific embodiment
In order that the technical problem to be solved, technical scheme and beneficial effect are clearer, clear, below tie
Close drawings and Examples, the present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only
In order to explain the present invention, it is not intended to limit the present invention.
Fig. 1 is the process of frequency offset correction provided in an embodiment of the present invention, and this process comprises the following steps：
S101：For the data receiving on an antenna of base station, according to each Tilestrip in every sub-channels
The response of all pilot sub-carriers, obtains the meansigma methodss of the response of different pilot sub-carriers.
Wherein, Tilestrip is the set of Tile in same frequency resources for the user, is one 2 dimension concept.Right
Take the carrier wave mapping mode of (PUSC) in subchannel section, it comprises (4*3*SlotDurNum) individual subcarrier.
S102：The meansigma methodss construction column vector of the response according to the different pilot sub-carriers obtaining, and the row according to construction
Vectorial structural matrix, is calibrated (scale) process to the matrix of construction.
Calibration scale process, one group of data will zoom in or out 2aTimes, to adapt to its memory range, wherein a is
Scale value, this 2aTimes, it is equivalent to move a position, a is just to represent to move to left, and negative indication moves to right,
The meansigma methodss construction column vector of the response according to the different pilot sub-carriers obtaining, including：
According to the numbering of OFDM symbol in time domain for the pilot sub-carrier, determine that different pilot sub-carrier meansigma methodss are corresponding
Serial number；
After getting the response meansigma methodss of different pilot sub-carriers, the response meansigma methodss of acquisition are divided into two groups, wherein
The meansigma methodss of the pilot sub-carrier response of Serial No. odd number are one group, the response of the pilot sub-carrier of Serial No. even number average
It is worth for one group, according to each set constructor column vector.
Included according to the column vector structural matrix of construction：
Column vector Z according to construction1, Z2, construct two matrixes, R1And R2, wherein,
Wherein zi=(y1, i+y4, i)/2, y1, iAnd y4, iResponse for different pilot sub-carriers；
Determine two matrix R of construction1And R2Sum, using this and as column vector construction matrix.
S103：According to the matrix after scale process, and the mark of the matrix after this process is carried out to the matrix after this process
Normalized.
Specifically in embodiments of the present invention, according to the matrix after scale process, and the mark of this matrix enters to this matrix
Row normalized, including：
Determine the sum of the real part of the diagonal entry of matrix after scale process, the mark using this with as matrix；
Determine the business of the mark of this matrix and this matrix, using this business as to the result after matrix normalization.
S104：According to the matrix after each Tilestrip normalized, determine the matrix after every antenna normalization
Meansigma methodss, and the meansigma methodss according to the matrix after the every antenna normalization determining, determine the matrix after this base station normalization
Meansigma methodss.
S105：The element in meansigma methodss according to the matrix after the normalization of base station, carries out fast Fourier transform, and according to
Result after fast Fourier transform carries out frequency offset correction.
Before carrying out frequency offset correction according to the result after fast Fourier transform, methods described also includes：
The column vector of the element construction in the meansigma methodss according to the matrix after normalization, and to this column vector fast Fourier
After conversion, the column vector of column vector real part construction, determines the position of maximum in the column vector of this real part construction；
According to the position of this maximum determining, carry out offset estimation, determine normalization frequency deviation.
Because the embodiment of the present invention is by the response of all pilot sub-carriers of each Tilestrip in every sub-channels,
Construct corresponding matrix, and the matrix of construction is normalized, determine the average of the matrix after every antenna normalization
Value, so that it is determined that the meansigma methodss of the matrix after the reception antenna normalization of this base station every, according to the square determining after the normalization of base station
The meansigma methodss of battle array, carry out fast Fourier transform, thus carrying out offset estimation, the result according to offset estimation carries out frequency deviation school
Just.It is achieved thereby that the estimation to ofdm system frequency deviation and correction.
Fig. 2 is the basic resources structural representation of uplink data channels in IEEE802.16e system, transverse axis in this Fig. 2
For time-domain OFDM symbol, the longitudinal axis is subcarrier in frequency domain, and including N number of physical resource unit, wherein background represents for shaded box and leads
Frequency carrier wave, other grids represent data carrier.The offset estimation that the embodiment of the present invention proposes and the scheme of correction are namely based on
This structure is realized.
Due to including 6 Tilestrip in sub-channels (Subchannel), and include on base station antenna
Many sub-channels.In embodiments of the present invention, first against the data receiving on a reception antenna of base station, therefrom select
Go out a Tilestrip, take out the response of all pilot sub-carriers in this Tilestrip, for example respectively y1,1, y1,2…y1,2N
And y4,1, y4,2…y4,2N, wherein yM, nMiddle m is m-th subcarrier on frequency domain, and n is n-th OFDM symbol in time domain.
For identical time-domain OFDM symbol, the response to different pilot sub-carriers is averaged.Wherein according to above-mentioned
Fig. 2 understands, this asks the process of response meansigma methodss to be to be added divided by 2 by the response of upper and lower two pilot sub-carriers, i.e. z1=
(y1,1+y4,1)/2, z2=(y1,2+y4,2)/2......z2N=(y1,10+y4,2N)/2.Specifically, different pilot sub-carriers are being obtained
Response meansigma methodss when, during fixed point is realized, the response of different pilot sub-carriers obtaining need to be moved to right 1, by the right side
The response of the pilot sub-carrier after shifting is added, and carries out except 2 afterwards, the as meansigma methodss of response.
When determining the meansigma methodss of response of different pilot sub-carriers, according to OFDM symbol in time domain for the pilot sub-carrier
Numbering, determine the corresponding serial number of meansigma methodss of the response of different pilot sub-carriers.When getting different pilot sub-carriers
After the meansigma methodss of response, the meansigma methodss of the response of acquisition are divided into two groups, wherein the sound of the pilot sub-carrier of Serial No. odd number
The meansigma methodss answered are one group, and the meansigma methodss of the response of the pilot sub-carrier of Serial No. even number are one group.
I.e. due to z1=(y1,1+y4,1)/2, z2=(y1,2+y4,2)/2......z2N=(y1,10+y4,2N)/2, are therefore inciting somebody to action
z1, z2...z2NWhen being divided into two groups, first group includes z1, z3... ..., z2N-1, second group includes z2, z4... ..., z2N.According to packet
The meansigma methodss of the response of different pilot sub-carriers comprising in each group afterwards, construct column vector.The column vector of wherein construction is divided
Wei not Z1=[z1, z3... ..., z2N-1]TAnd Z2=[z2, z4... ..., z2N]T, and two column vectors Z according to construction1, Z2Construction
Two matrix R1And R2, wherein,
Determine the sum of two matrixes of construction, that is, seek R1And R2Sum, using it with as R, R=R1+R2.Determining construction
Two matrixes and before, need matrix R1And R2In each element move to right 1, two matrixes after 1 will be moved to right
With as matrix R.
After determining matrix R, need to carry out scale process to matrix R, wherein matrix R is carried out with the mistake of scale process
Journey includes：Determine the element M of maximum absolute value in the real part of all elements and imaginary part in this matrix, wherein, M is the number more than 0
Value；Determine digit S1 that element M at most moves to left with respect to 32bit signed number；According to digit S1 determining, by this matrix R
The real part of all elements and empty step all move to left S1 position accordingly；Intercept the real part of all elements and imaginary part in this matrix after moving to left
High 16bit as scale process after matrix R '.After this matrix R has been carried out with scale process, the matrix obtaining can be used
R ' represents.
Matrix after needing after having obtained matrix R ' this scale is processed is normalized, wherein real in the present invention
The process applying this normalized in example includes：Matrix R ' after being processed according to this scale, and the mark of this R ' matrix, to this
Matrix R ' after scale is processed is normalized.
Wherein, according to the matrix R ' after this scale process, and the mark of this R ' matrix, the matrix after this scale is processed
R ' is normalized, including：Determine the sum of the real part of the diagonal entry of matrix R ' after scale process, should and make
Mark for matrix R '；Determine the business of the mark of this matrix R ' and this matrix R ', using this business as to the knot after matrix R ' normalization
ReallyI.e.Trace (R ') is the mark of matrix R '.
Specifically when realizing the normalization process of this matrix R ' by computer, can be diagonal according to matrix R ' first
The sum of the real part of line element, determines mark trace (R ') of this matrix R ', and calculating trace (R ') can with respect to 32bit signed number
With digit S2 at most moving to left, trace (R ') is moved to left (S2-15) position, and record scale (TileIdx)=- S2.Call 32
Position unsigned number division function, with 0x3fffffff divided by trace (R '), records result W, now the scope of W be 0x3fff～
0x7fff.The all elements of matrix R ' are all multiplied by W, obtain the matrix after normalization
It is the process carrying out for a Tilestrip in subchannel in above process, due to every sub-channels bag
Include 6 Tilestrip, processed according to said process for each Tilestrip, obtain and each Tilestrip pair
Matrix after the normalization answeredAnd to should matrix scale (TileIdx) value, for every sub-channels to this son
The matrix after the corresponding normalization of all Tilestrip in channel is averaged.
Wherein, for every sub-channels to the Matrix Calculating after the corresponding normalization of all Tilestrip in this subchannel
Meansigma methodss, including：
According in this subchannel, each Tilestrip is corresponding be normalized during preserve scale (TileIdx)
Value, search the minima of this corresponding scale of 6 Tilestrip (TileIdx), this minima be designated as scaleMin；
The corresponding Tilestrip of minima of this scale (TileIdx) be normalized after matrixDo not processed,
But the matrix after the corresponding normalization to other TilestripIn the real part of all elements and imaginary part, move to right corresponding position
Number, the digit wherein moving to right is the difference of the scale (TileIdx) and this scaleMin of this other Tilestrip, after moving to right
Matrix as this other Tilestrip corresponding unitized after matrix；
According to each Tilestrip corresponding unitized after matrix, determine this subchannel unitize after matrix flat
AverageAnd using this scaleMin as this subchannel scale (channelIdx), that is,RiFor each
Tilestrip corresponding unitized after matrix, the matrix after the normalization of the corresponding Tilestrip of this scaleMin is should
Tilestrip corresponding unitized after matrix.
It is directed to every sub-channels in said process, square is obtained according to the corresponding normalization of each Tilestrip in this subchannel
Battle array is it is determined that the matrix after the corresponding normalization of all Tilestrip in this subchannel is averaged.And wrap in every antenna
Include many sub-channels, be averaging according to the matrix after the corresponding normalization of all Tilestrip in the every sub-channels determining
Value, determines the meansigma methodss of the matrix after every antenna normalization.Wherein it is determined that the meansigma methodss of the matrix after every antenna normalization
Process, similar with the process of the meansigma methodss of the matrix after above-mentioned determination every sub-channels normalization, that is, be directed to every sub-channels
Corresponding scale (channelIdx), searches the minima of scale (channelIdx), for this scale (channelIdx)
Minima, determine the minima that each other subchannel is relative to this scale (channelIdx) unitized after matrix, pin
Corresponding to every sub-channels unitized after matrix, determine the meansigma methodss of the matrix after this antenna normalizationAnd should
The minima of scale (channelIdx) is as the scale (AntIdx) of this antenna.
After the meansigma methodss of matrix after determining every antenna normalization, according to the matrix after the normalization of every antenna
Meansigma methodss, determine the meansigma methodss of the matrix after this base station normalizationMatrix wherein after determining this base station normalization
During meansigma methodss, according to the meansigma methodss of the matrix after every antenna normalization, and it is directed to this antenna that every antenna preserves
Scale (AntIdx), determines the meansigma methodss of the matrix after this base station normalization, and return by every sub-channels with above-mentioned determination for detailed process
The process of the meansigma methodss of matrix after the meansigma methodss of the matrix after one change, and every antenna normalization of determination is similar to, here just not
Repeat one by one.
Concrete in embodiments of the present invention, when for every antenna, when determining the meansigma methodss of matrix after the normalization of base station,
Due to fixed-point number carry out the n times power that divisor is non-2 division relatively difficult, be thereby converted into being multiplied by it reciprocal and it is reciprocal
As table storage.Because base station could support up 8 antennas, form reciprocal is defined as：Q [8]=0x7fff, 0x3fff,
0x2aaa, 0x1fff, 0x1999,0x1555,0x1249,0x0fff }.
The meansigma methodss of matrix after determining base station normalizationAfterwards, the meansigma methodss according to the matrix after the normalization of base station
In element, carry out fast Fourier transform, and frequency offset correction carried out according to the result after fast Fourier transform.
Element in the meansigma methodss according to the matrix after the normalization of base station, when carrying out fast Fourier transform, structure first
Make column vector P, wherein this column vector is the column vector of 512*1, to the different element assignment in column vector, the wherein result of assignment
For：
P (0)=0
P (k)=0 (k >=N)
WhereinRepresentXth row in matrix, y arranges, according to above-mentioned assigned result, by non-assignment in this column vector P
Element be entered as 0.Column vector P after assignment is carried out with fast Fourier transform (FFT), and to the column vector P after FFT computing
Carry out the computing for the treatment of excess syndrome portion, i.e. Thelta=real (fft (P)), wherein, Thelta is the column vector of 512*1, afterwards by Thelta
128-383 element set to 0.
After determining column vector Thelta, determine the corresponding position of maximum in this column vector Thelta, that is, determine
[max, Idx]=max (Thelta) determine column vector Thelta in maximum corresponding position Idx when, can according under
State algorithm to realize：
After determining the corresponding position of the maximum in column vector Thelta, determine normalization frequency deviation, wherein this normalizing
Change frequency deviation Dq.Wherein, the calculating process of this normalization frequency deviation Dq includes：Δ θ=2 π (1/512) (Idx-1)/3
After determining normalization frequency deviation, you can all carrier waves to active user's scheduling, carry out phase place to correct
Frequency deviation, i.e. Si=Pi·e-j·Δθ·(i-1), wherein i represents i-th OFDM symbol, and Dq is normalization frequency deviation, PiBefore frequency offset correction
The value of carrier wave, Si is the value after correction.
And in fixed point is realized, carrying out power operation is the relatively difficult fortune that can avoid this complexity with look-up table
Calculate.Table-form making process is：Because the value of Idx is -128～127.And up could support up 21 symbols, wherein symbol 0 not
Need to do frequency offset correction, then the size of form is 256x20=5120B.
For symbol i：Calculate
I=cos ((2*3.14*Idx/3/512) * i) * 0x7fff；
Q=-sin ((2*3.14*Idx/3/512) * i) * 0x7fff；
Wherein Idx is -128～127.
And using I and Q as the high 16bit of form and low 16bit storage.
When carrying out frequency offset correction, only corresponding frequency offset correction value need to be found by Idx, then according to each subcarrier of symbol
It is multiplied with it one by one.
Fig. 3 is a kind of structural representation of the device of frequency offset correction provided in an embodiment of the present invention, and described device includes：
First normalized module 31, for the data receiving on an antenna for base station, according to every height
The response of all pilot sub-carriers of each Tilestrip in channel, obtains the meansigma methodss of the response of different pilot sub-carriers, root
According to the meansigma methodss construction column vector of the response of the different pilot sub-carriers obtaining, and the column vector structural matrix according to construction, right
The matrix of construction carries out scale process, the matrix after being processed according to scale, and the mark of this matrix carries out normalizing to this matrix
Change is processed；
Second normalized module 32, for according to the matrix after each Tilestrip normalized, determining every
The meansigma methodss of the matrix after antenna normalization, and the meansigma methodss according to the matrix after the every antenna normalization determining, determining should
The meansigma methodss of the matrix after the normalization of base station；
Frequency offset correction module 33, for the element in the meansigma methodss according to the matrix after the normalization of base station, carries out quick Fu
In leaf transformation, and frequency offset correction is carried out according to the result after fast Fourier transform.
Described first normalized module 31, specifically for the volume of the OFDM symbol in time domain according to pilot sub-carrier
Number, determine the different corresponding serial numbers of pilot sub-carrier meansigma methodss；After getting the response meansigma methodss of different pilot sub-carriers,
The response meansigma methodss of acquisition are divided into two groups, the meansigma methodss of the wherein pilot sub-carrier response of Serial No. odd number are one group, sequence
Row number is that the meansigma methodss of the pilot sub-carrier response of even number are one group, according to each set constructor column vector.
Described first normalized module 31, specifically for column vector Z according to construction1, Z2, construct two matrixes, R1
And R2, wherein,
Wherein zi=(y1, i+y4, i)/2, y1, iAnd y4, iResponse for different pilot sub-carriers；Determine two matrix R of construction1And R2's
With the matrix using this with as column vector construction.
Described first normalized module 31, specifically for determining in the real part of all elements and imaginary part in this matrix absolutely
To the element M that value is maximum, wherein M is the numerical value more than zero；Determine the position that element M at most moves to left with respect to 32bit signed number
Number S1；According to digit S1 determining, the real part of all elements in this matrix and empty step are all moved to left S1 position accordingly；Intercepting moves to left
In this matrix afterwards, as the matrix after scale process, wherein M is more than zero for the real part of all elements and the high 16bit of imaginary part
Numerical value.
Described first normalized module 31, the diagonal entry of the matrix after processing specifically for determination scale
The sum of real part, the mark using this with as matrix；Determine the business of the mark of this matrix and this matrix, using this business as to matrix normalizing
Result after change.
During the meansigma methodss of specific matrix after determining this base station normalization, because every sub-channels include 6
Tilestrip, is processed according to said process for each Tilestrip, is obtained return corresponding with each Tilestrip
Matrix after one changeAnd to should matrix scale (TileIdx) value, for every sub-channels in this subchannel
The corresponding normalization of all Tilestrip after matrix average.
Wherein, for every sub-channels to the Matrix Calculating after the corresponding normalization of all Tilestrip in this subchannel
Meansigma methodss, including：
According in this subchannel, each Tilestrip is corresponding be normalized during preserve scale (TileIdx)
Value, search the minima of this corresponding scale of 6 Tilestrip (TileIdx), this minima be designated as scaleMin；
The corresponding Tilestrip of minima of this scale (TileIdx) be normalized after matrixDo not processed,
But the matrix after the corresponding normalization to other TilestripIn the real part of all elements and imaginary part, move to right corresponding position
Number, the digit wherein moving to right is the difference of the scale (TileIdx) and this scaleMin of this other Tilestrip, after moving to right
Matrix as this other Tilestrip corresponding unitized after matrix；
According to each Tilestrip corresponding unitized after matrix, determine this subchannel unitize after matrix flat
AverageAnd using this scaleMin as this subchannel scale (channelIdx), that is,RiFor each
Tilestrip corresponding unitized after matrix, the matrix after the normalization of the corresponding Tilestrip of this scaleMin is should
Tilestrip corresponding unitized after matrix.
It is directed to every sub-channels in said process, square is obtained according to the corresponding normalization of each Tilestrip in this subchannel
Battle array is it is determined that the matrix after the corresponding normalization of all Tilestrip in this subchannel is averaged.And wrap in every antenna
Include many sub-channels, be averaging according to the matrix after the corresponding normalization of all Tilestrip in the every sub-channels determining
Value, determines the meansigma methodss of the matrix after every antenna normalization.Wherein it is determined that the meansigma methodss of the matrix after every antenna normalization
Process, similar with the process of the meansigma methodss of the matrix after above-mentioned determination every sub-channels normalization, that is, be directed to every sub-channels
Corresponding scale (channelIdx), searches the minima of scale (channelIdx), for this scale (channelIdx)
Minima, determine the minima that each other subchannel is relative to this scale (channelIdx) unitized after matrix, pin
Corresponding to every sub-channels unitized after matrix, determine the meansigma methodss of the matrix after this antenna normalizationAnd should
The minima of scale (channelIdx) is as the scale (AntIdx) of this antenna.
After the meansigma methodss of matrix after determining every antenna normalization, according to the matrix after the normalization of every antenna
Meansigma methodss, determine the meansigma methodss of the matrix after this base station normalizationMatrix wherein after determining this base station normalization
During meansigma methodss, according to the meansigma methodss of the matrix after every antenna normalization, and it is directed to this antenna that every antenna preserves
Scale (AntIdx), determines the meansigma methodss of the matrix after this base station normalization, and return by every sub-channels with above-mentioned determination for detailed process
The process of the meansigma methodss of matrix after the meansigma methodss of the matrix after one change, and every antenna normalization of determination is similar to, here just not
Repeat one by one.
Concrete in embodiments of the present invention, when for every antenna, when determining the meansigma methodss of matrix after the normalization of base station,
Due to fixed-point number carry out the n times power that divisor is non-2 division relatively difficult, be thereby converted into being multiplied by it reciprocal and it is reciprocal
As table storage.Because base station could support up 8 antennas, form reciprocal is defined as：Q [8]=0x7fff, 0x3fff,
0x2aaa, 0x1fff, 0x1999,0x1555,0x1249,0x0fff }.
Described device also includes：
Frequency deviation estimating modules 34, for the column vector of the element construction in the meansigma methodss according to the matrix after normalization, and
Column vector to column vector real part construction after this column vector fast Fourier transform, determines maximum in the column vector of this real part construction
The position of value；According to the position of this maximum determining, carry out offset estimation, determine normalization frequency deviation.
Described frequency offset correction module 33, specifically for according to Si=Pi·e-j·Δθ·(i-1)Carry out frequency offset correction, wherein i table
Show i-th OFDM symbol, Dq is normalization frequency deviation.
Embodiments provide a kind of method and device of frequency offset correction, the method passes through each in every sub-channels
The response of all pilot sub-carriers of Tilestrip, constructs corresponding matrix, and the matrix of construction is normalized,
Determine the meansigma methodss of the matrix after every antenna normalization, so that it is determined that the matrix after the reception antenna normalization of this base station every
Meansigma methodss, according to the meansigma methodss of the matrix determining after the normalization of base station, carry out fast Fourier transform, thus carry out frequency deviation estimating
Meter, the result according to offset estimation carries out frequency offset correction.It is achieved thereby that the estimation to ofdm system frequency deviation and correction.
Described above illustrate and describes the preferred embodiments of the present invention, but as previously mentioned it should be understood that the present invention not
It is confined to form disclosed herein, be not to be taken as the exclusion to other embodiment, and can be used for various other combinations, modification
And environment, and can be carried out by the technology or knowledge of above-mentioned teaching or association area in invention contemplated scope described herein
Change.And the change that those skilled in the art are carried out and change without departing from the spirit and scope of the present invention, then all should be in institute of the present invention
In attached scope of the claims.
Claims (15)
1. a kind of method of frequency offset correction is it is characterised in that include：
For the data receiving on an antenna of base station, according to all pilot tones of each Tilestrip in every sub-channels
The response of subcarrier, obtains the meansigma methodss of the response of different pilot sub-carriers, according to the response of the different pilot sub-carriers obtaining
Meansigma methodss construction column vector, and according to construction column vector structural matrix, to construction matrix carry out calibrate scale process,
According to the matrix after scale process, and the mark of the matrix after this process is normalized to the matrix after this process；
According to the matrix after each Tilestrip normalized, determine the meansigma methodss of the matrix after every antenna normalization, and
According to the meansigma methodss of the matrix after the every antenna normalization determining, determine the meansigma methodss of the matrix after this base station normalization；
The element in meansigma methodss according to the matrix after the normalization of base station, carries out fast Fourier transform, and according in quick Fu
Result after leaf transformation carries out frequency offset correction；
Wherein, described Tilestrip is the set of fragment Tile in same frequency resources for the user.
2. the method for claim 1 it is characterised in that according to obtain different pilot sub-carriers response meansigma methodss
Construction column vector, including：
According to the numbering of orthogonal frequency division multiplex OFDM symbol in time domain for the pilot sub-carrier, determine that different pilot sub-carriers are average
It is worth corresponding serial number；
After getting the response meansigma methodss of different pilot sub-carriers, the response meansigma methodss of acquisition are divided into two groups, wherein sequence
Number it is one group for the meansigma methodss of the pilot sub-carrier response of odd number, the meansigma methodss of the pilot sub-carrier of Serial No. even number response are
One group, according to each set constructor column vector.
3. method as claimed in claim 2 is it is characterised in that include according to the column vector structural matrix of construction：
Column vector Z according to construction1,Z2, construct two matrixes, R1And R2, wherein,
Wherein zi=(y1,i+y4,i)/2, y1,iAnd y4,iResponse for different pilot sub-carriers；
Determine two matrix R of construction1And R2Sum, using this and as column vector construction matrix；
Wherein, the meansigma methodss that Z responds for pilot sub-carrier, N is positive integer, and i is i-th OFDM symbol in time domain.
4. method as claimed in claim 3 it is characterised in that construction two matrix R1And R2And before, methods described is also
Including：
By two matrix R1And R2In each element move to right 1.
5. the method for claim 1 is it is characterised in that carry out scale process inclusion to the matrix of construction：
Determine the element M of maximum absolute value in the real part of all elements and imaginary part in this matrix；
Determine digit S1 that element M at most moves to left with respect to 32bit signed number；
According to digit S1 determining, the real part of all elements in this matrix and empty step are all moved to left S1 position accordingly；
Intercept in this matrix after moving to left the real part of all elements and the high 16bit of imaginary part as the matrix after scale process.
6. the method for claim 1 is it is characterised in that matrix after the described process according to scale, and this matrix
Mark is normalized to this matrix, including：
Determine the sum of the real part of the diagonal entry of matrix after scale process, the mark using this with as matrix；
Determine the business of the mark of this matrix and this matrix, using this business as to the result after matrix normalization.
7. the method for claim 1 is it is characterised in that described carry out frequency deviation according to the result after fast Fourier transform
Before correction, methods described also includes：
The column vector of the element construction in the meansigma methodss according to the matrix after normalization, and to this column vector fast Fourier transform
The column vector of column vector real part construction, determines the position of maximum in the column vector of this real part construction afterwards；
According to the position of this maximum determining, carry out offset estimation, determine normalization frequency deviation.
8. method as claimed in claim 7 is it is characterised in that carry out frequency offset correction according to the result after fast Fourier transform
Including：
According to Si=Pi·e-j·Δθ·(i-1)Carry out frequency offset correction, wherein i represents i-th orthogonal frequency division multiplex OFDM symbol, PiFor
The value of carrier wave, S before frequency offset correctioniFor the value after correction；
Wherein, the value of Δ θ=2 π (1/512) (Idx-1)/3, Idx is -128～127.
9. a kind of device of frequency offset correction is it is characterised in that described device includes：
First normalized module, for the data receiving on an antenna for base station, according in every sub-channels
The response of all pilot sub-carriers of each Tilestrip, obtains the meansigma methodss of the response of different pilot sub-carriers, according to acquisition
The response of different pilot sub-carriers meansigma methodss construction column vector, and according to construction column vector structural matrix, to construction
Matrix carries out calibrating scale process, the matrix after being processed according to scale, and the mark of this matrix is normalized to this matrix
Process；
Second normalized module, for according to the matrix after each Tilestrip normalized, determining that every antenna is returned
The meansigma methodss of the matrix after one change, and the meansigma methodss according to the matrix after the every antenna normalization determining, determine that this base station is returned
The meansigma methodss of the matrix after one change；
Frequency offset correction module, for the element in the meansigma methodss according to the matrix after the normalization of base station, carries out fast Fourier change
Change, and frequency offset correction is carried out according to the result after fast Fourier transform；
Wherein, described Tilestrip is the set of fragment Tile in same frequency resources for the user.
10. device as claimed in claim 9, it is characterised in that described first normalized module, is led specifically for basis
The numbering of orthogonal frequency division multiplex OFDM symbol in time domain for the frequency subcarrier, determines the different corresponding sequences of pilot sub-carrier meansigma methodss
Row number；After getting the response meansigma methodss of different pilot sub-carriers, the response meansigma methodss of acquisition are divided into two groups, wherein sequence
Number it is one group for the meansigma methodss of the pilot sub-carrier response of odd number, the meansigma methodss of the pilot sub-carrier of Serial No. even number response are
One group, according to each set constructor column vector.
11. devices as claimed in claim 10 it is characterised in that described first normalized module, specifically for basis
Column vector Z of construction1,Z2, construct two matrixes, R1And R2, wherein,
Wherein zi=(y1,i+y4,i)/2, y1,iAnd y4,iResponse for different pilot sub-carriers；Determine two matrix R of construction1With
R2Sum, using this and as column vector construction matrix；
Wherein, the meansigma methodss that Z responds for pilot sub-carrier, N is positive integer, and i is i-th OFDM symbol in time domain.
12. devices as claimed in claim 9, it is characterised in that described first normalized module, are somebody's turn to do specifically for determining
The element M of maximum absolute value in the real part of all elements and imaginary part in matrix, wherein M is the numerical value more than zero；Determine element M phase
Digit S1 that 32bit signed number is at most moved to left；According to digit S1 determining, by the real part of all elements in this matrix and
Empty step all moves to left S1 position accordingly；Intercept the real part of all elements and the high 16bit conduct of imaginary part in this matrix after moving to left
Matrix after scale process, wherein M is the numerical value more than zero.
13. devices as claimed in claim 9 it is characterised in that described first normalized module, specifically for determining
The sum of the real part of the diagonal entry of matrix after scale process, the mark using this with as matrix；Determine this matrix and this square
The business of the mark of battle array, using this business as to the result after matrix normalization.
14. devices as claimed in claim 9 are it is characterised in that described device also includes：
Frequency deviation estimating modules, for the column vector of the element construction in the meansigma methodss according to the matrix after normalization, and to this row
After vector fast Fourier transform, the column vector of column vector real part construction, determines the position of maximum in the column vector of this real part construction
Put；According to the position of this maximum determining, carry out offset estimation, determine normalization frequency deviation.
15. devices as claimed in claim 14 it is characterised in that described frequency offset correction module, specifically for according to Si=Pi·
e-j·Δθ·(i-1)Carry out frequency offset correction, wherein i represents i-th orthogonal frequency division multiplex OFDM symbol, Δ θ is normalization frequency deviation, PiFor
The value of carrier wave, S before frequency offset correctioniFor the value after correction；
Wherein, the value of Δ θ=2 π (1/512) (Idx-1)/3, Idx is -128～127.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110129370.1A CN102790738B (en) | 2011-05-18 | 2011-05-18 | Frequency offset correcting method and device |
| PCT/CN2011/081276 WO2012155459A1 (en) | 2011-05-18 | 2011-10-25 | Method and device for correcting frequency offset |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110129370.1A CN102790738B (en) | 2011-05-18 | 2011-05-18 | Frequency offset correcting method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102790738A CN102790738A (en) | 2012-11-21 |
| CN102790738B true CN102790738B (en) | 2017-02-08 |
Family
ID=47156042
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110129370.1A Active CN102790738B (en) | 2011-05-18 | 2011-05-18 | Frequency offset correcting method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102790738B (en) |
| WO (1) | WO2012155459A1 (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104639485B (en) * | 2013-11-12 | 2019-01-04 | 联咏科技股份有限公司 | Offset correction of carrier frequency method and offset correction of carrier frequency system |
| CN106059984A (en) * | 2015-12-10 | 2016-10-26 | 国网山东省电力公司烟台供电公司 | Digital phase-modulated signal carrier phase offset estimation method |
| CN107294678B (en) * | 2016-03-31 | 2020-10-02 | 上海诺基亚贝尔股份有限公司 | Method and communication device for channel estimation |
| US10627483B2 (en) * | 2016-07-09 | 2020-04-21 | Texas Instruments Incorporated | Methods and apparatus for velocity detection in MIMO radar including velocity ambiguity resolution |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101022442A (en) * | 2007-01-16 | 2007-08-22 | 西安交通大学 | Joint time synchronizing and frequency-offset estimating method in OFDM system |
| CN101505290A (en) * | 2009-03-17 | 2009-08-12 | 山东大学 | Improved frequency bias estimation method for wideband MIMO |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7197525B2 (en) * | 2002-11-26 | 2007-03-27 | Analog Devices, Inc. | Method and system for fixed point fast fourier transform with improved SNR |
-
2011
- 2011-05-18 CN CN201110129370.1A patent/CN102790738B/en active Active
- 2011-10-25 WO PCT/CN2011/081276 patent/WO2012155459A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101022442A (en) * | 2007-01-16 | 2007-08-22 | 西安交通大学 | Joint time synchronizing and frequency-offset estimating method in OFDM system |
| CN101505290A (en) * | 2009-03-17 | 2009-08-12 | 山东大学 | Improved frequency bias estimation method for wideband MIMO |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102790738A (en) | 2012-11-21 |
| WO2012155459A1 (en) | 2012-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9106280B2 (en) | Pilot design for massive MIMO communication | |
| CN102437987B (en) | The generation of channel state information reference signals sequence and mapping method and device | |
| US8964886B2 (en) | Method and device for transmitting a preamble in a wireless communication system | |
| CN102208972B (en) | Noise estimation for wireless communication | |
| CN103595664B (en) | Channel estimation methods and device in a kind of multiple receive antenna system | |
| CN101778068B (en) | Frequency domain position-confirming method and device of positioning reference signal | |
| CN102687441B (en) | Channel estimation for OFDM systems | |
| US20070217526A1 (en) | Transmission/reception apparatus and method for frequency domain signal processing in a smart antenna system | |
| EP2991236B1 (en) | Multiple-input multiple-output orthogonal frequency-division multiplexing communication system and signal compensation method | |
| CN102790738B (en) | Frequency offset correcting method and device | |
| US20130336282A1 (en) | Transmitter apparatus, receiver apparatus, communication system, communication method, and integrated circuit | |
| US20140023092A1 (en) | Cyclic shift delays in multi-user packets with resolvable very high throughput long training fields (vhtltfs) | |
| CN110537351A (en) | Measuring device and measurement method | |
| US20140269883A1 (en) | Non-linear time domain channel estimation in ofdm systems | |
| US10972236B2 (en) | Signal transmission method, signal reception method, apparatus, and system in wireless communication | |
| CN105490980A (en) | Carrier wave frequency deviation estimation method and system | |
| CN103873416B (en) | EVM (Error Vector Magnitude) phase estimating and compensating method | |
| CN106878229A (en) | IQ based on initial phase compensation is uneven to be estimated and compensation method and device | |
| US9143364B2 (en) | IQ imbalance estimation using broadcast signals | |
| CN102035768B (en) | Method and device for measuring carrier to interference and noise ratio | |
| CN101282198A (en) | Transmission method and terminal for upstream multi-antenna in TDD system | |
| US20190074947A1 (en) | Reference signal transmission method and apparatus | |
| US8982975B2 (en) | Method for signal estimation and compensation and apparatus using the same | |
| CN115812282A (en) | Determining signal quality using error vector magnitude | |
| CN102447658B (en) | Method and device for suppressing and merging interference |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |