CN105323199A - Uplink frequency offset estimation method - Google Patents

Abstract

The invention discloses an uplink frequency offset estimation method which comprises the following steps: extracting a cyclic prefix (CP) of each orthogonal frequency-division multiplexing (OFDM) symbol of an uplink subframe, and a data portion corresponding to the CP, wherein the CP portion is cpdata, and the data portion corresponding to the CP is taildata; carrying out zero padding on the extracted cpdata and taildata to M data points corresponding to the bandwidth, and then, carrying out fast Fourier transform on the M data points to obtain frequency domain data D1 and D2; extracting useful K points, cpdata_FD and taildata_FD respectively, in the frequency domain data; letting taildata_FD_neg= -taildata_FD; carrying out related operation on the taildata_FD_neg and cpdata_FD corresponding to user resources through the formula shown in the specification, wherein the N in the formula represents the number of subcarriers occupied by a user; and extracting phase of the correlation result R and estimating frequency offset of the user according to the phase.

Description

A kind of uplink frequency offset method of estimation

Technical field

The application relates to mobile communication technology field, particularly relates to a kind of uplink frequency offset method of estimation.

Background technology

When LTE upstream data sends, frequency needs skew 1/2 subcarrier, therefore need first to recover 1/2 subcarrier at receiving terminal to move, way common in prior art is: after upstream data receives, abandon CP data, carry out recovery 1/2 subcarrier to other data to move, in actual applications, it is realize by inquiring about the exponent arithmetic table pre-set that 1/2 subcarrier is moved.Now exponent arithmetic table only need store between phase region is the value of 0 ~ π.

In order to improve uplink frequency offset estimated accuracy, the Cyclic Prefix (CP) of OFDM (OFDM) symbol can be utilized to carry out uplink frequency offset estimation, and concrete steps as shown in Figure 1, comprising:

Step 101: extract the data division that CP and CP of each OFDM symbol of a sub-frame of uplink is corresponding, as shown in Figure 2, wherein CP part is the data division that cpdata, CP are corresponding is taildata.

Step 102: 1/2 sub-carrier offset is recovered to cpdata and taildata.

Step 103: suppose that bandwidth is 20M, to cpdata and the taildata zero padding of extracting to 2048 points, then carries out the FFT conversion of 2048, obtains frequency domain data D1 and D2.Extract 1200 points useful in frequency domain data, be respectively cpdata_FD and taildata_FD.

Step 104: cpdata_FD and taildata_FD corresponding to user resources carries out associative operation.

$R=\underset{m=1}{\overset{M}{\mathrm{\Σ}}}\mathrm{cpdata}\_\mathrm{FD}\left(m\right)*\mathrm{taildata}\_\mathrm{FD}{\left(m\right)}^{*}$

Wherein M is the subcarrier number of CU.

Step 105: get phase place to correlated results R, can be estimated the frequency deviation of user by phase place.

In the manner described above, then need to carry out recovery 1/2 subcarrier to CP data and move, correspondingly, can extend to negative value interval between the phase region of exponent arithmetic table, this just needs more memory space to store exponent arithmetic table, adds the complexity of macrooperation.

Summary of the invention

This application provides a kind of uplink frequency offset method of estimation, the complexity of frequency excursion algorithm can be reduced.

A kind of uplink frequency offset method of estimation that the embodiment of the present application provides, comprising:

The data division that A, the cyclic prefix CP extracting each orthogonal frequency division multiplex OFDM symbol of a sub-frame of uplink and CP are corresponding, wherein CP part is the data division that cpdata, CP are corresponding is taildata;

B, to cpdata and the taildata zero padding of extracting to M data point corresponding to bandwidth, then carry out the fast Fourier transform of M point, obtain frequency domain data D1 and D2; Extract K point useful in frequency domain data, be respectively cpdata_FD and taildata_FD;

C, make taildata_FD_neg=-taildata_FD;

D, cpdata_FD and taildata_FD_neg corresponding to user resources carry out associative operation:

$R=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_\mathrm{FD}\_\mathrm{neg}{\left(n\right)}^{*};$

Wherein N is the subcarrier number of CU;

E, phase place is got to correlated results R, gone out the frequency deviation of user by phase estimation.

Preferably, after described step B, perform following steps:

Cpdata_FD and taildata_FD corresponding to user resources carries out associative operation:

$R=-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_{\mathrm{FD}}^{*};$

Then step e is performed.

Preferably, after described step B, perform following steps:

Cpdata_FD and taildata_FD corresponding to user resources carries out associative operation:

$R=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_{\mathrm{FD}}^{*}$

Phase is got to correlated results R;

Make φ=φ+π, and judge φ value size, φ value is corrected as follows:

$\mathrm{\&phi;}=\left\{\begin{array}{cc}\mathrm{\&phi;}-2\mathrm{\&pi;}& \mathrm{\&phi;}>\mathrm{\&pi;}\\ \mathrm{\&phi;}& -\mathrm{\&pi;}<\mathrm{\&phi;}\&le;\mathrm{\&pi;}\\ \mathrm{\&phi;}+2\mathrm{\&pi;}& \mathrm{\&phi;}\&le;-\mathrm{\&pi;}\end{array}\right.;$

According to the frequency deviation of the phase estimation user after correction.

Preferably, if bandwidth is 20M, then M=2048, K=1200.

As can be seen from the above technical solutions, not needing the recovery it being carried out to 1/2 sub-carrier offset, but negative value is got to reach identical effect to related data, therefore without the need to storing exponent arithmetic table, reducing the complexity that uplink frequency offset is estimated.

Accompanying drawing explanation

Fig. 1 is uplink frequency offset method of estimation flow chart of the prior art;

Fig. 2 is the schematic diagram extracting OFDM symbol;

The uplink frequency offset method of estimation flow chart that Fig. 3 provides for the embodiment of the present application.

Embodiment

The technical conceive of the uplink frequency offset method of estimation that the application provides is: after extracting cpdata and taildata data, do not need the recovery it being carried out to 1/2 sub-carrier offset, can get negative value to reach identical effect to related data.Concrete principle is as follows:

$\mathrm{cpdata}{\left(n\right)}^{\&prime;}=\mathrm{cpdata}\left(n\right)*e^{\mathrm{j\&pi;}\frac{-\mathrm{Ncp}+n-1}{N}}---\left(1\right)$

$\begin{array}{c}\mathrm{taildata}{\left(n\right)}^{\&prime;}=\mathrm{taildata}\left(n\right)*e^{\mathrm{j\&pi;}\frac{N-\mathrm{Ncp}+n-1}{N}}\\ =\mathrm{taildata}\left(n\right)*e^{\mathrm{j\&pi;}\frac{-\mathrm{Ncp}+n-1}{N}}*e^{\mathrm{j\&pi;}}\\ =-\mathrm{taildata}\left(n\right)*e^{\mathrm{j\&pi;}\frac{-\mathrm{Ncp}+n-1}{N}}\end{array}---\left(2\right)$

Wherein, N is counting of IFFT, and Ncp is the length of OFDM symbol CP, n=1,2 ..., Ncp.Cpdata (n) and taildata (n) is the data of recovery 1/2 sub-carrier offset, and cpdata (n) ' and taildata (n) ' is the data not recovering 1/2 sub-carrier offset.

Suppose that channel is constant in an OFDM symbol, then have taildata (n)=cpdata (n), formula (2) can be written as:

$\begin{array}{c}\mathrm{taildata}{\left(n\right)}^{\&prime;}=-\mathrm{cpdata}\left(n\right)*e^{\mathrm{j\&pi;}\frac{-\mathrm{Ncp}+n-1}{N}}\\ =-\mathrm{cpdata}{\left(n\right)}^{\&prime;}\end{array}---\left(3\right)$

From formula (3), 1/2 sub-carrier offset can adopt the impact of above-mentioned frequency deviation estimating method and add negative sign to eliminate to data.

For making the know-why of technical scheme, feature and technique effect clearly, below in conjunction with specific embodiment, technical scheme is described in detail.

The uplink frequency offset method of estimation flow process that the embodiment of the present application provides as shown in Figure 3, comprises the steps:

Step 301: extract the data division that CP and CP of each OFDM symbol of a sub-frame of uplink is corresponding, as shown in Figure 2, wherein CP part is the data division that cpdata, CP are corresponding is taildata.

Step 302: suppose that bandwidth is 20M, to cpdata and the taildata zero padding of extracting to 2048 points, then carries out the FFT conversion of 2048, obtains frequency domain data D1 and D2.Extract 1200 points useful in frequency domain data, be respectively cpdata_FD and taildata_FD.

Step 303: make taildata_FD_neg=-taildata_FD

Step 304: cpdata_FD and taildata_FD_neg corresponding to user resources carries out associative operation.

$R=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_\mathrm{FD}\_\mathrm{neg}{\left(n\right)}^{*}$

Wherein N is the subcarrier number of CU.

Step 305: get phase place to correlated results R, is gone out the frequency deviation of user by phase estimation.

By getting negative value to replace the operation of " recovering 1/2 sub-carrier offset " to taildata_FD in this method.

According to another embodiment of the application, negative value can be got realize to cpdata_FD or R or other data that can reach same effect.Such as, after step 302, perform following process:

Step 303A: cpdata_FD and taildata_FD corresponding to user resources carries out associative operation: $R=-\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_{\mathrm{FD}}^{*}$

Then the operation of 305 is performed.

According to another embodiment of the application, negative value can not be got to any data, but when finally phase place being got to R, π be added to phase value, then phase value is limited in (-π, π] in interval.Detailed process is: after 302 steps, performs:

Step 303B: cpdata_FD and taildata_FD corresponding to user resources carries out associative operation:

$R=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_{\mathrm{FD}}^{*};$

Step 304B: phase is got to correlated results R;

Step 305B: make φ=φ+π;

Step 306B: judge φ value size, φ value is corrected:

$\mathrm{\&phi;}=\left\{\begin{array}{cc}\mathrm{\&phi;}-2\mathrm{\&pi;}& \mathrm{\&phi;}>\mathrm{\&pi;}\\ \mathrm{\&phi;}& -\mathrm{\&pi;}<\mathrm{\&phi;}\&le;\mathrm{\&pi;}\\ \mathrm{\&phi;}+2\mathrm{\&pi;}& \mathrm{\&phi;}\&le;-\mathrm{\&pi;}\end{array}\right.;$

According to the frequency deviation of the phase estimation user after correction.

The foregoing is only the preferred embodiment of the application; not in order to limit the protection range of the application; within all spirit in technical scheme and principle, any amendment made, equivalent replacements, improvement etc., all should be included within scope that the application protects.

Claims (4)

1. a uplink frequency offset method of estimation, is characterized in that, comprising:

The data division that A, the cyclic prefix CP extracting each orthogonal frequency division multiplex OFDM symbol of a sub-frame of uplink and CP are corresponding, wherein CP part is the data division that cpdata, CP are corresponding is taildata;

B, to cpdata and the taildata zero padding of extracting to M data point corresponding to bandwidth, then carry out the fast Fourier transform of M point, obtain frequency domain data D1 and D2; Extract K point useful in frequency domain data, be respectively cpdata_FD and taildata_FD;

C, make taildata_FD_neg=-taildata_FD;

D, cpdata_FD and taildata_FD_neg corresponding to user resources carry out associative operation:

$R=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_\mathrm{FD}\_\mathrm{neg}{\left(n\right)}^{*};$

Wherein N is the subcarrier number of CU;

E, phase place is got to correlated results R, gone out the frequency deviation of user by phase estimation.

2. method according to claim 1, is characterized in that, after described step B, performs following steps:

Cpdata_FD and taildata_FD corresponding to user resources carries out associative operation:

$R=-\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_{\mathrm{FD}}^{*};$

Then step e is performed.

3. method according to claim 1, is characterized in that, after described step B, performs following steps:

Cpdata_FD and taildata_FD corresponding to user resources carries out associative operation:

$R=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{cpdata}\_\mathrm{FD}\left(n\right)*\mathrm{taildata}\_{\mathrm{FD}}^{*}$

Phase is got to correlated results R;

Make φ=φ+π, and judge φ value size, φ value is corrected as follows:

$\mathrm{\&phi;}=\left\{\begin{array}{cc}\mathrm{\&phi;}-2\mathrm{\&pi;}& \mathrm{\&phi;}>\mathrm{\&pi;}\\ \mathrm{\&phi;}& -\mathrm{\&pi;}<\mathrm{\&phi;}\&le;\mathrm{\&pi;}\\ \mathrm{\&phi;}+2\mathrm{\&pi;}& \mathrm{\&phi;}\&le;-\mathrm{\&pi;}\end{array}\right.;$

According to the frequency deviation of the phase estimation user after correction.

4. the method according to claim 1,2 or 3, is characterized in that, if bandwidth is 20M, M=2048, K=1200.