TIME DOMAIN PILOT OF SINGLE-CARRIER MIMO SYSTEM AND SYNCHRONIZATION METHOD THEREOF

Abstract

The present invention discloses a time domain pilot design solution suitable for a single-carrier MIMO system. The design solution comprises a time domain pilot location design and a training sequence design. In the present invention, several identical ZCZ sequences are uniformly inserted into each of the data blocks in the same data stream to serve as training sequences, wherein the training sequences inserted into different data streams are different. In addition, the present invention also discloses a simple algorithm for pilot tracking and phase correction suitable for the time domain pilot design solution for the single-carrier MIMO system. The time domain pilot design solution for a single-carrier MIMO system and the algorithm for pilot tracking and phase correction as disclosed in the present invention can improve the performance of a system.

Claims

1. A time domain pilot method of a single-carrier MIMO system, comprising the following steps: (1) generating TS sequences to be inserted; (2) inserting the TS sequences into data blocks at intervals of an identical number of data symbols, wherein the TS sequences inserted into each data block are no less than one; and (3) after CSD operation and space mapping on the data blocks into which the TS sequences are inserted, inserting CPs before the transformed data blocks, wherein the last few digits of the CP are the TS sequences inserted into the data block.

2. The time domain pilot method of a single-carrier MIMO system of claim 1, wherein the TS sequences inserted into each data block in a same data stream are identical, while the TS sequences inserted into different data streams are different.

3. The time domain pilot method of a single-carrier MIMO system of claim 1, wherein the TS sequence adopts a ZCZ sequence.

4. A transmitting device for implementing time domain pilot of a single-carrier MIMO system, comprising a time domain pilot module, a CSD module, a space mapping module and a CP inserting module, wherein the time domain pilot module is used for generating TS sequences, and uniformly inserting the TS sequences in data blocks; the CSD module is used for performing CSD operation on the data blocks in which the TS sequences are inserted; the space mapping module is used for performing space mapping on the data blocks subjected to the CSD operation; and the CP inserting module is used for inserting CPs before the data blocks subjected to the space mapping.

5. A synchronization method based on time domain pilot of a single-carrier MIMO system, comprising the following steps: (1) performing FFT, equalization and IFFT transformation on received data blocks, and estimating a phase error at a central symbol position of each TS sequence in the data blocks; (2) utilizing the TS sequence in the CP for shifting, performing FFT, equalization and IFFT transformation on the shifted data blocks, and estimating a phase error at an initial position of data blocks; (3) estimating and compensating for a phase error of each data symbol in the data blocks.

6. The synchronization method based on time domain pilot of a single-carrier MIMO system of claim 5, wherein the method for estimating the phase error of each data symbol is as follows: supposing N.sup.TS represents the length of the TS sequence, N.sub.D represents the number of data symbols between every two adjacent TS sequences, M represents the number of TS sequences inserted into each data block; s[i],i=1, 2, . . . , N.sub.TS represents the inserted TS sequences; {circumflex over (x)}′[i], i=1, 2, . . . , N represents a time domain signal obtained after FFT, equalization and IFFT transformation of N points is performed on the received time domain signal y[i], i=1, 2, . . . , N; r.sub.m[i], i=1, 2, . . . , N.sub.TS, m=1, 2, . . . , M represents the i.sup.th pilot symbol of the m.sup.th TS sequence in the transformed data symbol block; and {circumflex over (x)}′[i], i=1, 2, . . . , N represents a time domain signal obtained after FFT, equalization and IFFT transformation of N points is performed on the received data symbol {y[−N.sub.TS+1], . . . , y[−1], y[0], y[1], y[2], . . . y[N−N.sub.TS+1]}, then the phase error at the central symbol position of each TS sequence in the data block is as follows: $\begin{array}{cc}{\theta }_m=\angle \left(\underset{i=1}{\overset{N_{\mathrm{TS}}}{.Math.}}.Math..Math.r_m\left[i\right].Math.s^{*}\left[i\right]\right),m=1,2,\mathrm{.Math.}.Math.,M& \left(\mathrm{formula}.Math..Math.12\right)\end{array}$ the phase error at the initial position of the data block is as follows: $\begin{array}{cc}{\theta }_0=\angle \left(\underset{i=1}{\overset{N_{\mathrm{TS}}}{.Math.}}.Math..Math.{\hat{x}}^{\prime }\left[i\right].Math.s^{*}\left[i\right]\right)& \left(\mathrm{formula}.Math..Math.14\right)\end{array}$ the phase error of each data symbol {circumflex over (x)}[i], i=1, 2, . . . , N in the data block is as follows: $\begin{array}{cc}\theta \left[i\right]=\{\begin{array}{cc}\frac{{\theta }_1-{\theta }_0}{N_D+N_{\mathrm{TS}}}.Math.\left(i-N_D-.Math.\frac{N_{\mathrm{TS}}+1}{2}.Math.\right)+{\theta }_1,& \mathrm{if}.Math..Math.\begin{array}{c}1\le i<N_D+\\ .Math.\frac{N_{\mathrm{TS}}+1}{2}.Math.\end{array}\\ \frac{{\theta }_m-{\theta }_{m-1}}{N_D+N_{\mathrm{TS}}}.Math.\left(\begin{array}{c}i-{\mathrm{mN}}_D-\\ \left(m-1\right).Math.N_{\mathrm{TS}}-\\ .Math.\frac{N_{\mathrm{TS}}+1}{2}.Math.\end{array}\right)+{\theta }_m,& \mathrm{if}.Math..Math.a\le i<b\end{array},.Math..Math.\mathrm{wherein}.Math.\text{.Math.}.Math..Math.a=\left(m-1\right).Math.N_D+\left(m-2\right).Math.N_{\mathrm{TS}}+.Math.\frac{N_{\mathrm{TS}}+1}{2}.Math.,.Math..Math.b={\mathrm{mN}}_D+\left(m-1\right).Math.N_{\mathrm{TS}}+.Math.\frac{N_{\mathrm{TS}}+1}{2}.Math..& \left(\mathrm{formula}.Math..Math.15\right)\end{array}$

7. The synchronization method based on time domain pilot of a single-carrier MIMO system of claim 5, wherein the method of phase error compensation is as follows:
{circumflex over ({circumflex over (x)})}[i]=x[i]□e.sup.−jθ[i], i=1, 2, . . . , N   (formula 16) wherein x[i], i=1, 2, . . . , N represents a time domain signal obtained after FFT, equalization and IFFT transformation of N points is performed on the received data symbol y[i], i=1, 2, . . . , N; and θ.sub.i, i=1, 2, . . . , N represents a phase error of each data symbol in the data blocks.

8. A receiving device for implementing synchronization based on a single-carrier MIMO system, comprising a shifting module, an FFT module, an equalization module, an IFFT module and a phase correction module, wherein the shifting module is used for performing shifting operation on received data blocks; the FFT module is used for performing FFT transformation on the received data blocks and the shifted data blocks; the equalization module is used for performing equalization on the data blocks subjected to the FFT transformation; the IFFT module is used for performing IFFT transformation on the equalized data blocks; and the phase correction module is used for performing phase error estimation and compensation on the data blocks subjected to the IFFT transformation.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0034] FIG. 1 is a block diagram of a transmitting device of a single-carrier MIMO system in the present invention;

[0035] FIG. 2 is a block diagram of a receiving device of a single-carrier MIMO system in the present invention;

[0036] FIG. 3 is a structural schematic diagram of data with pilot frequencies inserted in the present invention;

[0037] FIG. 4 is a schematic diagram of FFT operation on the shifted received data in the present invention;

[0038] FIG. 5 is a structural diagram of data in specific embodiments in the present invention;

[0039] FIG. 6 is a schematic diagram of FFT operation after shifting in specific embodiments of the present invention;

[0040] FIG. 7 is a simulation diagram of SISO phase noise performance in the embodiments of the present invention, wherein the simulation parameters include one transmitting antenna, one receiving antenna, one stream, and 16 QAM (Quadrature Amplitude Modulation); and

[0041] FIG. 8 a simulation diagram of MIMO phase noise performance in the embodiments of the present invention, wherein the simulation parameters include 2 transmitting antennas, 2 receiving antennas, 2 streams, and 16 QAM (Quadrature Amplitude Modulation).

DETAILED DESCRIPTION

[0042] With the millimeter wave wireless local area network IEEE 802.11aj (45 GHz) as an example, specific embodiments of the present invention are further described below in detail in conjunction with the accompanying drawings. It should be understood that these embodiments are merely used for illustrating the present invention, rather than limiting the scope of the present invention. After reading the present invention, various equivalent modifications made to the present invention by those skilled in the art fall within the claims appended in the present application.

[0043] A transmitting device of a single-carrier MIMO system as shown in FIG. 1 includes a time domain pilot module, a CSD module, a space mapping module and a CP inserting module, wherein the time domain pilot module is used for generating TS sequences, and uniformly inserting the TS sequences in data blocks; the CSD module is used for performing CSD operation on the data blocks in which the TS sequences are inserted; the space mapping module is used for performing space mapping on the data blocks subjected to the CSD operation; and the CP inserting module is used for inserting CPs before the data blocks subjected to the space mapping.

[0044] A receiving device of a single-carrier MIMO system as shown in FIG. 2 includes a shifting module, an FFT module, an equalization module, an IFFT module and a phase correction module, wherein the shifting module is used for performing shifting operation on received data blocks; the FFT module is used for performing FFT transformation on the received data blocks and the shifted data blocks; the equalization module is used for performing equalization on the data blocks subjected to the FFT transformation; the IFFT module is used for performing IFFT transformation on the equalized data blocks; and the phase correction module is used for performing phase error estimation and compensation on the data blocks subjected to the IFFT transformation.

[0045] FIG. 3 is a structural schematic diagram of data with pilot frequencies inserted in the present invention. In each data block of the same data stream of the present invention, namely, in each data symbol contained in each FFT operation, several identical training sequences (TSs) are uniformly inserted; the CP is inserted before each data block, the number of the TS sequences inserted into each data block is greater than or equal to 1, and the length of the inserted TS sequence is less than that of the inserted CP; the TS sequences inserted into each data block of the same data stream are identical, while the TS sequences inserted into different data streams are different; and the TS sequence adopts a ZCZ sequence.

[0046] Supposing that N represents the length of FFT operation, namely, the total length of a data block containing the pilot sequence and the data symbols, N.sub.TS represents the length of each section of separate TS sequence, N.sub.D represents the number of data symbols spaced between every two adjacent TS sequences, M represents the number of TS sequences inserted into each separate data block, and N.sub.CP represents the length of the cyclic prefix, then M, N.sub.TS, N, N.sub.D, and N.sub.CP are all integers and satisfy the following relationship:

MN.sub.TS+MN.sub.D=N,

N.sub.TS<N.sub.CP′,M≧1   (formula 1).

[0047] In the present embodiment, the specific steps of inserting time domain pilot frequencies into a transmitting terminal are as follows:

[0048] Step 1: determining the following parameter values: the total length N of each data block is 256, and the length includes the number of data symbols and the number of time domain pilot symbols, the length of CP (N.sub.CP) in the data blocks is 64, the number M of TS sequences inserted into each separate data block is 4, the length N.sub.TS of each TS sequence is 16, the number N.sub.D of data symbols spaced between every two adjacent TS sequences is 48; obviously, the parameters M, N.sub.TS, N, N.sub.D and N.sub.CP satisfy the relationship shown in formula 1;

[0049] Step 2: generating the TS sequence on the t.sup.th stream, namely, the ZCZ sequence with the length of N.sub.TS:

[0050] s.sub.t [1], s.sub.t [2], . . . , s.sub.t [i], . . . s.sub.t [N.sub.TS];

[0051] Step 3: after being subjected to a constellation mapper, inserting the ZCZ sequence s.sub.t [1], s.sub.t [2], . . . , s.sub.t [i], . . . s.sub.t [N.sub.TS] generated in step 2 into data blocks of corresponding data streams at intervals of N.sub.D data symbols;

[0052] Step 4: subjecting the data symbols with the time domain pilot inserted therebetween to the CSD and space mapping modules of a single-carrier MIMO transmitting system, and inserting a CP before each data block. The finally formed data structure is as shown in FIG. 5.

[0053] The phase offset models of the specific embodiments will be described below, which are shown as follows:

[00006]$\begin{array}{cc}\mathrm{PSD}\left(f\right)=\mathrm{PSD}\left(0\right)\left[\frac{1+{\left(f/f_z\right)}^2}{1+{\left(f/f_p\right)}^2}\right],& \left(\mathrm{formula}.Math..Math.11\right)\end{array}$

[0054] wherein PSD(0)=−85 dBc/Hz, PSD(∞)=−125 dBc/Hz, the pole frequency ƒ.sub.p=1 MHz and the zero frequency ƒ.sub.z=100 MHz .

[0055] In the present embodiment, the receiving terminal performs estimation and correction of a phase error on each data block in each stream, comprising the following specific steps:

[0056] Step 1: performing FFT of 256 points on the received time domain signal y[i], i=1, 2, . . . , 256, to obtain Y[k], k=1, 2, . . . , 256, namely

[00007]$\begin{array}{cc}Y\left[k\right]=\underset{i=1}{\overset{256}{.Math.}}.Math..Math.y\left[i\right].Math.e^{-j.Math.\frac{2.Math.\pi .Math..Math.\mathrm{ki}}{256}},k=1,2,\mathrm{.Math.}.Math.,256;& \left(\mathrm{formula}.Math..Math.12\right)\end{array}$

[0057] Step 2: equalizing the frequency domain signal Y[k], k=1, 2, . . . , 256 through a frequency domain equalizer to obtain {circumflex over (X)}[k], k=1, 2, . . . , 256, and supposing that the transmission function of the frequency domain equalizer is E(k) , then:

{circumflex over (X)}[k]=E(k)Y[k], k=1, 2, . . . , 256   (formula 13);

[0058] Step 3: transforming the signal {circumflex over (X)}[k], k=1, 2, . . . , 256 output from the equalizer into a time domain signal through IFFT transformation, wherein {circumflex over (x)}[i], i=1, 2, . . . , N represents a time domain signal obtained after {circumflex over (X)}[k],k=1, 2, . . . , N is subjected to IFFT, then:

[00008]$\begin{array}{cc}\hat{x}\left[i\right]=\frac{1}{N}.Math.\underset{k=1}{\overset{256}{.Math.}}.Math..Math.\hat{X}\left[k\right].Math.e^{j.Math.\frac{2.Math.\pi .Math..Math.\mathrm{ki}}{256}},i=1,2,\mathrm{.Math.}.Math.,256& \left(\mathrm{formula}.Math..Math.14\right)\end{array}$

[0059] wherein r.sub.m[i],i=1, 2, . . . , 16, m=1, 2, 3, 4 represents the i.sup.th pilot symbol of the m.sup.th TS sequence in the equalized current data symbol block, then r.sub.m[i] and {circumflex over (x)}[i] satisfy the following relationship:

r.sub.m[i]={circumflex over (x)}[48m+16(m −1)+i]

i=1, 2, . . . , 16, m=1, 2, 3, 4   (formula 15);

[0060] Step 4: calculating to obtain:

[00009]$\begin{array}{cc}{\theta }_m=\angle \left(\underset{i=1}{\overset{16}{.Math.}}.Math..Math.r_m\left[i\right].Math.s^{*}\left[i\right]\right),m=1,2,3,4;& \left(\mathrm{formula}.Math..Math.16\right)\end{array}$

[0061] Step 5: performing FFT of 256 points on the received data symbol {y[−15], . . . , y[−1], y[0], y[1], . . . , y[i], . . . y[N−15]}, to obtain Y′[k],k=1, 2, . . . , 256 after FFT operation, as shown in FIG. 6;

[0062] Step 6: performing E(k) equalization on Y′[k],k=1, 2, . . . , 256 through a frequency domain equalizer to obtain {circumflex over (X)}′[k],k=1, 2, . . . , 256 and {circumflex over (X)}′[k]=E(k)Y′[k], k=1, 2, . . . , 256;

[0063] Step 7: performing IFFT transformation on {circumflex over (X)}′[k],k=1, 2, . . . , 256 , to obtain the time domain signal {circumflex over (x)}′[i],i=1, 2, . . . , 256;

[0064] Step 8: calculating θ.sub.0, namely, a phase error at an initial position, by using a calculation method as follows:

[00010]$\begin{array}{cc}{\theta }_0=\angle \left(\underset{i=1}{\overset{16}{.Math.}}.Math..Math.{\hat{x}}^{\prime }\left[i\right].Math.s^{*}\left[i\right]\right);& \left(\mathrm{formula}.Math..Math.17\right)\end{array}$

[0065] Step 9: calculating the phase errorθ[i],i=1, 2, . . . , 256 corresponding to each data symbol in data symbol blocks by utilizing θ.sub.m, m=0, 1, 2, 3, 4, with the calculation method as follows:

[00011]$\begin{array}{cc}\theta \left[i\right]=\{\begin{array}{cc}\frac{{\theta }_1-{\theta }_0}{64}.Math.\left(i-48-9\right)+{\theta }_1,& \mathrm{if}.Math..Math.1\le i<57\\ \frac{{\theta }_m-{\theta }_{m-1}}{64}.Math.\left(\begin{array}{c}i-48.Math..Math.m-\\ 16.Math.\left(m-1\right)-9\end{array}\right)+{\theta }_m,& \mathrm{if}.Math..Math.a\le i<b\end{array},& \left(\mathrm{formula}.Math..Math.18\right)\end{array}$

[0066] wherein α=48(m−1)+16(m−2)+9. b=48m+16(m−1)+9;

[0067] Step 10: performing phase error compensation on {circumflex over (x)}[i], i=1, 2, . . . , 256 obtained in step 3 by θ[i],i=1, 2, . . . , 256 obtained in step 9, with the compensation method as follows:

{circumflex over ({circumflex over (x)})}[i]={circumflex over (x)}[i]□e.sup.−jθ[i], i=1, 2, . . . , 256   (formula 19).

[0068] To show the performance improvement of the present invention, the embodiments of the present invention also provide a simulation performance diagram, as shown in FIGS. 7 and 8. From the figures, it can be seen that the present invention can effectively track the phase and obviously improve the performance.