PILOT FREQUENCY POSITION DETERMINING METHOD BASED ON PILOT FREQUENCY INTERVAL OPTIMIZATION, AND TRANSCEIVER DEVICE

Abstract

The present invention discloses a pilot frequency position determining method based on pilot frequency interval optimization and a transceiver device. The method optimizes a pilot frequency position on the basis of better use of pilot frequency in a wireless system for sampling frequency synchronization and residual phase tracking. After an optimal pilot frequency position is obtained according to the method, a transmitting terminal inserts a pilot frequency sequence at a corresponding pilot frequency position, a receiving terminal learns the pilot frequency position and the pilot frequency sequence, and after channel equalization, deviation is tracked by means of coherent detection of a local sequence. For the aforementioned method, the present invention further provides a transceiver device of a related pilot frequency module in the wireless system. The pilot frequency position can be better determined without increasing system complexity, and the present invention significantly improves the system performance.

Claims

1. A pilot frequency position determining method based on pilot frequency interval optimization, comprising the following steps: step 1, determining a value set of pilot frequency intervals according to the distribution condition of subcarriers in an IEEE 802.11 system; step 2, obtaining corresponding pilot frequency position sequences and a set thereof for the pilot frequency intervals in the value set of pilot frequency intervals in step 1; step 3, obtaining a direct product of the value set of pilot frequency deviations and the pilot frequency position sequence set in step 2, estimating sampling frequency deviations and residual phase deviations of spatial data flows, and counting normalized mean squared errors of the sampling frequency deviations; and step 4, selecting a pilot frequency interval which ensures a minimum normalized mean squared error of the sampling frequency deviations in step 3 from the value set of pilot frequency intervals in step 1, wherein the pilot frequency position sequence in the pilot frequency position sequence set corresponding to the pilot frequency interval is an optimal pilot frequency position sequence.

2. The pilot frequency position determining method based on pilot frequency interval optimization of claim 1, comprising the following specific steps: step 1, determining the value set of pilot frequency intervals according to the distribution condition of the subcarriers in the IEEE 802.11 system, specifically comprises: obtaining the value set D={d.sub.1,d.sub.2, . . . , d.sub.m} of pilot frequency intervals d within a value range $\left\{d|N_{D.Math..Math.C}+1\le d\le .Math.\frac{N_{\mathrm{SP}}+N_{\mathrm{SD}}+N_{D.Math..Math.C}-1}{N_{\mathrm{SP}}-1}.Math.,d.Math..Math.\mathrm{mod}.Math..Math.2=0\right\},$ wherein d.sub.l expresses a pilot frequency subcarrier interval t=1, 2, . . . , m , and m expresses the number of elements of the set D; N.sub.DC expresses the number of DC null subcarriers, N.sub.SP expresses the number of pilot frequency subcarriers, and N.sub.SD expresses the number of data subcarriers in each OFDM symbol; └.Math.┘ expresses round down; and mod2 expresses a remainder of dividing 2; step 2, obtaining the corresponding pilot frequency position sequences and the set thereof for the pilot frequency intervals in the value set of pilot frequency intervals in step 1, specifically comprises: expressing a pilot frequency subcarrier position ki by the pilot frequency interval d on the condition that pilot frequencies are symmetrically distributed on a subcarrier sequence at equal intervals as $k_i=\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right).Math.d,$ i=1,2, . . . , N.sub.SP, and then obtaining a corresponding pilot frequency position sequence b.sub.t=[k.sub.1,k.sub.2, . . . , k.sub.N.sub.SP] for ∀d.sub.t∈D, wherein the pilot frequency position sequence set is B={.sub.1,b.sub.2, . . . , b.sub.m}, and the number of elements of the set B is equal to the number of the elements of the set D; step 3, obtaining the direct product of the value set of pilot frequency deviations and the pilot frequency position sequence set in step 2, estimating the sampling frequency deviations and the residual phase deviations of the spatial data flows, and counting the normalized mean squared errors of the sampling frequency deviations, specifically comprises: setting the value set of the sampling frequency deviations {tilde over (ω)} as S={s.sub.1, s.sub.2, . . . , s.sub.n}, wherein s.sub.l expresses a random value of {tilde over (ω)} on Gaussian distribution in which a mean is 0 and a variance is σ.sup.2, the value range of {tilde over (ω)} is [−½,½], n expresses the number of the elements of the set S, and l=1,2, . . . n; the direct product of S and B is expressed as S×B={(s.sub.l,b.sub.t)|∀s.sub.l∈S∀b.sub.t∈B} and then each element in S×B corresponds to a combination of a sampling frequency deviation and a frequency position sequence selected in simulation, estimating the sampling frequency deviations and the residual phase deviations of the spatial data flows, and counting the normalized mean squared errors of the sampling frequency deviations; and step 4, selecting the pilot frequency interval which ensures the minimum normalized mean squared error of the sampling frequency deviations in step 3 from the value set D of pilot frequency intervals in step 1, wherein the pilot frequency position sequence in the pilot frequency position sequence set B corresponding to the pilot frequency interval is the optimal pilot frequency position sequence.

3. The pilot frequency position determining method based on pilot frequency interval optimization of claim 2, wherein the estimating the sampling frequency deviations and the residual phase deviations of the spatial data flows in step 3 is specifically as follows: on the condition of not considering noise influence, the pilot frequency sequence sent by a transmitting terminal is set as $x=\left[x_{k_1},\mathrm{.Math.}.Math..Math.x_{k_i}.Math..Math.\mathrm{.Math.}.Math.,x_{k_{N_{\mathrm{SP}}}}\right],$ and the pilot frequency sequence received by a receiving terminal is set as $y=\left[y_{k_1},\mathrm{.Math.}.Math..Math.y_{y_i}.Math..Math.\mathrm{.Math.}.Math.,y_{k_{N_{\mathrm{SP}}}}\right],$ wherein k.sub.i expresses a pilot frequency subcarrier position, namely the corresponding position of the i.sup.th pilot frequency is the k.sup.th subcarrier, h.sub.k.sub.i expresses a corresponding frequency domain channel, and then a model of the sampling frequency deviation {tilde over (ω)} and the residual phase deviation {tilde over (φ)} on the k.sub.i.sup.th pilot frequency subcarrier is as follows: $\begin{array}{cc}{y}_{k_i}=h_{k_i}.Math.x_{k_i}.Math.e^{j\left(\frac{2.Math.\pi }{N_{\mathrm{FFT}}}.Math.\tilde{\omega }.Math..Math.k_i+\tilde{\varphi }\right)},i=1,2,\mathrm{.Math.}.Math.,N_{\mathrm{SP}}& \left(1\right)\end{array}$ wherein, N.sub.FFT expresses the number of subcarriers in one OFDM symbol; j is an imaginary number; after channel equalization, the phase deviation obtained by coherent detection at the pilot frequency subcarrier is:
{tilde over (θ)}.sub.k.sub.i=∠(h.sub.k.sub.i.sup.−1y.sub.k.sub.ix.sub.k.sub.i*) i=1, 2, . . . , N.sub.SP   (2) wherein, ${\tilde{\theta }}_{k_i}=\frac{2.Math.\pi }{N_{\mathrm{FFT}}}.Math.\tilde{\omega }.Math..Math.k_i+\tilde{\varphi },$ ∠ expresses figuring out a phase value, (.Math.).sup.−1 expresses inversion, and (.Math.)* expresses figuring out conjugation; for N.sub.SP pilot frequency subcarriers, $\tilde{\theta }={\left[{\tilde{\theta }}_{k_1},{\tilde{\theta }}_{k_2},\mathrm{.Math.}.Math.,{\tilde{\theta }}_{k_{N_{\mathrm{SP}}}}\right]}^T,$ k=[k.sub.1,k.sub.2, . . . ,k.sub.N.sub.SP].sup.T is set, 1 expresses an all-1 vector with a size N.sub.SP×1, then: $\tilde{\theta }=\frac{2.Math.\pi }{N_{\mathrm{FFT}}}.Math.\tilde{\omega }.Math..Math.k+\tilde{\varphi }.Math..Math.1,.Math.{.Math.\tilde{\theta }-\frac{2.Math.\pi }{N_{\mathrm{FFT}}}.Math.\tilde{\omega }.Math..Math.k-\tilde{\varphi }.Math.1.Math.}^2={\left(\tilde{\theta }-\frac{2.Math.\pi }{N_{\mathrm{FFT}}}.Math.\tilde{\omega }.Math..Math.k-\tilde{\varphi }.Math..Math.1\right)}^T.Math.\left(\tilde{\theta }-\frac{2.Math.\pi }{N_{\mathrm{FFT}}}.Math.\tilde{\omega }.Math..Math.k-\tilde{\varphi }.Math.1\right)$ is minimized according to the least squares criterion, partial derivatives of {tilde over (ω)} and {tilde over (φ)} are respectively figured out, and a partial derivative formula is set to be equal to 0, and thus a sampling frequency deviation estimation value {tilde over (ω)}′ and a residual phase deviation estimation value {tilde over (φ)}′ in one OFDM symbol are respectively: $\begin{array}{cc}{\tilde{\omega }}^{\prime }=\frac{N_{\mathrm{FFT}}}{2.Math.\pi }.Math.\frac{N_{\mathrm{SP}}.Math.k^T.Math.\tilde{\theta }-{\tilde{\theta }}^T.Math.1.Math.k^T.Math.1}{N_{\mathrm{SP}}.Math.k^T.Math.k-{\left(k^T.Math.1\right)}^2}.Math..Math.{\tilde{\varphi }}^{\prime }=\left({\tilde{\theta }}^T.Math.1-{\tilde{\omega }}^{\prime }.Math.k^T.Math.1\right)/N_{\mathrm{SP}}& \left(3\right)\end{array}$ wherein, (.Math.).sup.T expresses figuring out transposition, and (.Math.).sup.2 expresses figuring out a square; $k_i=\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right).Math.d$ is substituted in the formula 3 to obtain an estimated sampling frequency deviation and an estimated residual phase deviation, which are expressed by the pilot frequency interval d: $\begin{array}{cc}{\tilde{\omega }}^{\prime }=\frac{N_{\mathrm{FFT}}}{2.Math.\pi }.Math.\frac{1}{d}.Math.\frac{N_{\mathrm{SP}}.Math.\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.{\tilde{\theta }}_{k_i}\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right)-N_{\mathrm{SP}}.Math.\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.{\tilde{\theta }}_{k_i}.Math.\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right)}{N_{\mathrm{SP}}.Math.\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.{\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right)}^2-{\left(\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right)\right)}^2}.Math..Math..Math.{\tilde{\varphi }}^{\prime }=\left(\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.{\tilde{\theta }}_{k_i}-\frac{2.Math.\pi }{N_{\mathrm{FFT}}}.Math.{\tilde{\omega }}^{\prime }.Math.\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right).Math.d\right)/N_{\mathrm{SP}}& \left(4\right)\end{array}$ moreover, the pilot frequency positions are symmetrically distributed about 0, namely $\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right)=0,$ and the formula 4 is simplified into $\begin{array}{cc}{\tilde{\omega }}^{\prime }=\frac{N_{\mathrm{FFT}}}{2.Math.\pi }.Math.\frac{1}{d}.Math.\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.{\tilde{\theta }}_{k_i}\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right)/\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.{\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right)}^2.Math..Math.{\tilde{\varphi }}^{\prime }=\underset{i=1}{\overset{N_{\mathrm{SP}}}{.Math.}}.Math.{\tilde{\theta }}_{k_i}/N_{\mathrm{SP}}.& \left(5\right)\end{array}$

4. The pilot frequency position determining method based on pilot frequency interval optimization of claim 2, wherein the expression of the pilot frequency interval {circumflex over (d)} for ensuring the minimum normalized mean squared error of the sampling frequency deviations in step 4 is:
{circumflex over (d)}=arg min {E[|{tilde over (ω)}′−{tilde over (ω)}|.sup.2]/E[{tilde over (ω)}.sup.2]} wherein, arg min{.Math.} expresses a parameter value ensuring a minimum value in the brackets, |.Math.| expresses taking an absolute value, E|.Math.| expresses expectation, and E[|{tilde over (ω)}′−{tilde over (ω)}|.sup.2]/E[{tilde over (ω)}.sup.2] expresses the normalized mean squared errors of the sampling frequency deviations.

5. A transceiver device of a related pilot frequency module in a wireless system, comprising a transmitter and a receiver, wherein the transmitter comprises a pilot frequency position optimization module and a pilot frequency insertion module, and the receiver comprises a phase tracking module and a pilot frequency removal module; the pilot frequency position optimization module is used for optimizing a pilot frequency position sequence by simulation according to the pilot frequency position determining method based on pilot frequency interval optimization of claim 1 on the premise of symmetrically inserting pilot frequencies at equal intervals; the pilot frequency insertion module is used for inserting pilot frequency for zero filling to obtain a complete OFDM symbol to constitute a spatial data flow output according to the pilot frequency position sequence obtained by the pilot frequency position optimization module; the phase tracking module is used for carrying out least squares on sampling frequency deviations and residual phase deviations estimated by the pilot frequency position optimization module to estimate an estimated sampling frequency deviation and an estimated residual phase deviation of each OFDM symbol, and removing the sampling frequency deviation and the residual phase deviation of each OFDM symbol; and the pilot frequency removal module is used for removing pilot frequency subcarriers and null subcarriers according to the pilot frequency position sequence obtained by the pilot frequency position optimization module to obtain the spatial data flow output that only contains data.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0036] FIG. 1 is a flowchart of a pilot frequency position determining method based on pilot frequency interval optimization.

[0037] FIG. 2 is a schematic diagram of distribution of a position sequence of pilot frequency subcarriers.

[0038] FIG. 3 is an NMSE curve at a 540 MHz bandwidth, wherein 64 QAM modulation, ½ code rate, antenna configuration 1×1 and one flow are adopted, a signal to noise ratio is 20 dB, and a frame length is 4096 bytes.

[0039] FIG. 4 is an NMSE curve at a 540 MHz bandwidth, wherein 64 QAM modulation, ½ code rate, antenna configuration 4×4 and 4 flows are adopted, the signal to noise ratio is 27 dB, and the frame length is 4096 bytes.

[0040] FIG. 5 shows a subcarrier distribution condition in one OFDM symbol in an embodiment of the present invention.

[0041] FIG. 6 is a frame error rate performance gain chart of a pilot frequency position obtained by the method in the present invention at a 540 MHz bandwidth and a traditional pilot frequency position, wherein BPSK modulation, ½ code rate, antenna configuration 1×1 and one flow are adopted, and the frame length is 4096 bytes.

[0042] FIG. 7 is a frame error rate performance gain chart of a pilot frequency position obtained by the method in the present invention at a 540 MHz bandwidth and the traditional pilot frequency position, wherein 64 QAM modulation, ½ code rate, antenna configuration 1×1 and one flow are adopted, and the frame length is 4096 bytes.

DETAILED DESCRIPTION

[0043] The technical solutions of the present invention will be illustrated below in detail in combination with the accompany drawings. Commonly used technical terms in the field involved in the present invention are shown in the following table.

TABLE-US-00001 Technical terms English Chinese NMSE Normalized Mean Normalized Mean Squared Error Squared Error QAM Quadrature Amplitude Quadrature Amplitude Modulation Modulation OFDM Orthogonal Frequency Orthogonal Frequency Division Multiplexing Division Multiplexing BPSK Binary Phase Shift Binary Phase Shift Keying Keying MIMO Multiple-Input Multiple- Multiple-Input Multiple- output output SISO Single-Input Single-Output Single-Input Single- Output

[0044] The present invention designs a pilot frequency position determining method based on pilot frequency interval optimization, as shown in FIG. 1, including the following steps:

[0045] step 1, determining a value set of pilot frequency intervals according to the distribution condition of subcarriers in an IEEE 802.11 system, specifically:

[0046] obtaining the value set D={d.sub.1, d.sub.2, . . . , d.sub.m} of pilot frequency intervals d within a value range

[00016]$\left\{d|N_{D.Math..Math.C}+1\le d\le .Math.\frac{N_{\mathrm{SP}}+N_{\mathrm{SD}}+N_{D.Math..Math.C}-1}{N_{\mathrm{SP}}-1}.Math.,d.Math..Math.\mathrm{mod}.Math..Math.2=0\right\},$

wherein d.sub.t expresses a pilot frequency subcarrier interval, t=1,2, . . . , m , and m expresses the number of elements of the set D; N.sub.DC expresses the number of DC null subcarriers, N.sub.SP expresses the number of pilot frequency subcarriers, and N.sub.SD expresses the number of data subcarriers in each OFDM symbol; └.Math.┘ expresses round down; and mod 2 expresses a remainder of dividing 2;

[0047] step 2, obtaining corresponding pilot frequency position sequences and a set thereof for the pilot frequency intervals in the value set of pilot frequency intervals in step 1, specifically:

[0048] expressing a pilot frequency subcarrier position k.sub.i by the pilot frequency interval d on the condition that pilot frequencies are symmetrically distributed on a subcarrier sequence at equal intervals as

[00017]$k_i=\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right).Math.d,$

i=1, 2, . . . , N.sub.SP, as shown in FIG.2, and then obtaining a corresponding pilot frequency position sequence b.sub.t=[k.sub.1, k.sub.2, . . . , k.sub.N.sub.SP for ∀d.sub.t∈D, wherein the pilot frequency position sequence set is B={b.sub.1, b.sub.2, . . . , b.sub.m}, and the number of elements of the set B is equal to the number of the elements of the set D;

[0049] step 3, obtaining a direct product of the value set of pilot frequency deviations and the pilot frequency position sequence set in step 2, estimating sampling frequency deviations and residual phase deviations of spatial data flows, and counting normalized mean squared errors of the sampling frequency deviations, specifically:

[0050] setting the value set of the sampling frequency deviations {tilde over (ω)} as S={s.sub.1, s.sub.2, . . . , s.sub.n}, wherein S.sub.1 expresses a random value of {tilde over (ω)} on Gaussian distribution in which a mean is 0 and a variance is σ.sup.2, the value range of {tilde over (ω)} is [−½,½] n expresses the number of the elements of the set S, and l=1,2, . . . n; the direct product of S and B is expressed as S×B={(s.sub.l, b.sub.t)|∀s.sub.l ∈S∀b.sub.t∈B}, and then each element in S×B corresponds to a combination of a sampling frequency deviations and a frequency position sequence selected in simulation, estimating the sampling frequency deviations and the residual phase deviations of the spatial data flows, and counting the normalized mean squared errors of the sampling frequency deviations; and

[0051] step 4, selecting a pilot frequency interval which ensures a minimum normalized mean squared error of the sampling frequency deviations in step 3 from the value set D of pilot frequency intervals in step 1, wherein the pilot frequency position sequence in the pilot frequency position sequence set B corresponding to the pilot frequency interval is an optimal pilot frequency position sequence.

[0052] The specific embodiments of the present invention will be further illustrated in detail in combination with the accompany drawing with a millimeter wave wireless local area network IEEE 802.11 aj (45 GHz) as an example. It should be understood that these embodiments are merely used for illustrating the present invention rather than limiting the scope of the present invention, and modifications in various equivalent forms made by those skilled in the art to the present invention after reading the present invention shall all fall within the scope of the appended claims.

[0053] In the embodiment of the present invention, the IEEE 802.11 aj (45 GHz) supports two bandwidths 540 MHz and 1080 MHz, and the bandwidth 540 MHz is taken as an example. In the 540 MHz bandwidth, there are 256 subcarriers in total, N.sub.SD=168 data subcarriers, N.sub.SP=8 pilot subcarriers, 39 null subcarriers at a low frequency end, 38 null subcarriers at a high frequency end, and N.sub.DS=3 DC null subcarriers. Specific simulation parameter configuration is as shown in table 1.

TABLE-US-00002 TABLE 1 simulation parameter setting Parameter Value Antenna number 1 × 1 Flow induced   1 Bandwidth 540 MHz Multipath delay spread 50-60 ns Distinguishable number finding   18 Low density parity check coding degree 672 bits Channel implementation time 10000

[0054] With 540 MHz bandwidth and 64 QAM modulation mode and ½ code rate as an example, the specific flows of the pilot frequency position determining method based on pilot frequency interval optimization are as follows:

[0055] step 1, the value set D={d.sub.1, d.sub.2, . . . , d.sub.m} of pilot frequency intervals d is obtained within a value range

[00018]$\left\{d|N_{D.Math..Math.C}+1\le d\le .Math.\frac{N_{\mathrm{SP}}+N_{\mathrm{SD}}+N_{D.Math..Math.C}-1}{N_{\mathrm{SP}}-1}.Math.,d.Math..Math.\mathrm{mod}.Math..Math.2=0\right\}.$

[0056] In the embodiment, the value range of the pilot frequency intervals d is {d|4≦d≦25, d mod 2=0}, and then the value set thereof is D={4, 6, 8,10,12,14,16,18, 20, 22, 24}.

[0057] Step 2, the pilot frequency subcarrier position k.sub.i is expressed by the pilot frequency interval d as

[00019]$k_i=\left(i-\frac{N_{\mathrm{SP}}+1}{2}\right).Math.d,$

then the corresponding pilot frequency position sequence b.sub.t=[k.sub.1, k.sub.2, . . . , k.sub.N.sub.SP] is obtained for ∀d.sub.t∈D , and the pilot frequency position sequence set is B={b.sub.1, b.sub.2, . . . , b.sub.m}.

[0058] In the embodiment, the correspondingly obtained pilot frequency position sequences can be expressed as [±1/2d,±3/2d,±5/2d,±7/2d ], and the pilot frequency position sequence set is B={[±2, ±6, ±10, ±14], . . . ,[±12, ±36, ±60, ±84]}.

[0059] Step 3, the value set of the sampling frequency deviations {tilde over (ω)} is set as S={s.sub.1,s.sub.2, . . . ,s.sub.n}, a direct product of S and B is expressed as S×B={(s.sub.l,b.sub.t)|∀s.sub.l ∈S∀b.sub.t ∈B}, then each element in S×B corresponds to a combination of a sampling frequency deviation and a frequency position sequence selected in simulation, the sampling frequency deviations and the residual phase deviations of the spatial data flows are estimated, and the normalized mean squared errors of the sampling frequency deviations are counted.

[0060] Step 4, the pilot frequency interval which ensures the minimum normalized mean squared error of the sampling frequency deviations is selected from the set D, and the pilot frequency position sequence in the pilot frequency position sequence set B corresponding to the pilot frequency interval is the optimal pilot frequency position sequence.

[0061] In the embodiment, in the case that the residual phase deviations on the frames obey (0, ½) Gaussian distribution within a range of [−π/6,π/6], 6 conditions of pre-added sampling frequency deviations on OFDM symbols are simulated:

[0062] 1. obeying the (0, 1/24) Gaussian distribution within the range of [−⅛,⅛];

[0063] 2. obeying the (0, 1) Gaussian distribution within the range of [−⅛,⅛];

[0064] 3. obeying the (0, 1/12) Gaussian distribution within the range of [−¼,¼];

[0065] 4. obeying the (0, 1) Gaussian distribution within the range of[−¼,¼];

[0066] 5. obeying the (0, ⅙) Gaussian distribution within the range of [−½,½]; and

[0067] 6. obeying the (0, 1) Gaussian distribution within the range of [−½,½].

[0068] It can be seen from a simulation result FIG. 3 that the optimal pilot frequency position sequence is independent of the magnitude and variance of the pre-added sampling frequency deviation, and an optimal value is obtained at d=22. In order to facilitate comparison in one figure, an NMESE curve corresponding to the Gaussian distribution within the range [−⅛, ⅛] of the pre-added sampling frequency deviation in FIG. 3 is 10 times as large as a simulation value. As shown in FIG. 3 and FIG. 4, the lowest point of each of all the NMESE curves is at d=22 in the case of SISO and MIMO and is the optimal pilot frequency interval that can be selected in the system, at this time, the pilot frequency subcarrier sequence is {±11, ±33, ±55, ±77}, and the position distribution is as shown in FIG. 5.

[0069] In order to compare the performance difference between the pilot frequency position obtained by the method provided by the present invention and the traditional pilot frequency position, a performance comparison chart with the traditional pilot frequency position is also provided in the present embodiment. In this embodiment, the performance of the present invention is illustrated by comparison by taking a spatial flow, a BPSK modulation mode and a 64 QAM modulation mode and a code rate ½ at the 540 MHz bandwidth as an example, which is specifically as shown in FIG. 6 and FIG. 7. FIG. 6 is a frame error rate performance gain chart of the pilot frequency position obtained by the method in the present invention at the 540 MHz bandwidth and the traditional pilot frequency position, wherein binary phase shift keying (Binary Phase Shift Keying, BPSK) modulation, antenna configuration 1×1 and one flow are adopted, and the frame length is 4096 bytes; and FIG. 7 is a performance gain chart of the pilot frequency position obtained by the method in the present invention at the 540 MHz bandwidth and the traditional pilot frequency position, wherein 64 QAM modulation, antenna configuration 1×1 and one flow are adopted, and the frame length is 4096 bytes. Therefore, the pilot frequency position obtained by the pilot frequency position method provided by the present invention is better than the traditional pilot frequency position.

[0070] The present invention further designs a transceiver device of a related pilot frequency module in a wireless system, including a transmitter and a receiver, wherein the transmitter includes a pilot frequency position optimization module and a pilot frequency insertion module, and the receiver includes a phase tracking module and a pilot frequency removal module;

[0071] the pilot frequency position optimization module is used for optimizing a pilot frequency position sequence by simulation according to the pilot frequency position determining method based on pilot frequency interval optimization on the premise of symmetrically inserting pilot frequencies at equal intervals;

[0072] the pilot frequency insertion module is used for inserting pilot frequency for zero filling to obtain a complete OFDM symbol to constitute a spatial data flow output according to the pilot frequency position sequence obtained by the pilot frequency position optimization module;

[0073] the phase tracking module is used for carrying out least squares on sampling frequency deviations and residual phase deviations estimated by the pilot frequency position optimization module to estimate an estimated sampling frequency deviation and an estimated residual phase deviation of each OFDM symbol, and removing the sampling frequency deviation and the residual phase deviation of each OFDM symbol according to the deviations; and

[0074] the pilot frequency removal module is used for removing pilot frequency subcarriers and null subcarriers according to the pilot frequency position sequence obtained by the pilot frequency position optimization module to obtain the spatial data flow output that only contains data.

[0075] According to the method for optimizing the pilot frequency position based on the pilot frequency intervals, the position where the pilot frequency needs to be inserted in the system is set, and the pilot frequency position obtained by optimization in the system is fixed and is known to the transceiver device of the related pilot frequency module in the wireless system. In the embodiment, the pre-added sampling frequency deviations on the OFDM symbols obey the (0, 1/24) Gaussian distribution within the range of [−⅛, ⅛], and the pilot frequencies are distributed at {±11, ±33, ±55, ±77}.

[0076] Transmitting and receiving steps of the related pilot frequency module in the wireless system are as follows:

[0077] step 1: converting each spatial data flow into a matrix form with a size of N.sub.SYM×N.sub.SD according to rows, wherein each row only contains data in one OFDM symbol, matrix column sequence numbers are sequentially arranged from positive to negative and small to large of carrier mark numbers, N.sub.SYM expresses the number of the OFDM symbols contained in each spatial data flow, and N.sub.SD expresses the number of the data subcarriers of one OFDM symbol corresponding to a corresponding bandwidth; in the embodiment, N.sub.SYM=66;

[0078] step 2: operating the matrix according to the rows, inserting the pilot frequency at the pilot frequency position obtained by optimization, carrying out zero filling at a corresponding position of each row according to the number and positions of required null subcarriers in the system, wherein the data on the original position moves backward in sequence, the column sequence numbers are sequentially arranged from positive to negative and small to large of the carrier mark numbers, a subcarrier mark number

[00020]$0.Math.•.Math..Math.\frac{N_{\mathrm{FFT}}}{2}-1$

corresponds to the column sequence number

[00021]$0.Math.•.Math..Math.\frac{N_{\mathrm{FFT}}}{2}-1,$

and the subcarrier mark number

[00022]$-\frac{N_{\mathrm{FFT}}}{2}.Math.•-1$

corresponds to the column sequence number

[00023]$\frac{N_{\mathrm{FFT}}}{2}.Math.•.Math..Math.N_{\mathrm{FFT}}-1.$

At this time, each row of the matrix is a complete OFDM symbol, and the size of the matrix is N.sub.SYM×N.sub.FFT;

[0079] step 3: converting the matrix form with the size of N.sub.SYM×N.sub.FFT into the spatial data flow with the size of 1×(N.sub.SYM×N.sub.FFT) according to rows;

[0080] step 4: operating each spatial data flow by the receiving terminal after receiving the spatial data flows, and converting the spatial data flow into the matrix form with the size of N.sub.SYM×N.sub.FFT according to rows;

[0081] step 5: operating each OFDM symbol according to rows, estimating a frequency deviation value and a phase deviation value at each pilot frequency, estimating and removing a frequency deviation value and a phase deviation value in one OFDM symbol according to the formula 7, wherein the size of the matrix is N.sub.SYM×N.sub.FFT at the moment;

[0082] step 6: operating each OFDM symbol according to rows, removing the pilot frequencies and null carriers at corresponding positions, and moving the data forward in sequence to fill the null positions, wherein the size of the matrix is N.sub.SYM×N.sub.SD ; and

[0083] step 7: converting the matrix form with the size of N.sub.SYM×N.sub.SD into spatial data flows with lengths of 1×(N.sub.SYM×N.sub.SD) according to rows.

[0084] The foregoing descriptions are merely specific embodiments in the present invention, but the protection scope of the present invention is not limited thereto. Variations or substitutions that can be understood and thought by anyone who is familiar with this art without the technical scope disclosed by the present invention shall fall within the scope of the present invention, and thus the protection scope of the claims prevail over the protection scope of the present invention.