JP2019083442A - Frequency corrector, demodulation circuit, radio, and frequency correction method - Google Patents

Abstract

To provide a method of performing frequency correction with a small amount of calculation with good accuracy as a method of performing frequency correction by using a plurality of pilot symbols.SOLUTION: A frequency corrector includes a frequency correction value deriving unit that acquires phase information for each pilot symbol and stores the phase information in a memory, complex-multiplies each data symbols a plurality of times according to a modulation scheme to generate a pseudo pilot signal, acquires phase information on each generated pseudo pilot signal, stores the phase information in the memory, and takes the average to derive the frequency correction value with reference to a phase information group stored in the memory, and a frequency correction unit that corrects deviation of an input signal on the basis of the frequency correction value.SELECTED DRAWING: Figure 1

Description

  The present invention relates to symbol correction techniques performed in a receiver using a digital modulation scheme.

  In a receiver of a digital communication system, communication performance is improved by performing frequency correction of a carrier frequency (including a subcarrier frequency).

  Such a technique is described, for example, in Patent Document 1. Patent Document 1 discloses a technique for performing frequency correction for wireless communication. This patent document 1 describes that in the receiver, a frequency error is removed from the sample to obtain a frequency-corrected sample. It is also described that there is a residual error in frequency error estimation, that a phase error occurs in a frequency-corrected sample due to the residual error, and that the performance of wireless communication may be deteriorated due to the phase error. Further, Patent Document 1 describes an invention for correcting a phase error of a data symbol. Note that the frequency error suggests that it is due to the difference in oscillator frequency between the transmitter and the receiver, the Doppler shift, and the like.

  Patent Document 2 also describes a technique for correcting a phase error after estimating a frequency error on the receiver side and performing frequency error correction.

  In some current communication systems, pilot symbols are sequentially inserted so as to sandwich data symbols at the transmitter, and a plurality of pilot symbols are used in the receiver to calculate frequency correction values of the data symbols. More specifically, the receiver calculates the amount of phase shift of each pilot symbol from the plurality of received pilot symbols, and calculates the frequency correction value of the data symbol group by taking the average.

  Patent Document 3 describes one method of calculating a frequency correction value of a data symbol using a plurality of pilot symbols in a receiver and correcting the frequency of a carrier of the data symbol.

JP, 2012-182801, A Unexamined-Japanese-Patent No. 09-307525 JP 2001-339363 A

  In the method of calculating the frequency correction value of a data symbol using a plurality of pilot symbols in a receiver such as Patent Document 3, one of the pilot symbols to be averaged due to spatial noise or fading is greatly distorted compared to other pilot symbols. In the case of a drop, the average value is pulled large by the amount of deviation of the pilot symbol, and the error of the frequency correction value becomes large.

  Further, in the methods of Patent Document 1 and Patent Document 2, the phase difference at the corresponding symbol determination point between the ideal signal and the measurement signal is obtained after performing frequency error correction, and the phase error is corrected. In the phase error correction after the estimation of the phase error for each symbol and the frequency error correction (frequency error estimation), the circuit scale, the calculation resource consumption amount, the power consumption and the like become large.

  Further, in the case of communication in a narrow band in which the allocated communication band or the communication band to be used is limited, the radio frame is configured with a small number of pilot symbols. In this case, since the influence of each pilot symbol becomes relatively large, a method of calculating a more accurate frequency correction value is desired. On the other hand, in the correction methods of Patent Document 1 and Patent Document 2, the circuit scale, the calculation resource consumption amount, the power consumption and the like become large.

  The present invention has been made in view of the above problems, and as a method of frequency correction using a plurality of pilot symbols, a frequency corrector that corrects the frequency with a low calculation amount with good accuracy, a demodulation circuit, a wireless device, And provide a frequency correction method.

  The frequency corrector according to an embodiment of the present invention acquires phase information for each pilot symbol and stores the phase information in a memory, and performs complex multiplication of the data symbol for each data symbol multiple times according to the modulation scheme. A pilot signal is generated, each phase information of the generated pseudo pilot signal is acquired, and stored in the memory, and the phase information group stored in the memory is referenced to take an average and derive a frequency correction value. And a frequency correction unit that corrects the deviation of the input signal based on the frequency correction value.

  A demodulation circuit according to an embodiment of the present invention is characterized by including the above frequency corrector.

  Similarly, a radio set according to an embodiment of the present invention is characterized by comprising a demodulation circuit including the above frequency corrector.

  Further, in the frequency correction method by the frequency corrector according to one embodiment of the present invention, phase information for each pilot symbol is acquired and accumulated, and the data symbol is complex-multiplied several times according to the modulation scheme for each data symbol. Pseudo pilot signals are respectively generated, respective phase information of the generated pseudo pilot signals is acquired and accumulated, a frequency correction value is derived with reference to the accumulated phase information group, and the frequency correction value is calculated based on the frequency correction value. To compensate for the deviation of the input signal.

  According to the present invention, as a method of performing frequency correction using a plurality of pilot symbols, it is possible to provide a frequency corrector, a demodulation circuit, a wireless device, and a frequency correction method capable of performing frequency correction with low calculation amount and good accuracy.

It is a block diagram showing frequency corrector 1 of one embodiment. It is a flowchart which shows the frequency correction process of the frequency correction device 1 of one Embodiment. FIG. 2 is an explanatory view showing a configuration example of a demodulator including a frequency corrector 1 and a frame format example. FIG. 2 is a block diagram showing an example of the configuration of a frequency corrector 1; It is explanatory drawing which shows the receiving symbol map of QPSK, and the receiving symbol map after multiplication calculation. It is an explanatory view showing the order and time series of the phase data of a pilot signal and a false pilot signal. It is explanatory drawing which shows the simulation result at the time of calculating the frequency correction value whose Eb / N0 is 10 dB. It is explanatory drawing which shows the simulation result of the frequency correction value in case Eb / N0 is close to an ideal state of 100 dB.

  Embodiments of the present invention will be described based on the drawings.

First Embodiment
FIG. 1 is a block diagram showing a frequency corrector 1 of the embodiment.

  The frequency corrector 1 shown in FIG. 1 includes a frequency correction value deriving unit 10 and a frequency correction unit 20 as a part of a demodulation circuit of a receiver (wireless device). The frequency correction unit 1 receives the symbol string, calculates the frequency correction value by the frequency correction value deriving unit 10, corrects the frequency of the symbol string by the frequency correction unit 20, and outputs it. The pre-stage circuit and the post-stage circuit of the frequency corrector 1 may adopt an existing circuit configuration suitable for the communication system.

  Frequency correction value deriving unit 10 receives a symbol (a pilot symbol) of a pilot signal and a data signal (a data symbol) included in a symbol string, and derives a frequency correction value as described later with reference to each symbol. . Further, the frequency correction unit 20 corrects the frequency of the input signal based on the frequency correction value derived by the frequency correction value deriving unit 10.

  The frequency correction value deriving unit 10 first acquires phase information for each pilot symbol. At the same time, the frequency correction value deriving unit 10 performs complex multiplication on the IQ data of each symbol multiple times depending on the modulation scheme implemented on the transmitter side for each symbol input as a data section, thereby generating a pseudo pilot signal (one point Respectively). Next, the frequency correction value deriving unit 10 acquires each piece of phase information of the generated pseudo pilot signal for each symbol.

  In M-PSK (M-Phase-Shift Keying) (M = 4 for QPSK (Quadrature Phase Shift Keying), M = 8 for 8 PSK), the self symbol is complexed (M-1) times When multiplied, it converges to one point. The convergence point is used as a pseudo pilot signal to acquire phase information of the corresponding symbol.

  In the process up to this point, the frequency correction value derivation unit 10 generates phase information group of pilot symbols and symbol pilot information which is input as a data section obtained by generating a pseudo pilot signal from a symbol string sequentially input. Get a group.

  Next, the frequency correction value deriving unit 10 derives a frequency correction value with reference to the phase information group of each symbol input as the data section together with the phase information group of the pilot symbol.

  As described above, in the process of calculating the frequency correction value based on the phase of the pilot symbol input as the pilot signal, the frequency correction value derivation unit 10 uses the data symbols of the data section sandwiched by the pilot symbols to generate a pseudo pilot. Generate a signal and obtain its phase value. Then, the frequency correction value deriving unit 10 refers to the phase values of the pilot signal and the pseudo pilot signal according to the input timing of each symbol, and calculates the frequency correction value with an average of an arbitrary period. This arbitrary period may be, for example, between one pilot symbol or between multiple pilot symbols. Further, the number of generations of pseudo pilot signals can be up to the number of data symbols as the number of generation of at least one generated between pilot symbols. By increasing the number of samples, the frequency error can be averaged with more samples, and for example, the influence can be reduced even when noise is present on a specific pilot signal. Also, the averaging may be performed using, for example, the least squares method. In the case of averaging by the least squares method, a linear approximation may be derived with reference to the time series of each symbol and the phase transition of each symbol, and the slope of the straight line may be output as a frequency correction value.

  The frequency correction unit 20 performs frequency correction based on the frequency correction value derived by the frequency correction value deriving unit 10. This corrects the deviation of the input signal.

  By configuring as described above, as a method of performing frequency correction using a plurality of pilot symbols, it is possible to obtain a frequency corrector capable of performing frequency correction with good accuracy with a small amount of calculation.

  The frequency corrector 1 may be configured as a dedicated circuit or may be configured using a microcomputer. FIG. 2 is a flowchart showing the frequency correction process of the frequency corrector 1 of the embodiment.

  The frequency corrector 1 (frequency correction value deriving unit 10) receives the symbol string and detects the phase value of the pilot symbol (S101).

  The frequency corrector 1 also multiplies the data symbols of the received symbol string by the number of times according to the modulation scheme to generate a pseudo pilot signal, and detects the phase value of the multiplied symbol data (S102).

  The frequency corrector 1 calculates a frequency correction value with reference to the phase value of the pseudo pilot signal as well as the phase value of the pilot signal every arbitrary period (S103).

  The frequency corrector 1 (frequency corrector 20) corrects the frequency of the symbol string using the frequency correction value calculated every arbitrary period (S104).

  Thus, the frequency corrector 1 provides sample points for deriving the frequency correction value also in the data symbol section to derive an accurate frequency correction value.

  Note that an existing method may be used for the frequency correction method performed on each symbol. In addition, as the symbol string to be corrected, the frequency correction value may be applied to the data symbol of the derived arbitrary period in accordance with the data delay tolerance or the like, or the frequency correction derived to the data symbol of the next arbitrary period A value may be applied.

  By operating the frequency correction value deriving unit 10 as described above, even if the pilot signal used for frequency correction in the radio frame can not be sufficiently secured, the modulation side is changed in the demodulation process. It is possible to calculate a highly accurate frequency correction value from existing received radio waves.

  As a result, as a method of frequency correction using a plurality of pilot symbols, it is possible to provide a frequency correction method that can perform frequency correction with a low amount of calculation and good accuracy.

  Next, the present invention will be described by showing a configuration example of a demodulator and a frequency corrector.

  FIG. 3 is an explanatory view showing a configuration example of a demodulator including the frequency corrector 1 and a frame format example. FIG. 4 is a block diagram showing a configuration example of the frequency corrector 1.

  The demodulator shown in FIG. 3A includes an analog receiver 2, an orthogonal demodulator 3, a band limiting filter 4, a synchronization word detector 5, a frequency corrector 1, a symbol timing detector 6, and a symbol / bit converter 7. , Frame remover 8. The frequency correction unit 1 shown in FIG. 4 includes the frequency correction value deriving unit 10, the phase conversion unit 11, the first storage unit 12, the complex multiplication unit 13, the phase conversion unit 14, the second storage unit 15, and an approximation. It is configured by the straight line calculation unit 16. In the frame format shown in FIG. 3B, nine data symbols, which are user data, are arranged in each DATA section between pilot signals (pilot symbols).

  In the demodulator, AGC (Automatic Gain Control) is performed in the analog reception unit 2, and IQ data output from the orthogonal demodulator 3 is taken in a packet unit, passed through the band limiting filter 4, and then the synchronization word detector 5 To detect the synchronization word as a reference. Then, the demodulator corrects the frequency shift component of the next incoming IQ data with the frequency corrector 1, extracts the data at the symbol timing with the symbol timing detector 6, and sends it to the symbol / bit converter 7. Convert to bit data. Finally, the demodulator removes extra frames in the frame remover 8 and outputs it as user data.

  The frequency corrector 1 converts IQ data of the pilot symbol of the input symbol string into phase information in the phase converter 11, stores the converted phase information in the first storage unit 12, and approximates the accumulated phase information group Input to the straight line calculation unit 14. At the same time, in the frequency correction unit 1, the complex multiplication unit 13 multiplies the data symbols of the input symbol string by the number of times based on the modulation scheme. Then the IQ data converges to one point. The frequency corrector 1 uses the converged IQ data as a pseudo pilot signal. The frequency correction unit 1 converts the converged IQ data into phase information by the phase conversion unit 14, stores the converted phase information in the second storage unit 15, and inputs the accumulated phase information group to the approximate straight line calculation unit 16 Do. Then, the frequency correction unit 1 calculates the inclination of the approximate straight line for the phase information group by the approximate straight line calculation unit 16 by the least squares method. The inclination that is the calculation result is corrected as the frequency shift component to the IQ data input to the frequency correction unit 20.

The slope of the linear approximation by the least squares method can be calculated using Equation 1. N is the number of symbols for performing linear approximation, X is a symbol position on the frame, and Y is phase information.

  Here, the operation of the frequency correction value deriving unit 10 will be described in detail by taking the QPSK modulation method (M = 4) as an example. FIG. 5 is an explanatory view showing a received symbol map of QPSK and a received symbol map after multiplication calculation.

  The phase converter 11 sequentially converts IQ data of pilot symbols from input IQ data of the frequency corrector 1 into phase information.

  In parallel, the complex multiplier 13 complex-multiplies the self symbol from the input IQ data 3 (M-1) times. Then, as shown in FIG. 5A, the QPSK symbols divided into four points converge to one point as shown in FIG. 5B. Next, the phase conversion unit 14 sequentially converts the IQ data converged to one point into phase information.

  The approximate straight line calculation unit 16 arranges and averages each phase data of each pseudo pilot symbol obtained by shifting each phase data of each pilot signal by one sample. At this time, since the approximate straight line calculation unit 16 multiplies the phase of each pseudo pilot symbol by (M-1) in the complex multiplication processing (the component of the frequency shift is also multiplied by (M-1)), the modulation method is The phase value is divided on the basis of the number of times (M-1) multiplied in accordance with.

  In the explanatory diagram shown in FIG. 6A, when each phase data of the pilot symbol group is A, B, C, D, E, F, and G, the phase data group shifted by one sample is A + 1, Phase data group shifted by 2 samples by B + 1, C + 1, D + 1, E + 1, F + 1, G + 1, A + 2, B + 2, C + 2, D + 2, E + A sequence of phase data is generated by similarly shifting 2 to 3 samples by 2 to 9 and 2 to 2 to visualize the arrangement.

  The explanatory diagram shown in FIG. 6B shows a state in which each phase data of the pilot signal and the pseudo pilot signal is arranged in time series. Thus, the phase information groups of the pilot signal and the pseudo pilot signal are arranged at equal intervals.

  For the phase information groups arranged at equal intervals in accordance with this time series, linear approximation is performed using each phase data group, and the slopes of the respective approximate straight lines are averaged. This makes it possible to calculate a highly accurate frequency correction value.

  In this manner, pilot symbol positions are denoted as A, B, C, ... on the radio frame, and A + 1, B + 1, C + 1, ... A + 2, B + 2, C + 2. It is possible to obtain good frequency correction values by defining and using each data symbol and performing linear approximation for each data symbol group according to the distance from the pilot signal and taking the average of the respective slopes. . As a result, even if the carrier to noise ratio (C / N) is deteriorated, the error of the frequency shift calculation result can be reduced.

  FIG. 7 shows, as an example, the calculation result (a) and calculation of this method when the frequency correction value is calculated using only the pilot symbol with respect to the reception value of Eb / N0 (energy per bit to noise power spectral density ratio) 10 dB. It is a simulation result which shows a result (b). FIG. 8 shows the calculation result of the frequency correction value when Eb / N0 is close to the ideal state of 100 dB. The equation shown in the upper right of the figure is the equation of a straight line as a result of the straight line approximation, and the coefficient of the X term shows the slope.

  As shown in FIG. 7A, when the frequency correction value is calculated using only the pilot symbol, y = 0.0263x + 0.1244. On the other hand, in the present method, it was calculated that y = 0.0205x + 0.3156. It can be understood that the present method is closer to the ideal state when this inclination is compared with y = 0.021x + 0.0018 which indicates a frequency correction value (slope) in the case close to the ideal state shown in FIG.

  Since the pilot symbol alone is used to calculate the frequency correction value in a state where the number of samples is small, the error of the calculated value of the frequency correction value becomes larger than the result at the ideal value due to the influence of noise. As a result, when Eb / N0 is deteriorated, frequency correction causes the deterioration of BER (Bit Error Rate).

  In this method, complex multiplication is performed (M-1) times in accordance with M of the M-PSK communication method to generate a pseudo pilot signal so that all symbols can be regarded as known symbols, and a frequency correction value is calculated. To derive.

  In addition to linear approximation of the pilot symbol group, linear approximation is similarly performed on the data symbol group, and a frequency correction value is derived from the slope of the straight line. The number of samples can simply be increased by at most data symbols. The results show that this method is closer to the ideal value.

  As described above, the frequency corrector to which the present invention is applied can perform frequency correction with a low amount of calculation with good accuracy as a method of frequency correction using a plurality of pilot symbols. Similarly, according to the present invention, as a method of frequency correction using a plurality of pilot symbols, it is possible to provide a demodulation circuit, a wireless device, and a frequency correction method capable of performing frequency correction with a low amount of calculation with good accuracy.

  The present invention has been described by exemplifying the embodiment. However, the specific configuration of the present invention is not limited to the above-described embodiment, and any changes without departing from the scope of the present invention are included in the present invention. For example, modifications such as separation and merging of block configurations and replacement of procedures in the above-described embodiment are free as long as the purpose of the present invention and the functions to be described are satisfied, and the above description does not limit the present invention.

Also, some or all of the above embodiments may be described as follows. The following appendices do not limit the present invention at all.
[Supplementary Note 1]
In the process of calculating the frequency correction value based on the phase of the pilot symbol input as the pilot signal,
In addition to acquiring phase information for each pilot symbol,
IQ data for each symbol input as a data section is multiplied a plurality of times according to the modulation scheme to generate pseudo pilot signals, and phase information of the generated pseudo pilot signals for each symbol is acquired,
A frequency correction value deriving unit that derives a frequency correction value with reference to a phase information group for each symbol input as a data section obtained by generating a pseudo pilot signal together with a phase information group of a pilot symbol;
A frequency correction unit that performs frequency correction based on the frequency correction value derived by the frequency correction value deriving unit;
A frequency corrector characterized by comprising.

[Supplementary Note 2]
A pilot signal phase converter for acquiring phase information of a pilot symbol from a pilot symbol input as a pilot signal;
A complex multiplication unit which multiplies IQ data of the input data symbol by the number of times based on the modulation scheme and converges;
A data signal phase converter for converting IQ data of converged data symbols into phase information;
An approximate straight line calculation unit that calculates the inclination of the approximate straight line indicated by the phase information of the pilot symbol and the phase information of the data symbol as a frequency correction value indicating a frequency shift component;
A frequency correction unit that performs frequency correction on the data symbol input based on the frequency correction value calculated by the approximate straight line calculation unit;
A frequency corrector characterized by comprising.

[Supplementary Note 3]
When deriving the frequency correction value, the phase value is divided based on the number of times the generated phase information of the pseudo pilot signal for each symbol is multiplied according to the modulation scheme, and then added to the phase information group of pilot symbols. The frequency corrector as set forth in claim 1 or 2, characterized in that a frequency correction value is calculated.

[Supplementary Note 4]
When deriving the frequency correction value, for each phase information of the generated pseudo pilot signal for each symbol, between the phase information of any pilot symbol and the phase information of the next pilot symbol, along the time series, The phase information of the pseudo pilot signal obtained by dividing the phase value based on the number of times of multiplication according to the modulation scheme is inserted at a position shifted by an arbitrary number of symbols to derive an approximate straight line of the least squares method. The frequency corrector according to any one of 1 to 3.

[Supplementary Note 5]
The phase information of the pseudo pilot signal is generated for all symbols input as a data section, and the frequency correction values are derived from the average of the inclinations of the pilot signal and the pseudo pilot signal arranged at equal intervals. 5. A frequency corrector according to any one of 4.

[Supplementary Note 6]
The phase information of the pseudo pilot signal is generated for the symbol in the middle of the data section sandwiched by the pilot symbols, and the frequency correction value is derived from the average of the inclinations of the pilot signal and the pseudo pilot signal arranged at equal intervals. The frequency compensator according to any one of supplementary notes 1 to 4, characterized by the above.

[Supplementary Note 7]
The phase information of the pseudo pilot signal is generated for all the symbols input to the plurality of data sections, the phase information group of data symbols of the plurality of data sections is acquired for each shift amount from each pilot symbol, and Each linear approximation line is calculated from the phase information group of data symbols having the same shift amount, and the slope of the straight line obtained by averaging the linear approximation lines of each shift amount is calculated as a frequency correction value. The frequency corrector according to any one of 1 to 4.

[Supplementary Note 8]
A demodulation circuit including the frequency corrector according to any one of appendices 1 to 7.

[Supplementary Note 9]
A radio set comprising a demodulation circuit including the frequency corrector according to any one of appendices 1 to 7.

[Supplementary Note 10]
Each phase information of each pilot symbol input as a pilot signal is acquired,
IQ data for each symbol input as a data section is multiplied a plurality of times according to the modulation scheme to generate pseudo pilot signals, and phase information of the generated pseudo pilot signals for each symbol is acquired,
The frequency correction value is derived with reference to the phase information group for each symbol input as a data section obtained by generating the pseudo pilot signal together with the phase information group of the pilot symbol,
A frequency correction method using a frequency corrector characterized by performing frequency correction based on the derived frequency correction value.

  The present invention can be used as a demodulator in a time division multiple access (TDMA) communication system radio, a minimum shift keying (MSK) modulation, and a quadrature amplitude modulation (QAM) modulation.

1 frequency corrector 10 frequency correction value deriving unit 11 phase converter (for pilot signal)
12 first storage unit 13 complex multiplication unit 14 phase conversion unit (for data signal)
15 second storage unit 16 approximate straight line calculation unit 20 frequency correction unit

Claims (8)

The phase information of each pilot symbol is acquired and stored in a memory, and the data symbol is complex-multiplied several times according to the modulation scheme for each data symbol to generate a pseudo pilot signal, and the generated pseudo pilot signal A frequency correction value deriving unit for acquiring each phase information and accumulating it in the memory, referring to the phase information group accumulated in the memory, taking an average to derive a frequency correction value;
A frequency correction unit that corrects the deviation of the input signal based on the frequency correction value;
A frequency corrector characterized by comprising. A pilot signal phase converter for acquiring first phase information from pilot symbols;
A first storage unit that stores the first phase information;
A complex multiplication unit which multiplies the data symbols by the number of times based on a modulation scheme and converges the data symbols;
A data signal phase converter for converting converged data symbols into second phase information;
A second storage unit that stores the second phase information;
An approximate straight line calculation unit that calculates as a frequency correction value indicating a frequency shift component using the first phase information and the second phase information;
A frequency correction unit that corrects the deviation of the input signal based on the frequency correction value;
A frequency corrector characterized by comprising.   When deriving the frequency correction value, the phase value is divided based on the number of times of complex multiplication according to the modulation method for each phase information of the generated pseudo pilot signal for each symbol, and then the phase information group of pilot symbols is obtained. The frequency corrector according to claim 1, further comprising calculating a frequency correction value.   When deriving the frequency correction value, for each phase information of the generated pseudo pilot signal for each symbol, between the phase information of any pilot symbol and the phase information of the next pilot symbol, along the time series, The phase information of a pseudo pilot signal whose phase value is divided based on the number of times of complex multiplication according to the modulation scheme is inserted at a position shifted by an arbitrary number of symbols to derive an approximate straight line of the least squares method. The frequency corrector according to claim 3.   The phase information of the pseudo pilot signal is generated for all the symbols input to the plurality of data sections, the phase information group of data symbols of the plurality of data sections is acquired for each shift amount from each pilot symbol, and Each linear approximation line is calculated from the phase information group of data symbols having the same shift amount, and the slope of the straight line obtained by averaging the linear approximation lines of each shift amount is calculated as a frequency correction value. Item 4. The frequency corrector according to item 3 or 4.   A demodulation circuit including the frequency corrector according to any one of claims 1 to 5.   A radio set comprising the demodulation circuit according to claim 6. Acquire and accumulate phase information for each pilot symbol,
Each of the data symbols is complex-multiplied a plurality of times according to the modulation scheme to generate the pseudo pilot signal.
Acquire and accumulate phase information of each of generated pseudo pilot signals;
The frequency correction value is derived with reference to the accumulated phase information group,
And correcting the deviation of the input signal based on the frequency correction value.

Images