JP2011029686A - Transmission path estimating device and ofdm demodulator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To optimize the width of a transmission path estimation window by controlling the transmission path estimation window. <P>SOLUTION: A transmission path estimating device 20 includes an IFFT means for obtaining a delay profile, by performing IFFT with respect to a pilot symbol SP extracted from an OFDM signal IN; a generation means for determining a path presence range from a delay profile and generating a transmission path estimation window; and an operation means for performing transmission path estimation operation from the transmission path estimation window and the delay profile to obtain the result of transmission path estimation. The generation means includes a path detecting section 22 for detecting a path S22 from a delay profile S21; and an estimation window generating section 23 for generating the transmission path estimation window S23 from the path S22. The operation means includes a multiplier 24 for multiplying the transmission path estimation window S23 and the delay profile S21 to calculate a multiplication result S24; and an FFT part 25 for performing FFT, with respect to the multiplication result S24 to obtain a transmission path estimation result S25. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a transmission path estimator in OFDM demodulation that uses fast Fourier transform (hereinafter referred to as “FFT”) such as OFDM (orthogonal frequency division multiplexing) for demodulation, and an OFDM demodulator using the same. .

  The transmission path estimation method in OFDM demodulation is a technique applicable to, for example, an equalizer for a terrestrial integrated digital broadcasting service (hereinafter referred to as “ISDB-T”), an equalizer in OFDM, and the like.

  2. Description of the Related Art Conventionally, in a transmission path estimation method in OFDM demodulation, for example, a desired signal of an OFDM signal (scattered pilot (hereinafter referred to as “pilot symbol” in ISDB-T) is extracted) and their amplitude frequency characteristics and Calculation of transfer function estimation is performed using the phase frequency characteristics. Further, a delay profile (amount of power with respect to delay time, hereinafter simply referred to as “power amount”) can be obtained from the pilot symbols.

  As a method for obtaining a delay profile, for example, in Patent Document 1 below, after receiving an OFDM signal, performing frequency conversion by FFT and demodulating, a pilot symbol is extracted, and inverse fast Fourier transform (hereinafter referred to as “IFFT”). ) To find the delay profile.

  2A, 2B, and 2C are diagrams for explaining a transmission path estimation method in conventional OFDM demodulation. 2A and 2B, the horizontal axis is time (t), the vertical axes of FIGS. 2B and 2C are power (p), and the horizontal axis of FIG. 2C is frequency (f). ).

  As shown in FIG. 2A, the OFDM signal 1 has an effective OFDM symbol 1a and a guard interval GI added to the head thereof. A pilot symbol SP is extracted from the OFDM signal 1 to perform transfer function estimation calculation. In the transfer function estimation calculation, for example, as shown in FIG. 2C, pilot symbols SP (n−2), SP (n−1), SP (n) around the estimation target carrier (ie, estimation symbol 2). ), SP (n + 1), and SP (n + 2), the transfer function is estimated using the amplitude frequency characteristic and the phase frequency characteristic. At this time, the transfer path estimation window width affects the transfer function estimation calculation. The maximum value of the transmission path estimation window width W is determined according to the pilot symbol SP interval. For example, in channel estimation using ISDB-T 4 symbols (symbols), pilot symbols SP exist at intervals of three carriers, so the channel estimation window width W at the time of estimation is (effective OFDM symbol length ÷ 3). is there. If a path component deviating from the transmission path estimation window width exists on the delay profile, the accuracy of the transmission path estimation calculation decreases. Conversely, when all path components are sufficiently included in the transmission path estimation window width, the smaller the transmission path estimation window width W, the narrower the noise band and the noise resistance can be improved.

JP 2000-115087 A

  A transmission path estimation method using pilot symbols SP in OFDM demodulation includes an estimation method (1) in which pilot symbols SP are interpolated in the frequency direction, and a method (2) in which a delay profile is subjected to FFT calculation as follows. There is.

(1) Interpolation Interpolation Estimation Method The interpolation interpolation estimation method uses a low-pass filter (hereinafter referred to as “LPF”), and has the following characteristics regarding the passband of the filter.

  By narrowing the filter pass band, the noise band can be narrowed to suppress the estimation error, but this is a case where the path existence area is limited to a narrow range, and conversely multipath fading. Thus, when the path existence area is wide, the estimation error can be suppressed by interpolating the ripple caused by multipath fading with the surrounding carriers by widening the filter pass band. That is, the estimation error and the estimable path existence range are in a trade-off relationship.

(2) Method of Estimating from Delay Profile In the method of estimating from the delay profile, the estimation error is improved and deteriorated by multiplying the delay profile by a window function and performing an FFT operation. The window width of this window function also has the same characteristics as the filter band in the interpolation, and the estimation error and the estimable path existence range are in a trade-off relationship.

  FIGS. 3A and 3B are diagrams for explaining the problems of the conventional transmission path estimator. In FIGS. 3A and 3B, the horizontal axis represents time (t) and the vertical axis represents power (p).

  In the conventional transmission path estimator, the number of interpolated pilot symbols SP needs to be appropriately increased due to ripples generated in the frequency domain in multipath fading. The larger the path delay amount, the better the transmission path estimation window width W1, W2 becomes wider. On the other hand, since the path distribution is not constant in the actual reception environment, the maximum width is used as the transmission path estimation window width W in order to maximize the amount of path delay that can be demodulated.

  However, in order to optimize the transmission path estimation window width W, a mechanism for dynamically determining the path distribution and controlling the transmission path estimation window is necessary, but there is a trade-off between the estimation error and the estimable path existence range. Therefore, it has been difficult to realize an accurate mechanism as described above in the conventional transmission path estimator.

  A transmission path estimator according to a first aspect of the present invention includes IFFT means for obtaining a delay profile by performing IFFT on a pilot symbol extracted from an OFDM signal subjected to OFDM modulation, and a path existence range from the delay profile. A path determination / estimation window generating means for generating a transmission path estimation window, and a transmission path estimation for performing a transmission path estimation operation from the generated transmission path estimation window and the delay profile. And an arithmetic means.

  A transmission path estimator according to a second aspect of the invention is a pilot symbol extracting means for generating an FFT input data window from an OFDM signal subjected to OFDM modulation and extracting a pilot symbol by an FFT operation, and for the extracted pilot symbol. IFFT means for obtaining a delay profile by performing IFFT, path determination / estimation window generating means for determining a path existence range from the delay profile and generating a transmission path estimation window, the generated transmission path estimation window, and Transmission path estimation calculation means for obtaining a transmission path estimation result by performing transmission path estimation calculation from a pilot symbol; and FFT window base position control means for controlling a time base position in the input data window of the FFT calculation. It is characterized by.

  An OFDM demodulator according to a third invention equalizes the OFDM signal using the transmission path estimator according to the first or second invention and the transmission path estimation result obtained by the transmission path estimator. And a container.

  According to the transmission path estimator of the first invention and the OFDM demodulator of the third invention, the window profile of the FFT transmission path estimation window is dynamically controlled by determining the path existence range using the delay profile. As a result, transmission path estimation with the best estimation error according to the path existence range at the time of reception can be realized.

  According to the transmission path estimator of the second invention and the OFDM demodulator of the third invention, the same effect as that of the first invention can be obtained, and furthermore, filter coefficient control for the purpose of reducing arithmetic circuits, and this filter Optimal FFT window base point position control for improving the control effect can also be realized.

FIG. 1 is a schematic configuration diagram showing an OFDM demodulator having a transmission path estimator in OFDM demodulation according to Embodiment 1 of the present invention. FIG. 2 is a diagram for explaining a transmission path estimation method in conventional OFDM demodulation. FIG. 3 is a diagram for explaining a problem of a conventional transmission path estimator. FIG. 4 is a waveform diagram for explaining the estimation window generation operation of FIG. FIG. 5 is a waveform diagram for explaining the delay profile path detection threshold calculation of FIG. FIG. 6 is a waveform diagram for explaining the maximum power of the delay profile of FIG. FIG. 7 is a waveform diagram for explaining the gain control value in the previous stage of FIG. FIG. 8 is a waveform diagram for explaining the calculation of the delay profile path window width of FIG. FIG. 9 is a schematic configuration diagram illustrating an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the second embodiment of the present invention. FIG. 10 is a schematic configuration diagram illustrating an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the third embodiment of the present invention. FIG. 11 is a waveform diagram showing the estimation window generation of FIG. FIG. 12 is a waveform diagram showing the transmission path estimation of FIG. FIG. 13 is a schematic configuration diagram illustrating an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the fourth embodiment of the present invention. FIG. 14 is a waveform diagram showing the transmission path estimation of FIG. FIG. 15 is a schematic configuration diagram showing an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the fifth embodiment of the present invention. FIG. 16 is a waveform diagram showing the FFT window base point control of FIG. FIG. 17 is a waveform diagram showing the FFT window base point control of FIG.

  Modes for carrying out the present invention will become apparent from the following description of the preferred embodiments when read in light of the accompanying drawings. However, the drawings are only for explanation and do not limit the scope of the present invention.

(Configuration of Example 1)
FIG. 1 is a schematic configuration diagram illustrating an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the first embodiment of the present invention.

  This OFDM demodulator is provided in an OFDM receiver, for example, and is an auto gain control unit (automatic gain control unit, hereinafter referred to as “AGC unit”) that controls the gain for an OFDM signal IN that is an OFDM-modulated received signal. ) 11 and an FFT window extraction unit 12 is connected to this output side. The FFT window extraction unit 12 extracts the FFT window S12 from the output signal S11 of the AGC unit 11, and the FFT unit 13 is connected to this output side. The FFT unit 13 performs an FFT process on the extracted FFT window S12, and a pilot symbol extraction unit (SP extraction unit) 14 and an equalizer 31 are connected to the output side. The pilot symbol extraction unit 14 extracts a pilot symbol SP from the output signal S13 of the FFT unit 13, and a transmission path estimator 20 is connected to the output side.

  The transmission path estimator 20 determines a path existence range from the delay profile S21 to generate an optimal transmission path estimation window S23, and obtains a transmission path estimation result S25 by transmission path estimation calculation using the transmission path estimation window S23. IFFT means (for example, IFFT section) 21, path determination / estimation window generation means (for example, path detection section 22 that is path detection means, and estimation window generation section 23 that is estimation window generation means), It comprises transmission path estimation calculation means (for example, a multiplier 24 as multiplication means and an FFT unit 25 as FFT means).

  The IFFT unit 21 is connected to the output side of the pilot symbol extraction unit 14, and performs IFFT processing on the pilot symbol SP extracted by the pilot symbol extraction unit 14 to obtain a delay profile S21. On the side, a path detector 22 and a multiplier 24 are connected. The path detection unit 22 detects the path S22 from the delay profile S21 based on a signal-to-noise ratio (hereinafter referred to as “SNR”), and an estimation window generation unit 23 is connected to the output side. The estimation window generator 23 generates a transmission path estimation window S23 from the detected path S22, and a multiplier 24 is connected to the output side.

  The multiplier 24 multiplies the transmission path estimation window S23 and the delay profile S21 to obtain a multiplication result S24 in order to correct the delay profile S21, and an FFT unit 25 is connected to this output side. . The FFT unit 25 performs an FFT process on the multiplication result S24 to obtain a transmission path estimation result S25, and an equalizer 31 is connected to the output side.

  The equalizer 31 performs an equalization process on the output signal S13 of the FFT unit 13 using the transmission path estimation result S25, and outputs an output signal OUT that can be OFDM demodulated. A demodulator that is not connected is connected.

(Operation of Example 1)
4 is a waveform diagram for explaining the estimation window generating operation of FIG. 1, FIG. 5 is a waveform diagram for explaining delay profile path detection threshold calculation of FIG. 1, and FIG. 6 is a waveform diagram for explaining the maximum power of the delay profile of FIG. FIG. 7 is a waveform diagram for explaining the gain control value in the previous stage of FIG. 1, and FIG. 8 is a waveform diagram for explaining the calculation of the delay profile path window width in FIG. The horizontal axis of these FIGS. 4-8 is time (t), and a vertical axis | shaft is electric power (p).

  When the OFDM demodulator of FIG. 1 receives an OFDM signal that is OFDM-modulated, the gain of the OFDM signal IN is controlled by the AGC unit 11, and an FFT window is extracted from the output signal S 11 of the AGC unit 11. The FFT window S12 is extracted by the unit 12. An FFT process is performed on the extracted FFT window S12 by the FFT unit 13, and an output signal S13 after this process is sent to the equalizer 31 and the pilot symbol extraction unit 14. The equalizer 31 performs an equalization process on the output signal S13 of the FFT unit 13 based on the transmission path estimation result S25, and outputs an output signal OUT that can be OFDM demodulated. The output signal OUT is demodulated by a demodulation unit (not shown) or the like. The SNR detector 32 detects the SNR from the output signal OUT, and provides it to the path detector 22 in the transmission path estimator 20.

  The pilot symbol extraction unit 14 extracts a pilot symbol SP from the output signal S13 of the FFT unit 13 and provides the pilot symbol SP to the IFFT unit 21 in the transmission path estimator 20. The IFFT unit 21 performs an IFFT process on the pilot symbol SP, outputs a delay profile S 21 having a waveform as shown in FIG. 4, and supplies the delay profile S 21 to the path detection unit 22 and the multiplier 24.

  The path detection unit 22 and the estimation window generation unit 23 center the peak time of each path S22 detected by the method described below on the basis of the specific width W calculated by the method described below as the transmission path estimation window S23. To do.

As shown in FIG. 5, the path detection unit 22 obtains a path determination threshold value TH and detects a path S22 exceeding the threshold value TH. The optimum value of the threshold TH for path determination varies depending on the transmission path state. In the first embodiment, threshold generation is performed using the element of the following equation (1).
Threshold TH = basic threshold thresh × threshold coefficient f + threshold offset (1)
However, the basic threshold value thresh; a value proportional to the maximum power of the delay profile S21
Threshold coefficient: Value proportional to the previous gain control value
Threshold offset: Total power of delay profile S21 × SP demodulation result
SNR estimate ÷ IFFT points (known)
Or
Maximum power of delay profile S21 x SP demodulation result
SNR estimate ÷ IFFT points (known)

As shown in FIG. 6, the basic threshold thresh in equation (1) can be obtained from the following equation (2) based on the maximum power max_power of the delay profile S21.
Basic threshold thresh = max_power / K (2)
Where K is a coefficient
For example, judge up to K = 2: 3dB Loss path
K = 4: 6dB Loss path is judged

  The threshold coefficient f in Expression (1) is a value proportional to the gain control value in the previous stage. As shown in FIG. 7A, if the gain control value is the gain of the AGC unit 12, the noise (noise power) increases when the AGC gain is large. Therefore, as shown in FIG. 7B, the threshold level is increased by multiplying the basic threshold thresh by the threshold coefficient f, thereby reducing the influence of noise and facilitating path detection. The path S22 detected by the path detection unit 22 is sent to the estimation window generation unit 23.

  As shown in FIG. 8, the optimum values of the transmission path estimation window widths W1, W2, W3, W4, W5,. Therefore, the estimation window generator 23 of the first embodiment calculates the transmission path estimation window width W based on the element of the following equation (3) to generate the window width.

In Expression (3), the window width coefficient i is a set having 0 to 6 elements among the following 6 elements (1) to (6).
Element (1): Value proportional to or related to the “power value of the relevant path” Element (2); Proportional to the “maximum power of the delay profile and the power ratio of the relevant path”
Is a value related to the magnitude element (3); proportional to the ratio of the total power of the delay profile to the power of the path.
Or value related to magnitude element (4); value proportional to "previous gain control value" or value related to magnitude element (5); inversely proportional to "SNR estimated value based on SP demodulation result" or magnitude related
Value element (6); inversely proportional to “Doppler frequency estimate”
Value

  The window function of the transmission path estimation window S23 of FIG. 4 generated in this way by the estimation window generator 23 is multiplied by the delay profile S21 by the multiplier 24. The multiplication result S24 is subjected to FFT processing by the FFT unit 25 to obtain a transmission path estimation result S25, which is sent to the equalizer 31. The equalizer 31 equalizes the output signal S13 of the FFT unit 13 based on the transmission path estimation result S25.

(Effect of Example 1)
According to the OFDM demodulator having the transmission path estimator 20 of the first embodiment, the delay profile S21 is used to determine the existence range of the path S22 and to dynamically control the window function of the FFT transmission path estimation window S23. As a result, transmission path estimation with the best estimation error according to the path existence range at the time of reception can be realized.

(Configuration of Example 2)
FIG. 9 is a schematic configuration diagram illustrating an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the second embodiment of the present invention, and is an element common to the elements in FIG. 1 illustrating the OFDM demodulator according to the first embodiment. Are marked with a common reference.

  The OFDM demodulator of the second embodiment has a transmission path estimator 20A having a configuration different from that of the transmission path estimator 20 in the OFDM demodulator of the first embodiment. In the transmission path estimator 20A of the second embodiment, instead of the multiplier 24 and the FFT unit 25 in the transmission path estimator 20 of the first embodiment, a finite impulse response (hereinafter “FIR”) is obtained from the window function of the transmission path estimation window S23. An FFT means (for example, FFT unit) 26 for generating a coefficient S26 and an FIR filter 27 for obtaining a transmission path estimation result S27 from the generated FIR coefficient S26 and the delay profile S21 are provided.

  The FFT unit 26 and the FIR filter 27 constitute transmission path estimation calculation means. Other configurations are the same as those of the first embodiment.

(Operation of Example 2)
In the transmission path estimator 20A of the second embodiment, the FIR coefficient S26 is calculated by the FFT section 26 from the window function of the transmission path estimation window S23 obtained by the estimation window generating section 23, and this FIR coefficient S26 is calculated by the FIR filter. 27. In the FIR filter 27, a transmission path estimation result S 27 is obtained by interpolation using the pilot symbol SP extracted by the pilot symbol extraction unit 14 and is given to the equalizer 31. Other operations are the same as those in the first embodiment.

(Effect of Example 2)
According to the OFDM demodulator having the transmission path estimator 20A according to the second embodiment, the delay profile S21 is used, the path detection unit 22 and the estimation window generation unit 23 determine the existence range of the path S22, and the FFT unit. 26 and the FIR filter 27 dynamically control the interpolation interpolation filter coefficients of the LPF, so that the transmission path estimation with the best estimation error according to the path existence range at the time of reception can be realized. In addition, since the FFT unit 26 and the FIR filter 27 are used, the calculation amount can be reduced as compared with the first embodiment.

(Configuration of Example 3)
FIG. 10 is a schematic configuration diagram illustrating an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the third embodiment of the present invention. FIG. 10 illustrates the OFDM demodulator according to the first and second embodiments. Elements common to elements are denoted by common reference numerals.

  The OFDM demodulator of the third embodiment includes a transmission path estimator 20B having a configuration different from that of the transmission path estimator 20A in the OFDM demodulator of the second embodiment. In the transmission path estimator 20B of the third embodiment, instead of the estimation window generator 23 in the transmission path estimator 20A of the second embodiment, an estimation window generator (for example, an estimation window generator) 23B having a different function from this is used. Is provided. The estimation window generation unit 23B generates a line-symmetric transmission path estimation window S23B from the path S22 detected by the path detection means (for example, path detection unit) 22, and supplies it to the FFT unit 26.

  The path detection unit 22 and the estimation window generation unit 23B constitute path determination / estimation window generation means. Other configurations are the same as those of the second embodiment.

(Operation of Example 3)
11 is a waveform diagram showing the estimation window generation of FIG. 10, and FIG. 12 is a waveform diagram showing the transmission path estimation of FIG. In these FIG.11 and FIG.12, a horizontal axis is time (t) and a vertical axis | shaft is electric power (p).

  In the transmission path estimator 20B of the third embodiment, the path detection unit 22B and the estimation window generation unit 23B determine the path existence range from the delay profile S21 obtained by the IFFT unit 21, and determine the optimal transmission path estimation window S23B. Further, the transmission path estimation result S27 is obtained by performing the transmission path estimation calculation using the generated optimal transmission path estimation window S23B by the FFT unit 26 and the FIR filter 27.

  That is, in the path detection unit 22, as shown in FIG. 11, the specific time calculated by the same method as in the first and second embodiments, centering on the peak time of each path detected by the same method as in the first and second embodiments. Let the width be the primary estimation windows A, B, C. In the FFT unit 26, the transmission path estimation window S23B is aligned with the end (right end in FIG. 11) of the primary estimation window C that is the maximum distance (for example, distance C) among the distances A, B, and C from the time 0 position. And By using a line-symmetric window function in the estimation window generation unit 23B, the FIR coefficient generated by the FFT unit 26 is only a real component.

  That is, as shown in FIG. 12, the FFT unit 26 calculates the FIR coefficient S26 from the line-symmetric window function that is the transmission path estimation window S23B obtained by the estimation window generation unit 23B, and supplies the FIR coefficient S26 to the FIR filter 27. The FIR filter 27 obtains the transmission path estimation result S27 by interpolation using the pilot symbol SP extracted by the pilot symbol extraction unit 14 using only the real component FIR coefficient S26, and sends it to the equalizer 31. give. Other operations are the same as those in the first and second embodiments.

(Effect of Example 3)
According to the OFDM demodulator having the transmission path estimator 20B of the third embodiment, the delay profile S21 is used to determine the existence range of the path S22 and to dynamically control the interpolation interpolation filter coefficient of the LPF. Therefore, it is possible to realize a transmission path estimation with the best estimation error according to the path existence range at the time of reception. Furthermore, with respect to the LPF interpolation interpolation estimation method, filter coefficient control is performed for the purpose of reducing the number of arithmetic circuits, so that the calculation amount of the FIR filter 27 can be reduced.

(Configuration of Example 4)
FIG. 13 is a schematic configuration diagram illustrating an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the fourth embodiment of the present invention, and is an element common to the elements in FIG. 10 illustrating the OFDM demodulator according to the third embodiment. Are denoted by common reference numerals.

  The OFDM demodulator of the fourth embodiment includes a transmission path estimator 20C having a configuration different from that of the transmission path estimator 20B in the OFDM demodulator of the third embodiment. In the transmission path estimator 20C of the fourth embodiment, instead of the FFT unit 26 in the transmission path estimator 20B of the third embodiment, selection means (for example, FIR coefficient selection section) 28 having a different configuration is provided. Yes. The FIR coefficient selection unit 28 has FIR coefficients corresponding to a finite number of window width patterns as a table, and is based on the line-symmetric transmission path estimation window S23B generated by the estimation window generation unit 23B. From the width pattern, the FIR coefficient S28 of the minimum pattern having a width equal to or larger than the window function obtained in the third embodiment is selected and given to the FIR filter 27.

  The FIR coefficient selection unit 28 and the FIR filter 27 constitute transmission path estimation calculation means. Other configurations are the same as those of the third embodiment.

(Operation of Example 4)
FIG. 14 is a waveform diagram showing the transmission path estimation of FIG. In FIG. 14, the horizontal axis represents time (t) and the vertical axis represents power (p).

  In the transmission path estimator 20C of the fourth embodiment, the FIR coefficient selection unit 28 has FIR coefficients corresponding to a finite number of window width patterns a, b, c, d, e as a table, as shown in FIG. Based on the line-symmetric transmission path estimation window S23B generated by the estimation window generation unit 23B, the finite number of window width patterns a, b, c, d, e in the table are obtained in the third embodiment. The FIR coefficient S28 having the minimum pattern (for example, pattern c) having a width equal to or greater than the window function thus selected is selected and applied to the FIR filter 27. The FIR filter 27 obtains the transmission path estimation result S27 by interpolation using the pilot symbol SP extracted by the pilot symbol extraction unit 14 using the selected FIR coefficient S28, and provides it to the equalizer 31. Other operations are the same as those in the third embodiment.

(Effect of Example 4)
According to the OFDM demodulator having the transmission path estimator 20C of the fourth embodiment, there are the same effects as the third embodiment. Furthermore, since the FIR coefficient selection unit 28 is configured to select the FIR coefficient stored in the table, the FIR coefficient generation calculation in the FFT unit 26 of the third embodiment can be omitted.

(Configuration of Example 5)
FIG. 15 is a schematic configuration diagram illustrating an OFDM demodulator having a transmission path estimator in OFDM demodulation according to the fifth embodiment of the present invention, and is an element common to the elements in FIG. 10 illustrating the OFDM demodulator according to the third embodiment. Are denoted by common reference numerals.

  In the OFDM demodulator of the fifth embodiment, instead of the FFT unit 13, the pilot symbol extracting unit 14, and the transmission path estimator 20B in the OFDM demodulator of the third embodiment, pilot symbol extracting means (for example, different configuration) (for example, , FFT input generation unit 41 and FFT unit 13D), and transmission path estimator 20D.

  The FFT input generation unit 41 is connected to the output side of the FFT window extraction unit 12 and generates an FFT input signal S41 from the FFT window S12 extracted by the FFT window extraction unit 12 based on the correction information S30. An FFT unit 13D is connected to the output side. The FFT unit 13D performs an FFT process on the FFT input signal S41, outputs a pilot symbol SP, and supplies the pilot symbol SP to the equalizer 31 and the transmission path estimator 20D. The equalizer 31 equalizes the input pilot symbol SP with the transmission path estimation result S27 output from the transmission path estimator 20D, and outputs an output signal OUT that can be OFDM demodulated.

  In the transmission path estimator 20D of the fifth embodiment, in the transmission path estimator 20B of the third embodiment, FFT window base point position control means (for example, a path center deviation calculation section 29, connected to the output side of the path detection section 22). In addition, a filter 30) which is a filter means is newly added. The path center deviation calculation unit 29 calculates a path center deviation S29 from the path S22 detected by the path detection unit 22 and supplies the path center deviation S29 to the filter 30. The filter 30 filters the path center deviation S29 to obtain the correction information S30 and supplies it to the FFT input generation unit 41. The filter 30 includes a digital filter such as an FIR filter. Other configurations are the same as those of the third embodiment.

(Operation of Example 5)
16 and 17 are waveform diagrams showing the FFT window base point position control of FIG. 16 and 17, the horizontal axis represents time (t) and the vertical axis represents power (p).

  In the OFDM demodulator of the fifth embodiment, similarly to the third embodiment, the path detection unit 22 and the estimation window generation unit 23B determine the path existence range from the delay profile S21 obtained by the IFFT unit 21. After generating the optimum transmission path estimation window S23B, the FFT section 26 and the FIR filter 27 perform a transmission path estimation calculation using the generated optimum transmission path estimation window S23B to obtain the transmission path estimation result S27. . Further, in the fifth embodiment, in order to further optimize the transmission path estimation window S23B to be generated, the path center deviation calculation unit 29, the filter 30, and the FFT input generation unit 41 perform an FFT calculation on the input data window. Time base point position control (FFT window base point position control) is performed.

  The plurality of paths S22 detected by the path detection unit 22 exist at different distances A, B, and C from the position of time 0, as shown in FIG. Here, the center of the path S22 of the distance B and the path S22 of the distance C is the center P1 of the plurality of paths S22, but the center P1 of this path is shifted from the position of time 0.

Therefore, the path center deviation calculation unit 29 calculates the position of the center P1 of the peak time of all paths S22, calculates the deviation S29 of the current delay profile S21 with respect to the time 0 position, and provides it to the filter 30. In the example of FIG.
(Deviation between path center P1 and time 0 position) S29 = (distance C−distance B) / 2
It has become. As shown in FIG. 17, the filter 30 performs an appropriate filtering on the value of the calculated deviation S29 between the path center P1 and the time 0 position to obtain an average value, and obtains an average value from the filtered path center P1 and the time 0. The value of the deviation S29 from the position is fed back as the base time position correction information S30 in the FFT input generation unit 41.

  By such correction of the base time position, the path center P2 of the delay profile S21 generated by the IFFT unit 21 after the FFT processing by the FFT unit 13D converges to the time 0 position, so that the transmission path estimation according to the third embodiment is performed. The width of the window S23B can be minimized. Other operations are the same as those in the third embodiment.

(Effect of Example 5)
According to the OFDM demodulator having the transmission path estimator 20D of the fifth embodiment, the same effect as that of the third embodiment can be obtained. Further, the filter coefficient control for the purpose of reducing the arithmetic circuit and the filter control effect can be improved. The optimum FFT window base point position control can be realized.

  Instead of this, the FFT unit 26 in the transmission path estimator 20D may be provided with the FIR coefficient selection unit 28 of the fourth embodiment, whereby the same effect as the fourth embodiment is obtained.

(Modification)
This invention is not limited to the said Examples 1-5, A various utilization form and deformation | transformation are possible.

  For example, the OFDM demodulator configuration other than the transmission path estimators 20 to 20D can be changed to a circuit configuration other than illustrated.

11 AGC unit 12 FFT window extraction unit 13, 13D, 25, 26 FFT unit 14 Pilot symbol extraction unit 20, 20A, 20B, 20C, 20D Transmission path estimator 21 IFFT unit 22 Path detection unit 23, 23B Estimation window generation unit 24 Multiplier 27 FIR filter 28 FIR coefficient selection unit 29 Path center shift calculation unit 30 Filter 31 Equalizer 41 FFT input generation unit

Claims (10)

Inverse fast Fourier transform means for obtaining a delay profile by performing inverse fast Fourier transform on a pilot symbol extracted from an OFDM signal modulated by OFDM;
A path determination / estimation window generating means for determining a path existence range from the delay profile and generating a transmission path estimation window;
A transmission path estimation calculation means for performing a transmission path estimation calculation from the generated transmission path estimation window and the delay profile, and obtaining a transmission path estimation result;
A transmission path estimator characterized by comprising: The path determination / estimation window generation means includes:
Path detection means for detecting a path from the delay profile;
An estimation window generating means for generating the transmission path estimation window from the detected path,
The transmission path estimation calculation means includes
Multiplication means for multiplying the generated transmission path estimation window and the delay profile to calculate a multiplication result;
The transmission path estimator according to claim 1, further comprising: a fast Fourier transform section that performs a fast Fourier transform on the calculated multiplication result to obtain the transmission path estimation result. The path determination / estimation window generation means includes:
Path detection means for detecting a path from the delay profile;
An estimation window generating means for generating the transmission path estimation window from the detected path,
The transmission path estimation calculation means includes
Fast Fourier transform means for performing a fast Fourier transform on the generated transmission path estimation window to generate a finite impulse response coefficient;
2. The transmission path estimator according to claim 1, further comprising a finite impulse response filter that obtains the transmission path estimation result by interpolation using pilot symbols from the generated finite impulse response coefficient and the delay profile. . The path determination / estimation window generation means includes:
Path detection means for detecting a path from the delay profile;
An estimation window generating means for generating the transmission path estimation window line-symmetric from the detected path,
The transmission path estimation calculation means includes
Fast Fourier transform means for performing a fast Fourier transform on the generated line-symmetric transmission path estimation window to generate a finite impulse response coefficient;
2. The channel estimation according to claim 1, further comprising: a finite impulse response filter that obtains the channel estimation result from the generated finite impulse response coefficient and the pilot symbol by interpolation using the pilot symbol. vessel. The path determination / estimation window generation means includes:
Path detection means for detecting a path from the delay profile;
An estimation window generating means for generating the transmission path estimation window line-symmetric from the detected path,
The transmission path estimation calculation means includes
Selection means for selecting a finite impulse response coefficient stored in advance based on the generated line-symmetric transmission path estimation window;
2. The transmission path estimation according to claim 1, further comprising a finite impulse response filter that obtains the transmission path estimation result from the selected finite impulse response coefficient and the pilot symbol by interpolation using the pilot symbol. vessel. Pilot symbol extraction means for generating an input data window for fast Fourier transform from an OFDM signal modulated by OFDM and extracting pilot symbols by fast Fourier transform calculation;
Inverse fast Fourier transform means for performing an inverse fast Fourier transform on the extracted pilot symbols to obtain a delay profile;
A path determination / estimation window generating means for determining a path existence range from the delay profile and generating a transmission path estimation window;
A transmission path estimation calculation means for performing a transmission path estimation calculation from the generated transmission path estimation window and the pilot symbol, and obtaining a transmission path estimation result;
Fast Fourier transform window base position control means for controlling a time base position in the input data window of the fast Fourier transform operation;
A transmission path estimator characterized by comprising: The path determination / estimation window generation means includes:
Path detection means for detecting a path from the delay profile;
An estimation window generating means for generating the transmission path estimation window line-symmetric from the detected path,
The transmission path estimation calculation means includes
Fast Fourier transform means for performing a fast Fourier transform on the generated line-symmetric transmission path estimation window to generate a finite impulse response coefficient;
7. The transmission path estimation according to claim 6, further comprising a finite impulse response filter that obtains the transmission path estimation result from the generated finite impulse response coefficient and the pilot symbol by interpolation using the pilot symbol. vessel. The path determination / estimation window generation means includes:
Path detection means for detecting a path from the delay profile;
An estimation window generating means for generating the transmission path estimation window line-symmetric from the detected path,
The transmission path estimation calculation means includes
Selection means for selecting a finite impulse response coefficient stored in advance based on the generated line-symmetric transmission path estimation window;
7. The transmission path estimation according to claim 6, further comprising a finite impulse response filter that obtains the transmission path estimation result from the selected finite impulse response coefficient and the pilot symbol by interpolation using the pilot symbol. vessel. The fast Fourier transform window base point position control means,
A path center deviation calculating means for obtaining a center position of peak times of all paths detected by the path detecting means and calculating a path center deviation from a time zero position of the current delay profile;
Filter means for filtering the calculated path center shift, and feeding back a value of the shift between the filtered path center and the time zero position as correction information of the time base position in the input data window;
The transmission path estimator according to claim 7 or 8, characterized by comprising: The transmission path estimator according to any one of claims 1 to 9,
An equalizer for equalizing the OFDM signal using the transmission path estimation result obtained by the transmission path estimator;
An OFDM demodulator characterized by comprising:

Images