JP2003218827A - Receiver and transmission channel estimation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a receiver capable of meeting a feedforward circuit capable of performing accurate transmission channel estimation even under the influence of a convolution of a sinc function. <P>SOLUTION: The receiver includes a transmission channel estimator 4 calculating a frequency characteristic by the use of a known symbol included in a reception signal, an IFFT section 6 converting the frequency characteristic to a time response, a waveform shaping section 7 which subtracts the time response to one time before from the converted time response and shapes a waveform using a threshold determined corresponding to the maximum amplitude value of the subtracted result, and a frequency characteristic update section 10 updating the frequency characteristic by adding a difference between the frequency characteristic of the time response after the waveform shaping and the frequency characteristic obtained from the IFFT section 6 to the frequency characteristic of the time response posterior to the waveform shaping. The processing in the IFFT section 6 having an input of the frequency characteristic after the update, the waveform shaping section 7, and the frequency characteristic update section 10 is repeatedly performed. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a receiver for receiving signals transmitted from a parent station, base station, relay station, etc., such as TV broadcast signals and communication signals,
In particular, the FFT (Fast Fourier Transfor
(m: fast Fourier transform) or IFFT (Inverse FFT: inverse fast Fourier transform) is used to eliminate the influence of convolution of the sinc function that occurs when a time response is obtained, and a receiver and a channel estimation method for obtaining a more accurate time response. It is about.

[0002]

2. Description of the Related Art A conventional transmission path estimation method will be described below. Generally, when a wireless device obtains a frequency characteristic by using a pilot signal inserted in advance and obtains a time response by IFFT processing based on the result, F
When the FT size and the signal bandwidth are different, it is equal to multiplication by the rectangular window of the signal bandwidth on the frequency axis. Therefore, on the time axis, the sinc function of the rectangular window corresponding to the signal bandwidth is convoluted with the original time response. In this case, this convolution causes the signal to exist at a timing that does not exist in the original time response.

On the other hand, Japanese Patent Laid-Open No. 2000-34124
Japanese Unexamined Patent Application Publication No. 2 discloses an example in which a threshold value is set and the time response at a timing of a level below the threshold value is forcibly set to 0 to eliminate the influence of the sinc function.

Here, the above publication will be briefly described. FIG. 34 is a diagram showing a configuration of a repeater in terrestrial digital broadcasting, and FIG. 35 is a diagram showing a configuration of a frequency band expansion circuit in the repeater. In FIG. 34, 10
Reference numeral 1 is a frequency band expansion circuit, and in FIG.
Reference numeral 1 is an absolute value calculation circuit, 112 is a maximum value detection circuit, 113 is a threshold value defining circuit, 114 is a comparator, and 115 is a gate circuit.

[0005] In the terrestrial digital broadcasting repeater, a replica of the interference signal is generated by estimating the transmission path of the sneak interference signal from the transmission antenna to the reception antenna, and the replica is subtracted from the reception signal to obtain the interference signal. Suppress the effects of. At this time, the inverse Fourier transform circuit
Inverse Fourier transform is performed on the frequency characteristic, and the time response is obtained as a result. The frequency band expansion circuit 101 removes the influence of the convolution of the sinc function that occurs in that case, and the time response after the removal is applied to the coefficient update circuit. Sending out.

Next, the frequency band expansion circuit 101 will be described with reference to FIG. The signal output from the inverse Fourier transform circuit is output from the frequency band expansion circuit 101 at the same time as the gate circuit.
Is sent to the absolute value calculation circuit 111 in which the absolute value of the time response in the observation section is obtained.

The maximum value detection circuit 112 finds the maximum absolute value within the observation section, and the threshold value defining circuit 113
The threshold value is determined based on the maximum value. Comparator 11
In 4, the output signal of the absolute value calculation circuit 111 is compared with the threshold value, and the signal at the timing below the threshold value is forcibly set to 0. This can eliminate the influence of the sinc function.

[0008]

However, the above
In the conventional transmission path estimation method described in Japanese Patent Application Laid-Open No. 2000-341242, a method is adopted in which the influence of the sinc function is removed by updating the error in the feedback circuit. Therefore, when this is applied to a feedforward circuit, there is a problem that if the level of the signal that should originally exist is low, the signal is removed and the characteristics deteriorate.

The present invention has been made in view of the above, and can be applied to a feedforward circuit.
An object of the present invention is to obtain a receiving apparatus and a channel estimation method that can perform channel estimation more accurately even when there is an influence of convolution of an inc function.

[0010]

[Means for Solving the Problems]
In order to achieve the object, in the receiving device according to the present invention, the frequency characteristic calculating means for calculating the frequency characteristic by using the known symbols included in the received signal (the FFT unit 1, the pilot signal of the embodiment described later, (Corresponding to the extraction unit 2, the pilot signal generation unit 3, and the transmission path estimation unit 4), and frequency / time conversion means (IF) for converting the frequency characteristic into a time response.
(Corresponding to the FT unit 6) and the previous time response from the above time response, waveform shaping is performed using a threshold value determined according to the maximum amplitude value of the subtraction result, and the time after waveform shaping is performed. Updating means for updating the frequency characteristic in the signal band by adding the difference between the frequency characteristic of the response and the frequency characteristic of the frequency characteristic calculating means to the frequency characteristic of the time response after the waveform shaping (the waveform shaping section 7). Corresponding to the determination unit 8, the FFT unit 9, and the frequency characteristic updating unit 10), and the frequency / time conversion unit converts the frequency characteristic output from the frequency characteristic calculation unit into a time response, and It is characterized in that the updated frequency characteristic is converted into a time response.

In the receiving device according to the next invention,
The frequency characteristic calculation means for calculating frequency characteristics using known symbols included in the received signal, the frequency / time conversion means for converting the frequency characteristics into a time response, and the maximum amplitude value of the time response are determined. Performing waveform shaping using the threshold value, the difference between the frequency characteristic of the time response after waveform shaping and the frequency characteristic by the frequency characteristic calculating means, and an updating means for updating the frequency characteristic in the signal band,
The frequency / time conversion means may convert the frequency characteristic output from the frequency characteristic calculation means into a time response, and then convert the updated frequency characteristic into a time response.

In a receiving device according to the next invention,
Frequency characteristic calculation means for calculating frequency characteristics using known symbols included in the received signal, frequency / time conversion means for converting the frequency characteristics into a time response, and an initial waveform (zero value for the first time) from the time response. Subtracting, performing a waveform shaping using a threshold value determined according to the maximum amplitude value of the subtraction result, the frequency response of the time response after waveform shaping, updating means for updating the frequency characteristics in the signal band, Equipped with
The frequency / time conversion means may convert the frequency characteristic output from the frequency characteristic calculation means into a time response, and then convert the updated frequency characteristic into a time response.

In the receiving device according to the next invention, the updating means re-updates the frequency characteristic based on the time response obtained by converting the updated frequency characteristic, and the frequency / time converting means and the updating means are used. It is characterized in that the signal processing is repeatedly executed a predetermined number of times.

In the receiving device according to the next invention, when the known symbol is repeatedly inserted at a constant cycle on the frequency axis and the time axis, the frequency characteristic calculating means calculates the frequency characteristic in units of a plurality of symbols. It is characterized by

In the receiving device according to the next invention,
An interpolation interpolating means for interpolating the frequency characteristic of a sample (carrier) having no value is provided, and the frequency / time converting means converts the frequency characteristic after the interpolation to a time response.

In the receiving device according to the next invention,
An integrating means for averaging the frequency characteristics over time is provided.

In the receiving device according to the next invention, the updating means may use the maximum amplitude value of the time response instead of the maximum amplitude value.
A threshold value is determined according to the power of the time response.

In the transmission path estimation method according to the next invention, a frequency characteristic calculation step of calculating a frequency characteristic using a known symbol included in a received signal, and a frequency / time conversion of the frequency characteristic into a time response. The conversion step and the previous time response are subtracted from the time response, waveform shaping is performed using a threshold value determined according to the maximum amplitude value of the subtraction result, and the frequency characteristic of the time response after waveform shaping is performed. And an updating step of updating the frequency characteristic in the signal band by adding the difference between the frequency characteristic by the frequency characteristic calculating means and the frequency characteristic of the time response after the waveform shaping, the frequency / time In the converting step, after converting the frequency characteristic output in the frequency characteristic calculating step into a time response, the updated frequency characteristic is converted into a time response. It is characterized in.

In the transmission path estimation method according to the next invention, a frequency characteristic calculation step of calculating a frequency characteristic by using a known symbol included in a received signal, and a frequency / time conversion of the frequency characteristic into a time response. Converting step, performing waveform shaping using a threshold value determined according to the maximum amplitude value of the time response, by the difference between the frequency characteristic of the time response after waveform shaping and the frequency characteristic by the frequency characteristic calculating means, An updating step of updating the frequency characteristic within the signal band, wherein the frequency / time converting step converts the frequency characteristic output by the frequency characteristic calculating step into a time response, and then updates the updated frequency characteristic. It is characterized by conversion into a time response.

In the transmission path estimation method according to the next invention, a frequency characteristic calculating step of calculating a frequency characteristic using known symbols included in a received signal, and a frequency / time conversion of the frequency characteristic into a time response. The conversion step and the initial waveform (0 value at the first time) are subtracted from the time response, waveform shaping is performed using a threshold value determined according to the maximum amplitude value of the subtraction result, and the time response of the time response after the waveform shaping is performed. An updating step of updating the frequency characteristic within the signal band by the frequency characteristic, wherein the frequency / time converting step converts the frequency characteristic output by the frequency characteristic calculating step into a time response, and then performs the updating. The frequency characteristic of is converted into a time response.

In the transmission path estimation method according to the next invention, in the updating step, the frequency characteristic is re-updated based on the time response obtained by converting the updated frequency characteristic, and the frequency / time conversion step and the updating step are performed. It is characterized in that the signal processing is repeatedly executed a predetermined number of times.

In the transmission path estimation method according to the next invention, when the known symbol is repeatedly inserted at a constant cycle on the frequency axis and the time axis, the frequency characteristic is calculated in units of a plurality of symbols in the frequency characteristic calculation step. It is characterized by calculating.

A transmission path estimation method according to the next invention includes an interpolation step of interpolating the frequency characteristic of a sample (carrier) having no value, and in the frequency / time conversion step, the interpolation step is performed. It is characterized in that the frequency characteristic after interpolation is converted into a time response.

The transmission path estimation method according to the next invention is characterized by including an integration step of averaging the frequency characteristics in time.

In the transmission path estimation method according to the next invention, in the updating step, the threshold value is determined according to the power of the time response instead of the maximum amplitude value of the time response. .

[0026]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of a receiving apparatus and a transmission path estimating method according to the present invention will be described below in detail with reference to the drawings. The present invention is not limited to this embodiment.

Embodiment 1. 1 is a diagram showing a configuration of a first embodiment of a receiving apparatus according to the present invention. In FIG. 1, reference numeral 1 is an FFT unit for converting a received signal from a time axis signal to a frequency axis signal, and 2 is a known amplitude / phase characteristic for each subcarrier, which is inserted according to a predetermined rule. A pilot signal extraction unit for extracting a pilot signal, 3 is a pilot signal generation unit, and 4 is a frequency characteristic obtained by dividing the output signal of the pilot signal extraction unit 2 by the output signal of the pilot signal generation unit 3. Reference numeral 5 is a transmission path estimation unit, 5 is a frequency characteristic memory unit that stores the output signal of the transmission path estimation unit 4, and 6 is an IFFT unit that converts the output signal of the frequency characteristic memory unit 5 from a frequency axis signal to a time axis signal. 7 is a waveform shaping unit that shapes the output signal of the IFFT unit 6, 8 is a determination unit that determines based on the output signal of the waveform shaping unit 7, and 9 is a waveform shaping unit. An FFT unit for converting the output signal of 7 from the time axis signal to the frequency axis signal, and 10 is a frequency characteristic updating unit for calculating the updated frequency characteristic from the output signal of the frequency characteristic memory unit 5 and the output signal of the FFT unit 9. Is.

The output control signal of the judging section 8 is IFF.
The switch a for selecting whether the signal input to the T unit 6 is output to the frequency characteristic memory unit 5 or the frequency characteristic updating unit 10 is controlled, and the output destination of the waveform shaping unit 7 is changed to
A switch b for selecting whether to use the FFT unit 9 or an external output as an output time response is controlled.

First, the FFT section 1 converts the received signal from the time axis signal to the frequency axis signal, and inputs the conversion result to the pilot signal extraction section 2. Pilot signal extraction unit 2
Then, the pilot signal is extracted according to a predetermined rule. The transmission path estimation unit 4 calculates the estimated transmission path by dividing the output signal of the pilot signal extraction unit 2 by the output signal of the pilot signal generation unit 3. When this is expressed by an equation, H (z) = Y (z) / X (z) (1). Here, X (z) represents the frequency characteristic of the transmitted signal, Y (z) represents the frequency characteristic of the received signal, and H (z)
(Z) represents the frequency characteristic of the transmission line response.

The frequency characteristic obtained by the transmission path estimation unit 4 is stored in the frequency characteristic memory unit 5 and at the same time IF
The signal is sent to the FT section 6, and the IFFT section 6 converts the frequency axis signal into a time axis signal and sends the converted signal to the waveform shaping section 7.

Now, the operation of the waveform shaping section 7 will be described in detail. FIG. 2 is a diagram showing the configuration of the waveform shaping section 7. In FIG. 2, 201 is a waveform subtraction unit, 202 is an absolute value calculation unit, 203 is a maximum value detection unit, and 204 is a maximum value detection unit.
Is a threshold value defining unit, 205 is a comparing unit, and 20
6 is a gate circuit, 207 is a waveform adding unit, and 2
Reference numeral 08 is a memory unit. The output signal of the IFFT unit 6 is l (t, i). However, t represents time t, and i
Represents the i-th time waveform at time t.

The output signal l (t, i) is supplied to the waveform subtraction unit 20.
1 and the difference signal from the previous time response m (t, i−1) stored in the memory unit 208 is obtained. m (t, i) = 1 (t, i) -m (t, i-1) (2)

The difference signal m (t, i) is supplied to the gate circuit 20.
6 is sent to the absolute value calculation unit 202 at the same time, and the absolute value calculation unit 202 obtains the absolute value of the time response in each sample. The maximum value detection unit 203 calculates the maximum value in the observation section based on the absolute value, and the threshold value definition unit 20
In 4, the threshold value th1 for waveform shaping is determined based on the maximum value, and the threshold value is transmitted to the comparison unit 205 and the determination unit 8.

The comparison unit 205 compares the output signal of the absolute value calculation unit 202 with the threshold value th1 determined by the threshold value defining unit 204, and sends the comparison result to the gate circuit 206.

The gate circuit 206 performs processing based on the comparison result of the comparison section 205. That is, when the absolute value of the time response of the sample is larger than the threshold value th1, it is output as it is, and when it is smaller, it is forcibly output as 0.

The output signal of the gate circuit 206 is transmitted to the waveform addition section 207, and the waveform addition section 207 adds the output signal of the gate circuit 206 and the output signal of the memory section 208, and the addition result is output time response. Alternatively, it is output as an output signal to the FFT unit 9. At the same time, the addition result is
It is stored in the memory unit 208 as a new memory signal.
The output signal of the waveform addition unit 207 can be expressed by the following equation (3). m (t, i) = m (t, i) + m (t, i-1) (3)

Further, the output signal t of the threshold value defining unit 204
h1 is transmitted to the determination unit 8, and the determination unit 8 determines whether th1 is larger or smaller than a preset threshold TH. For example, when th1 is larger than TH, the switch a is connected to the frequency characteristic updating unit 10 and the switch b is connected to the FFT unit 9. Also t
When h1 is smaller than TH, the switch a is connected to the frequency characteristic memory unit 5, and the switch b is connected to the waveform shaping unit 7.
Is connected so that the output signal of is output as the output time response.

When the output signal of the waveform shaping section 7 is transmitted to the FFT section 9, the FFT section 9 converts the output signal of the waveform shaping section 7 from a time axis signal to a frequency axis signal,
The conversion result is transmitted to the frequency characteristic updating unit 10.

Next, the operation of the frequency characteristic updating section 10 will be described in detail. FIG. 3 is a diagram showing a configuration of the frequency characteristic updating unit 10, and 211 is a frequency characteristic difference calculating unit,
Reference numeral 212 is a frequency characteristic difference addition unit.

In the frequency characteristic difference calculation unit 211, the FFT
A difference within the signal band between the frequency characteristic of the output of the unit 9 and the frequency characteristic stored in the frequency characteristic memory unit 5 is obtained, and the frequency characteristic difference addition unit 212 adds the difference to the frequency characteristic of the output of the FFT unit 9. The frequency characteristic in the signal band is updated by and is sent to the IFFT unit 6 again. When this is expressed by an equation, H '(z) = H' (z) + (H (z) -H '(z)) (4). Note that H ′ (z) is the FFT unit 9 after waveform shaping.
It is the frequency characteristic of the output.

Thereafter, the IFFT section 6 converts the output signal of the frequency characteristic updating section 10 from the frequency axis signal to the time axis signal again, and the waveform shaping section 7 shapes the output signal of the IFFT section 6 and further, In the FFT unit 9, the output signal of the waveform shaping unit 7 is converted from the time axis signal to the frequency axis signal, and such loop processing is repeatedly performed.

Then, the determination unit 8 causes the waveform shaping unit 7
When it is determined that the output signal th1 of the threshold value defining unit 204 in the above is smaller than the threshold value TH, the switch a
And the switch b is controlled, and the output signal of the waveform shaping section 7 is output as an output time response. At the same time, the value in the memory unit 208 in the waveform shaping unit 7 is reset and becomes the initial value 0.

After that, when a new frequency characteristic is output from the transmission path estimation unit 4 and stored in the frequency characteristic memory unit 5, the waveform shaping operation is performed again.

Next, the above operation will be specifically described with reference to the drawings. FIG. 4 is a diagram showing a two-wave model of time response. Further, FIG. 5 is a diagram showing a frequency characteristic of the output of the transmission path estimation unit 4, and this frequency characteristic is stored in the frequency characteristic memory unit 5. At this time, the time response which is converted from the frequency axis signal to the time axis signal by the IFFT section 6 and is input to the waveform shaping section 7 can be expressed as shown in FIG.

Note that in FIG. 5, since the signal bandwidth and the FFT size are different, it can be considered as equivalent to multiplication by a rectangular window on the frequency axis. Therefore, in the time response of FIG. A waveform obtained by convolving the corresponding sinc function is output. As a result, although there should originally be only an impulse-shaped waveform, a waveform corresponding to the side lobe of the sinc function is output to other samples.

Further, the absolute value calculation unit 202 obtains the amplitude value of each sample point in the observation range of this time response,
The maximum value detection unit 203 obtains the maximum value of each amplitude value.
Then, the threshold value defining unit 204 determines the threshold value th1. FIG. 7 shows the comparison unit 205 and the gate circuit 20.
6 is a diagram showing a waveform in which a time response of a threshold value th1 or less is forcibly set to 0 by 6. FIG. Here, only the first wave that exceeds the threshold value is output.

Since the FFT size and the signal bandwidth of the threshold value th1 determined by the threshold value defining unit 204 are known, and the shape of the sinc function to be convolved is also known, the threshold value th1 is based on them. Set. Specifically, considering that the maximum sidelobe peak in the sinc function is smaller than the mainlobe peak by about 13 dB, the threshold is further determined.
Considering the relationship between the size and the signal bandwidth, and that there is a sampling point whose level exceeds the maximum sidelobe peak due to noise, etc., the maximum value and 13 dB from that value
It is desirable to set it only between the down level.

Then, the judgment section 8 sets the threshold value defining section 20.
When it is determined that the output signal th1 of No. 4 is larger than TH, the waveform-shaped output signal of the gate circuit 206 is input to the memory unit 208. At the same time, the gate circuit 206
The output signal of is converted from the time axis signal into the frequency axis signal by the FFT unit 9 and transmitted to the frequency characteristic updating unit 10. FIG. 8 is a diagram showing a frequency axis signal output from the FFT unit 9.

As is clear from FIG. 8, by removing the influence of the convolution of the sinc function (removing the side lobes), the frequency characteristic of only one wave can be obtained. In addition, the frequency characteristic outside the signal band is also extrapolated.

However, since the initial value 0 is set in the memory unit 208 in the waveform shaping unit 7 at the time of the first waveform shaping when the waveform of the output signal of the frequency characteristic memory unit 5 is shaped, Subtractor 201, waveform adder 207
In, the input signal is output as it is.

Further, the frequency characteristic updating section 10 which has received the frequency axis signal output from the FFT section 9 obtains the difference within the signal band between the frequency axis signal and the signal stored in the frequency characteristic memory section 5. Then, the difference is added to the frequency axis signal output from the FFT unit 9 to update the frequency characteristic within the signal band. As a result, the signal component of the second wave deleted by the first waveform shaping is included in the updated frequency characteristic. FIG. 9 is a diagram showing a frequency characteristic in which the signal component of the second wave deleted by the first waveform shaping is updated.

The IFFT section 6 again converts the updated frequency characteristic from the frequency axis signal into the time axis signal, and sends it to the waveform shaping section 7. FIG. 10 is a diagram showing a time axis signal output from the IFFT unit 6.

In the waveform subtraction unit 201 in the waveform shaping unit 7,
The difference from the first waveform shaping signal stored in the memory unit 208 is calculated. FIG. 11 is a diagram showing a differential signal output from the waveform subtraction unit 201.

Here again, the absolute value calculating unit 202 obtains the amplitude value of each sample point in the observation range of this time response, and the maximum value detecting unit 203 obtains the maximum value of the amplitude values.
Then, the threshold value defining unit 204 sets the new threshold value th.
Determine one. FIG. 12 is a diagram showing a waveform in which the time response of the threshold value th1 or less is forcibly set to 0 in the comparison unit 205 and the gate circuit 206.

Then, the waveform addition section 207 adds the output signal of the gate circuit 206 and the output signal of the memory section 208. FIG. 13 is a diagram showing a waveform output from the waveform addition unit 207. At this point, the ideal waveforms of the first and second waves are obtained.

In the present embodiment, the above processing is performed with the threshold value t.
This is repeated until h1 becomes smaller than the threshold value TH in the determination unit 8, and waveform shaping is performed.

As described above, in the present embodiment, waveform shaping is performed using the threshold value determined according to the maximum amplitude value in the time response, the frequency characteristic of the time response is obtained, and the original frequency characteristic is obtained. Since it is configured to perform the waveform shaping of the time response again by updating with the difference between and, it is possible to perform the transmission path estimation more accurately even with the feedforward type.

The above-described operation is applicable to all receivers in which the transmission channel is estimated by FFT or IFFT and the FFT size and the signal bandwidth are different, and the multi-carrier transmission system, single carrier transmission system, etc. It is not limited by the transmission method.

Embodiment 2. Next, the transmission path estimation method according to the second embodiment will be described. The configuration of the receiving device is the same as that of the above-described first embodiment. Here, only the operation different from that of the first embodiment will be described.

FIG. 14 is a diagram showing pilot signals in terrestrial digital broadcasting. The pilot signal may be inserted in a constant cycle as shown in FIG. In this case, the pilot signal extraction unit 2 extracts the pilot signal according to the regularity, and the transmission path estimation unit 4
The frequency characteristic is obtained based on the extracted pilot signal.

In FIG. 14, 12 in the frequency axis direction.
For each carrier, every 4 symbols in the time axis direction,
A pilot symbol is inserted in each. Therefore, for example, if the frequency characteristic is obtained in units of 4 symbols, the pilot signal will be inserted every 3 carriers in the frequency axis direction.

When the above-described operation of the first embodiment is performed using the pilot signal, the output signal of IFFT section 6 has a time response as shown below. FIG. 15 shows the IF
It is a figure which shows the time-axis signal of FT section 6 output.

In this case, since frequency characteristics exist for every three carriers, aliasing occurs in the time response, and the time response is repeated at intervals of 1/3 of the FFT size. Further, in this case, the FFT size has a width of Nth power of 2 and not a multiple of 3, so that the folded waveform exists during the sampling timing. And the second and third folded waveforms are
It will be output as a waveform with a shifted timing.
However, the first waveform (t = 0 to FFT size /
3) is output at the sampling timing, so if the range is set as the observation range,
The operation shown in can be performed.

In the present embodiment, since the frequency characteristic memory unit 5 stores the frequency characteristic having every third value, the frequency characteristic updating unit 10 also updates the three frequency characteristics. It needs to be calculated every other time. As described above, in the present embodiment, it is possible to support an OFDM (Orthogonal Frequency Division Multiplexing) signal such as digital terrestrial broadcasting.

Embodiment 3. Next, the transmission path estimation method according to the third embodiment will be described. The configuration of the receiving device is the same as that of the above-described first embodiment. In the receiving device according to the present embodiment, the H / W scale is reduced by simplifying the waveform shaping unit 7 and the frequency characteristic updating unit 10 of the receiving device according to the first embodiment described above.

FIG. 16 is a diagram showing the configuration of the waveform shaping section 7. The waveform shaping unit 7 of FIG. 16 does not include the waveform subtraction unit 201 of FIG. 2, and the memory unit 208 includes the waveform addition unit 2 of FIG.
Exchanges signals with 07. FIG. 17 is a diagram showing the configuration of the frequency characteristic updating unit 10. In FIG. 17, the frequency characteristic difference addition unit 212 in FIG. 3 does not exist.
In addition, in FIG. 16 and FIG. 17, FIG.
The same components as those in FIG. 3 are designated by the same reference numerals and the description thereof will be omitted. Here, only the operation different from that of the first embodiment will be described. Further, here, for convenience of description, the two-wave model shown in FIG. 4 is used.

For example, the signal stored in the frequency characteristic memory section 5 is converted from the frequency axis signal into the time axis signal by the IFFT section 6, and the time axis signal is transmitted to the waveform shaping section 7. The time axis signal at this time can be expressed as shown in FIG.

In the waveform shaping section 7, the processing of the waveform subtraction section 201 is deleted from the operation of the first embodiment described above, that is, the output signal of the IFFT section 6 is directly converted into the absolute value calculation section 20.
2 and thereafter, in the same procedure as in the first embodiment, si
Eliminate the effects of convolution of the nc function. The time axis signal at this time can be expressed as shown in FIG.

After that, the threshold value defining unit 20 in the judging unit 8
When it is judged that the output of 4 is larger than the threshold value TH,
The output signal of the waveform shaping section 7 is converted into a frequency axis signal by the FFT section 9, and the frequency axis signal is transmitted to the frequency characteristic updating section 10. The frequency axis signal at this time is shown in FIG.
Can be expressed as

The frequency characteristic updating unit 10 outputs the difference within the signal band between the frequency characteristic from the frequency characteristic memory unit 5 and the new frequency characteristic output from the FFT unit 9. Figure 18
6 is a diagram showing an output signal of a frequency characteristic difference calculation unit 211. FIG.

The output signal of the frequency characteristic updating unit 10 is IF
The FT unit 6 converts the frequency axis signal into a time axis signal, and transmits the time axis signal to the waveform shaping section 7. FIG. 19 shows a time-axis signal output from the IFFT unit 6, and the waveform here is a waveform in which the signal component truncated by the previous waveform shaping is affected by the convolution of the sinc function.

Further, inside the waveform shaping section 7, the maximum value in the observation section is obtained to determine the threshold value, and the sample below the threshold value is forcibly set to 0, and the result is added to the waveform addition section 20.
Send to 7. FIG. 20 is a diagram showing a time-axis signal in the case where samples below the threshold value are forcibly set to 0.

The waveform adding section 207 adds the output signal of the memory section 208 and the output signal of the gate circuit 206 to update the time response. FIG. 21 shows the waveform addition unit 207.
It is a figure which shows the time response in. Hereinafter, the first embodiment
The same process as is repeated.

As described above, in the present embodiment, the same effect as that of the first embodiment described above can be obtained, and the H / W scale can be further reduced.

Fourth Embodiment Next, the transmission path estimation method according to the fourth embodiment will be described. FIG. 22 is a diagram showing the configuration of the receiving device according to the fourth embodiment of the present invention. In the third embodiment described above, the time axis signal is updated by calculating the difference between the frequency axis signals, but in the present embodiment, the frequency characteristic memory unit 5 to the frequency characteristic updating unit 10 is updated. The configuration is such that there is no signal line, the difference between time-axis signals is taken, and the signal obtained by performing waveform shaping on the difference signal is updated.

FIG. 23 is a diagram showing the configuration of the waveform shaping section 7. In FIG. 23, 209 is an initial waveform memory unit. The difference from the waveform shaping section 7 of FIG. 2 is that an initial waveform memory section 209 is newly inserted. The same components as those in FIG. 2 are designated by the same reference numerals, and the description thereof will be omitted. Further, here, for convenience of description, the two-wave model shown in FIG. 4 is used.

For example, the signal stored in the frequency characteristic memory section 5 is converted from the frequency axis signal into the time axis signal by the IFFT section 6 and transmitted to the waveform shaping section 7. The time axis signal at this time can be expressed as shown in FIG.

In the waveform shaping section 7, the signal stored in the initial waveform memory section 209 by the waveform subtraction section 201 and the I
The difference from the output signal from the FFT unit 6 is taken. In addition,
The new frequency characteristic is input to the frequency characteristic memory unit 5,
When the output signal of the IFFT unit 6 is input for the first time, since the 0 value is stored in the initial waveform memory unit 209, the input signal level of the waveform subtraction unit 201 is output as it is. Other processes are the same as those in the first embodiment described above. The initial waveform memory unit 209 is
When the output signal of the FFT unit 6 is input for the first time, the output of the waveform subtraction unit 201 is controlled to be input.

When the judging section 8 judges that the output of the threshold value defining section 204 is larger than the threshold value TH, the FFT section 9 outputs the output signal of the waveform shaping section 7 from the time axis signal to the frequency axis signal. And is input to the frequency characteristic updating unit 10. The frequency axis signal at this time can be expressed as shown in FIG.

The frequency characteristic updating section 10 extracts the frequency axis signal from the FFT section 9 within the signal band and sets it as a new frequency axis signal. FIG. 24 is a diagram showing frequency characteristics in the frequency characteristic updating unit 10.

The updated frequency axis signal is transferred to the IFFT unit 6
The frequency axis signal is converted into a time axis signal by and is transmitted to the waveform shaping section 7. FIG. 25 is a diagram showing a time axis signal output from the IFFT unit 6.

Since the time axis signal output from the IFFT section 6 is a signal obtained by convolving the sinc function with the signal extracted by the first processing of the waveform shaping section 7, the initial waveform memory section in the waveform subtraction section 201. The time axis signal of FIG. 25 is subtracted from the time axis signal stored in 209, and the signal component truncated in the first processing is output. FIG. 26 is a diagram showing a time axis signal output from the waveform subtraction unit 201.

In the comparison unit 205 and the gate circuit 206, the signal component of each sample below the threshold value newly determined by the same processing as the first time is forcibly set to 0, and the result is transmitted to the waveform addition unit 207. To do. FIG. 27 is a diagram showing a time axis signal output from the gate circuit 206.

The waveform adding section 207 adds the time axis signal of FIG. 27 and the signal stored in the memory section 208.
Since the signal stored in the memory unit 208 is the signal shown in FIG. 7, the waveform adding unit 207 outputs the signal as shown in FIG. The threshold value T of the determination unit 8
When it is determined that the threshold value th1 in waveform shaping is smaller than H, the initial waveform memory unit 209 is reset to the initial value 0.

As described above, in the present embodiment, the difference between the time-axis signals is obtained, the signal obtained by performing waveform shaping on the difference signal is updated, and the time-response waveform is shaped again. Even with the feedforward type, more accurate channel estimation can be performed.

Embodiment 5. Next, the transmission path estimation method according to the fifth embodiment will be described. FIG. 29 is a diagram showing the configuration of the fifth embodiment of the receiving apparatus according to the present invention.
1 is an interpolation processing unit.

In the third embodiment described above, the transmission channel is estimated by using the pilot signal inserted according to the predetermined rule, and the frequency characteristic is calculated. However, in the present embodiment, the transmission channel estimation unit is used. The output signal of No. 4 is input to the interpolation processing unit 11, and the frequency characteristic is obtained by interpolating the frequency characteristic of the sample (carrier) having no value.

The interpolation interpolation here is, for example, 1
Secondary interpolation, quadratic interpolation, and interpolation interpolation by convolution of the ideal sinc function on the frequency axis are used.

As a result, in this embodiment, the same effect as that of the third embodiment can be obtained, and moreover, the transmission path can be estimated more accurately.

Sixth Embodiment Next, the transmission path estimation method according to the sixth embodiment will be described. FIG. 30 is a diagram showing the configuration of the sixth embodiment of the receiving apparatus according to the present invention.
Reference numeral 2 is an integration processing unit.

In the first to fifth embodiments described above, the transmission path estimation is performed using the frequency characteristics of the output of the transmission path estimation unit 4 and the output of the interpolation processing unit 11. However, in the present embodiment, the transmission path estimation is performed. The integration processing unit 12 temporally averages the output signal of the estimation unit 4 and the output signal of the interpolation processing unit 11 to obtain the frequency characteristic.

As a result, in this embodiment, the same effects as those of the above-described first to fifth embodiments can be obtained, the influence of noise can be further reduced, and the transmission path can be estimated more accurately.

As an averaging method, for example,
The configuration may be performed by a moving average using an FIR filter as shown in FIG. 31, or may be performed by a configuration using an IIR filter as shown in FIG.

Seventh Embodiment Next, the transmission path estimation method according to the seventh embodiment will be described. The configuration of the receiving device is the same as in the above-described first to sixth embodiments.

FIG. 33 is a diagram showing the configuration of the waveform shaping section 7, and 210 is a power calculation section. Note that, in FIG. 33, the same components as those in FIGS. 2, 16 and 23 described above are denoted by the same reference numerals and the description thereof will be omitted. Here, only the operation different from that of the first embodiment will be described.

In the first to sixth embodiments described above, the absolute value (amplitude) of the time response is used as the comparison target at the time of waveform shaping, but in the present embodiment, the power value (squared) of the time response is used. Value) is calculated, and the waveform is shaped by comparing the power value with a threshold value determined based on this power value. When the power value is calculated, it is not necessary to calculate the square root used when calculating the absolute value (amplitude), and the H / W circuit can be simply configured accordingly.

As a result, in this embodiment, the same effects as those of the above-described first to sixth embodiments can be obtained, and further, the H / W scale can be reduced as compared with the case where the absolute value is obtained and compared.

[0098]

As described above, according to the present invention, waveform shaping is performed by using the threshold value determined according to the maximum amplitude value in the time response, and the frequency characteristic of the time response is obtained. The frequency characteristic in the signal band is updated based on the difference from the frequency characteristic of, and the updated frequency characteristic is converted into a time response. As a result, there is an effect that it is possible to obtain a receiving device capable of accurately estimating a transmission path even if it is a feedforward type.

According to the next invention, it is possible to further obtain a receiving device capable of reducing the H / W scale.
Has the effect.

According to the next invention, the difference between the time-axis signals is obtained, the signal obtained by performing waveform shaping on the difference signal is updated, and the updated frequency characteristic is converted into the time response. As a result, even if it is a feedforward type,
It is possible to obtain a receiving device capable of accurately estimating a transmission path.

According to the next invention, the signal processing by the frequency / time converting means and the updating means is repeatedly executed a predetermined number of times. As a result, there is an effect that it is possible to obtain a receiving device that can perform channel estimation more accurately.

According to the next invention, there is an effect that it is possible to obtain a receiver capable of supporting an OFDM (Orthogonal Frequency Division Multiplex) signal such as a terrestrial digital broadcast.

According to the next invention, it is further possible to obtain the receiving device capable of accurately estimating the transmission path.

According to the next invention, since the influence of noise can be reduced, it is possible to obtain a receiving apparatus capable of estimating the transmission path with higher accuracy.

According to the next invention, it is possible to obtain a receiving apparatus capable of reducing the H / W scale as compared with the case where the absolute value is obtained and compared.

According to the next invention, waveform shaping is performed by using the threshold value determined according to the maximum amplitude value in the time response, the frequency characteristic of the time response is obtained, and the difference from the original frequency characteristic is calculated. Update the frequency characteristics within the signal band,
Since the updated frequency characteristic is converted into the time response, there is an effect that the transmission path can be estimated more accurately even in the feedforward type.

According to the next invention, the H / W scale can be further reduced.

According to the next invention, the difference between the time-axis signals is calculated, the signal obtained by performing waveform shaping on the difference signal is updated, and the updated frequency characteristic is converted into the time response. Even with the forward type, there is an effect that the transmission path can be estimated more accurately.

According to the next invention, since the signal processing by the frequency / time conversion step and the updating step is repeatedly executed a predetermined number of times, there is an effect that the transmission path can be estimated more accurately.

According to the next invention, it is possible to deal with an OFDM (Orthogonal Frequency Division Multiplex) signal such as a terrestrial digital broadcast.

According to the next invention, the effect that the transmission path can be accurately estimated can be obtained.

According to the next invention, since the influence of noise can be reduced, there is an effect that the transmission path can be estimated more accurately.

According to the next invention, there is an effect that the H / W scale can be reduced as compared with the case where the absolute value is obtained and compared.

[Brief description of drawings]

FIG. 1 is a first through third embodiment of a receiving apparatus according to the present invention.
It is a figure which shows the structure of.

FIG. 2 is a diagram showing a configuration example of a waveform shaping section.

FIG. 3 is a diagram illustrating a configuration example of a frequency characteristic updating unit.

FIG. 4 is a diagram showing a two-wave model of time response.

FIG. 5 is a diagram showing frequency characteristics of an output of a transmission path estimation unit.

FIG. 6 is a diagram showing a time axis signal output from an IFFT unit.

FIG. 7 shows a threshold value th obtained by a comparison unit and a gate circuit.
It is a figure which shows the waveform which forcedly set the time response of 1 or less to 0.

FIG. 8 is a diagram showing a frequency axis signal output from an FFT unit.

FIG. 9 is a diagram showing frequency characteristics in a frequency characteristic updating unit.

FIG. 10 is a diagram showing a time axis signal output from an IFFT unit.

FIG. 11 is a diagram showing a differential signal output from the waveform subtraction unit.

FIG. 12 is a diagram showing a waveform in which the time response of the threshold value th1 or less is forcibly set to 0 in the comparison unit and the gate circuit.

FIG. 13 is a diagram showing a waveform of an output of a waveform addition unit.

FIG. 14 is a diagram showing a pilot signal in terrestrial digital broadcasting.

FIG. 15 is a diagram showing a time axis signal output from an IFFT unit.

FIG. 16 is a diagram showing a configuration example of a waveform shaping section.

FIG. 17 is a diagram illustrating a configuration example of a frequency characteristic updating unit.

FIG. 18 is a diagram showing an output signal of a frequency characteristic difference calculation unit.

FIG. 19 is a diagram showing a time axis signal output from the IFFT unit.

FIG. 20 is a diagram showing a time-axis signal when samples below a threshold are forcibly set to 0.

FIG. 21 is a diagram showing a time response in the waveform adding unit.

FIG. 22 is a diagram showing a configuration of a fourth embodiment of a receiving apparatus according to the present invention.

FIG. 23 is a diagram showing a configuration example of a waveform shaping section.

FIG. 24 is a diagram showing frequency characteristics in a frequency characteristic updating unit.

FIG. 25 is a diagram showing a time axis signal output from an IFFT unit.

FIG. 26 is a diagram showing a time-axis signal output from a waveform subtractor.

FIG. 27 is a diagram showing a time-axis signal output from a gate circuit.

FIG. 28 is a diagram showing an output signal of the waveform addition unit.

FIG. 29 is a diagram showing a configuration of a fifth embodiment of a receiving device according to the present invention.

[Fig. 30] Fig. 30 is a diagram illustrating the configuration of a sixth embodiment of a receiving device according to the present invention.

FIG. 31 is a diagram showing an example of an averaging method by the integration processing unit.

FIG. 32 is a diagram showing an example of an averaging method by the integration processing unit.

FIG. 33 is a diagram showing a configuration example of a waveform shaping section.

FIG. 34 is a diagram showing a configuration of a repeater in terrestrial digital broadcasting.

FIG. 35 is a diagram showing a configuration of a frequency band expansion circuit in the repeater.

[Explanation of symbols]

1 FFT section, 2 pilot signal extraction section, 3 pilot signal generation section, 4 transmission path estimation section, 5 frequency characteristic memory section, 6 IFFT section, 7 waveform shaping section, 8 determination section, 9
FFT section, 10 frequency characteristic updating section, 11 interpolation processing section, 12 integration processing section, 201 waveform subtracting section, 202 absolute value calculating section, 203 maximum value detecting section, 204 threshold value defining section, 205 comparing section, 206 gate circuit , 207
Waveform addition unit, 208 memory unit, 209 initial waveform memory unit, 210 power calculation unit, 211 frequency characteristic difference calculation unit, 212 frequency characteristic difference addition unit.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Toshimasa Hamada             2-3 2-3 Marunouchi, Chiyoda-ku, Tokyo             Inside Ryo Electric Co., Ltd. F term (reference) 5K022 DD01 DD13 DD18 DD33 DD34

Claims (16)

[Claims] 1. A frequency characteristic calculating means for calculating a frequency characteristic using a known symbol included in a received signal, a frequency / time converting means for converting the frequency characteristic into a time response, and one before the time response. The time response is subtracted, waveform shaping is performed using a threshold value determined according to the maximum amplitude value of the subtraction result, and the difference between the frequency characteristic of the time response after waveform shaping and the frequency characteristic by the frequency characteristic calculation means. Is added to the frequency characteristic of the time response after the waveform shaping to update the frequency characteristic in the signal band, and the frequency / time conversion means outputs the frequency characteristic from the frequency characteristic calculation means. The receiving device is characterized by converting the frequency characteristic into a time response and then converting the updated frequency characteristic into a time response. 2. A frequency characteristic calculation means for calculating frequency characteristics using known symbols included in a received signal, a frequency / time conversion means for converting the frequency characteristics into a time response, and a maximum amplitude value of the time response. An updating means for performing waveform shaping using a threshold value determined in accordance with the difference between the frequency characteristic of the time response after waveform shaping and the frequency characteristic by the frequency characteristic calculating means, and updating the frequency characteristic in the signal band. The receiving device, wherein the frequency / time conversion unit converts the frequency characteristic output from the frequency characteristic calculation unit into a time response, and then converts the updated frequency characteristic into a time response. . 3. A frequency characteristic calculating means for calculating a frequency characteristic using a known symbol included in a received signal, a frequency / time converting means for converting the frequency characteristic into a time response, and an initial waveform (first time) from the time response. Value is subtracted, the waveform is shaped by using the threshold value determined according to the maximum amplitude value of the subtraction result, and the frequency characteristic within the signal band is updated by the frequency characteristic of the time response after the waveform shaping. Updating means for converting the frequency characteristic output from the frequency characteristic calculating means into a time response, and then converting the updated frequency characteristic into a time response. Receiving device. 4. The updating means is configured to re-update the frequency characteristic based on a time response obtained by converting the updated frequency characteristic, and the signal processing by the frequency / time converting means and the updating means is performed a predetermined number of times. The receiving device according to claim 1, wherein the receiving device is repeatedly executed. 5. The frequency characteristic calculation means calculates the frequency characteristic in units of a plurality of symbols when the known symbol is repeatedly inserted at a constant cycle on the frequency axis and the time axis. The receiving device according to any one of 1 to 4. 6. An interpolation interpolating means for interpolating the frequency characteristic of a sample (carrier) having no value, wherein the frequency / time converting means converts the frequency characteristic after the interpolating into a time response. The receiving device according to claim 5, wherein: 7. The receiving apparatus according to claim 1, further comprising: an integrating unit that averages the frequency characteristics over time. 8. The updating means determines the threshold value according to the power of the time response instead of the maximum amplitude value of the time response. The receiving device described in 1. 9. A frequency characteristic calculation step of calculating a frequency characteristic using a known symbol included in a received signal, a frequency / time conversion step of converting the frequency characteristic into a time response, and a frequency / time conversion step immediately before the time response. The time response is subtracted, waveform shaping is performed using a threshold value determined according to the maximum amplitude value of the subtraction result, and the difference between the frequency characteristic of the time response after waveform shaping and the frequency characteristic by the frequency characteristic calculation means. Is added to the frequency characteristic of the time response after the waveform shaping to update the frequency characteristic in the signal band, and the frequency / time conversion step outputs the frequency characteristic in the frequency characteristic calculation step. A transmission path estimation method comprising: converting the frequency characteristic into a time response, and then converting the updated frequency characteristic into a time response. 10. A frequency characteristic calculation step of calculating a frequency characteristic using a known symbol included in a received signal, a frequency / time conversion step of converting the frequency characteristic into a time response, and a maximum amplitude value of the time response. An updating step of performing waveform shaping using a threshold value determined in accordance with the difference between the frequency characteristic of the time response after waveform shaping and the frequency characteristic by the frequency characteristic calculating means, and updating the frequency characteristic in the signal band. In the frequency / time conversion step, the frequency characteristic output in the frequency characteristic calculation step is converted into a time response, and then the updated frequency characteristic is converted into a time response. Estimation method. 11. A frequency characteristic calculation step of calculating a frequency characteristic using a known symbol included in a received signal, a frequency / time conversion step of converting the frequency characteristic into a time response, and an initial waveform (initial waveform) from the time response. Value is subtracted, the waveform is shaped by using the threshold value determined according to the maximum amplitude value of the subtraction result, and the frequency characteristic within the signal band is updated by the frequency characteristic of the time response after the waveform shaping. In the frequency / time conversion step, the frequency characteristic output from the frequency characteristic calculation step is converted into a time response, and then the updated frequency characteristic is converted into a time response. And a transmission path estimation method. 12. In the updating step, the frequency characteristic is re-updated based on the time response obtained by converting the updated frequency characteristic, and the signal processing by the frequency / time converting step and the updating step is repeatedly executed a predetermined number of times. The transmission path estimation method according to claim 9, 10, or 11. 13. The frequency characteristic is calculated in a unit of a plurality of symbols in the frequency characteristic calculating step when the known symbol is repeatedly inserted at a constant cycle on a frequency axis and a time axis. 9-12
The transmission path estimation method described in any one of 1. 14. A sample (carrier) having no value
2. An interpolating step of interpolating the frequency characteristic of the above is included, and the frequency characteristic after the interpolating is converted into a time response in the frequency / time converting step.
3. The transmission path estimation method described in 3. 15. The transmission path estimation method according to claim 9, further comprising an integration step of averaging the frequency characteristics in time. 16. The threshold value is determined according to the power of the time response instead of the maximum amplitude value of the time response in the updating step.
5. The transmission path estimation method described in any one of 5.