JP2003060616A - Adaptive routing cancellation apparatus and method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an adaptive routing cancellation apparatus that has a simple structure and can be miniaturized easily. SOLUTION: A successive adaptive filter 53 is used to cancel routing that is generated by an SFN relay apparatus 51 in an OFDM. The successive adaptive filter 53 successively corrects a filter coefficient every time sample values are inputted, predicts routing signals, and outputs them to an adder 52. The adder 52 subtracts the prediction value from the input signal, thus cancelling the routing signal, passing signals from an antenna 1 at a master station as it is, and transmitting the signal from a transmission antenna 7.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an adaptive wraparound canceling device, and more particularly to an adaptive wraparound canceling device for canceling sneak in a terrestrial digital broadcast wave relay device based on the OFDM system.

[0002]

[Prior Art] OF adopted in domestic terrestrial digital broadcasting
DM (Orthogonal Frequency D
Ivision Multiplexing is a SFN (S) that is strong against multipath and can also be used for mobile reception.
The single frequency network (single frequency network) has various characteristics. At the time of SFN, there are a method of using the broadcast wave itself for distribution from the master station to the relay station, and a method of distributing by another line such as an optical fiber. The former is desirable in terms of cost, but performance degradation due to sneaking from the transmitting antenna of the relay station to the receiving antenna, especially the sneaking loop gain is 1
If it becomes above, the problem of signal increase and oscillation will occur.

In order to deal with this problem, various methods have been proposed in which a filter for canceling the sneak-up is installed in the relay station. One of such methods is a method of placing a filter for canceling loop-in, reproducing a pilot signal (SP) included in an OFDM signal, estimating a loop-in characteristic based on the pilot signal (SP), and constructing a filter coefficient. But this is the method to call. The other is an adaptive method in which a filter coefficient is determined so as to determine the correlation of a transmission signal and make it non-correlated. The former has been verified that relatively good performance can be obtained although OFDM demodulation is required. The latter has the merit that, in principle, there is no need for OFDM demodulation, so the structure may be simplified.

In these methods, a filter coefficient is updated in a block manner (usually for each OFDM symbol) (see, for example, Japanese Patent Laid-Open Nos. 2001-160794 and 11-355160). ing. Further, as an example in which this kind of sneak-cancellation device is applied in addition to the relay device, Japanese Patent Laid-Open Nos. 7-199971 and 10-257584 are disclosed.

[0005]

However, the adaptive sneak-canceling device used in the conventional relay device has a drawback that the device is complicated and it is difficult to miniaturize it because the filter characteristic is calculated in a block manner. It was

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an adaptive wraparound cancel device which is used in a relay device and has a simple structure and can be easily downsized.

[0007]

In order to solve the above-mentioned problems, the present invention is an adaptive sneak-cancellation device which cancels sneak by using it in a relay device of a transmission signal based on the OFDM system, which device is adapted sequentially. A filter,
An adder for adding the output signal of the successive adaptive filter and the received signal from the receiving antenna, outputting the addition result to the successive adaptive filter, and outputting the added signal as a transmission signal. According to the present invention, it is possible to provide an adaptive wraparound canceling device having the above-described configuration, which has a simple structure and can be easily downsized.

Another invention according to the present invention is an adaptive wraparound canceling method for canceling a sneak in a relay device for a transmission signal based on an OFDM system, wherein the relay device has a successive adaptive filter and an adder. The canceling method includes an adding step of obtaining a difference between a received signal from a receiving antenna and a predicted value which is an output of the successive adaptive filter by the adder, and inputs the addition result to the successive adaptive filter. And a rounding signal path characteristic estimating step of estimating the characteristics of the rounded signal path fed back from the output side to the input side of the repeater by the successive adaptive filter. According to another invention of the present invention, the same effects as those of the above-described present invention can be obtained by the above-mentioned configuration.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention is a wrap-around canceller for a terrestrial digital broadcast repeater using a successive adaptive filter, which does not require OFDM demodulation and has QAM (Qua
draw Amplitude Modulat
ion / DQPSK (Differential Qu)
Adorature Phase Shift Keyi
ng) It operates regardless of the modulation method, operates even if there is a large sneak gain, and is capable of significantly simplifying and downsizing the device. The present invention is characterized in that the update of the filter coefficient is not processed blockwise, but is performed for each sample value.

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is an overall configuration diagram of a relay system using an adaptive wraparound cancel device according to the present invention. Referring to the figure, the relay system includes a master station antenna 1, a multipath transmission line 2, and a reception antenna 3.
The relay device 5 and the transmitting antenna 7 are included. The relay device 5 includes an adaptive wraparound cancel device including an adder 52 and a successive adaptive filter 53.

Next, the operation of the relay system will be briefly described. The broadcast wave transmitted from the antenna 1 of the master station is received by the reception antenna 3 via the multipath transmission path 2, and is further input to the relay device 5 via the input terminal 51. Further, it is shown that the broadcast wave is transmitted from the transmitting antenna 7 to the outside via the output terminal 56 of the relay device 5, and a part of the broadcast wave wraps around to the receiving antenna 3 of the relay device 5 via the sneak path 8. ing.

That is, the multipath reflection, distortion, and noise are added to the received signal from the antenna 1 of the master station, and the sneak signal from the output of the repeater 5 itself is similarly subjected to multipath reflection, distortion, and noise. There is noise. In either case, the transmission path is FIR (Finite Impulse).
It would be reasonable to assume that it could be approximated with a Response filter.

Here, (1 + M) shows the multipath transmission line 2 from the master station, and (F) shows the loop-in transmission line 8. Now, let us say that the master station signal is X, the output signal of the relay device 5 is S, and the noise included in the received signal (sum of noise generated at various points) is represented by N. A correction filter for coping with the wraparound is represented by T. Then, in the relay device 5, the respective signals are related by (X (1 + M) + SF + N) T = S ... (1).

Here, assuming that the output S of the relay device is equal to the master station signal X and the noise N is ignored, the correction filter T
Becomes T = 1 / (1-MF) (2), and ideally, if FIR filter 53 having the characteristic of H = M + F is arranged as shown in FIG. The characteristics can be canceled.

That is, assuming that the master station signal is X, the sneak signal from the transmitting antenna 7 is Y, and the output signal (predicted signal) from the FIR filter 53 is Z, the output of the adder 52 = X + Y−Z (3) The FIR is calculated so that (Y−Z) is minimized.
It suffices to set the coefficient of the filter 53.

Next, the structure and operation of the adaptive wraparound cancel device according to the present invention will be described in detail. FIG. 2 is a detailed configuration diagram of the adaptive wraparound cancel device according to the present invention. The same components as those in FIG. 1 are designated by the same reference numerals and the description thereof will be omitted. Referring to the figure, the adaptive wraparound cancel device is provided in the relay device 5, and has an input terminal 51, an adder 52, a successive adaptive filter 53, a stabilizing filter 54, an amplifier 55, and an output. And a terminal 56. Then, the input terminal 51 and the first input side of the adder 52 are connected, the output side of the adder 52 is connected to the input side of the stabilizing filter 54 and the first input side of the successive adaptive filter 53, and the stabilizing filter The output side of 54 is connected to the input side of the amplifier 55 and the second input side of the successive adaptive filter 53, the output side of the amplifier 55 is connected to the output terminal 56, and the output side of the successive adaptive filter 53 is the second input of the adder 52. Connected to the side.

The broadcast wave sent from the antenna 1 of the master station passes through the transmission path 2 and is received by the receiving antenna 3 and input to the relay device 5 via the input terminal 51. In the relay device 5, the difference between the signal from the input terminal 51 and the prediction signal output from the successive adaptive filter 53 is created by the adder 52, passes through the stabilizing filter 54, is power-amplified by the amplifier 55, and then is transmitted from the transmitting antenna 7. Sent. The signal from the transmitting antenna 7 is further transmitted to the antenna 9 for relay or home reception.
Sent to. In addition, a part of the signal from the transmitting antenna 7 enters the receiving antenna 3 of the relay device through the sneak path 8 to form an unnecessary feedback loop.

Next, the operation of the adaptive wraparound cancel device will be described with reference to FIGS. Figure 3
3 is a flowchart showing the operation of the adaptive wraparound cancel device. In FIG. 2, the signal transmitted from the antenna 1 of the master station reaches the reception antenna 3 of the relay device through the transmission path 2. At that time, if there is reflection, it becomes a multi-pass (ghost). Further, the signal output from the relay device 5 is also input to the receiving antenna 3 through the transmission line 8. Relay device 5
Is the difference between the signal from the input terminal 51 and the predicted value which is the output of the successive adaptive filter 53, calculated by the adder 52 (S in FIG. 3).
1) The error is input to the stabilizing filter 54 and the successive adaptive filter 53 (S2 in FIG. 3). Stabilizing filter 54
Output is power-amplified by the amplifier 55 and output terminal 5
It is sent to the transmitting antenna 7 via 6 and is also input to the successive adaptive filter 53 (S3 in FIG. 3). The successive adaptive filter 53 estimates the overall characteristics of the amplifier 55, the transmitting antenna 7, the sneak path 8, the receiving antenna 3, and the input terminal 51. Based on the output signal of the stabilizing filter 54, the successive adaptive filter 53 outputs from the receiving terminal 51. The input signal is predicted (S4 in FIG. 3). Then, the adder 52 subtracts the predicted value from the input signal (S1 in FIG. 3) to cancel the sneak-through signal, pass the signal from the master station as it is, and transmit the signal from the transmission antenna 7. Function as.
As described above, every time the signal transmitted from the antenna 1 of the master station reaches the receiving antenna 3 of the relay device, S1 to S4 in FIG.
The process up to is repeated.

Next, the structure and operation of the successive adaptive filter 53 will be described. FIG. 4 is a block diagram of an example of the successive adaptive filter 53. Referring to the figure, the successive adaptive filter 53 includes a first input terminal 531 to which the output signal of the adder 52 is input, a second input terminal 532 to which the output signal of the stabilizing filter 54 is input, and a transmission signal storage unit. 533, a control unit 534, a coefficient storage unit 535, and an output terminal 536.

Then, the first input terminal 531 and the controller 53
4 is connected to the second input side, the second input terminal 532 is connected to the input side of the transmission signal storage section 533, and the output side of the transmission signal storage section 533 is connected to the first input side of the control section 534. The output side of the coefficient storage section 535 is connected to the third input side of the control section 534, the first output side of the control section 534 is connected to the output terminal 536, and the second output side of the control section 534 and the coefficient are connected. The input side of the storage unit 535 is connected. or,
The output terminal 536 is connected to the second input side of the adder 52.

Next, the operation of the successive adaptive filter 53 will be described with reference to FIGS. 4, 5 and 6. 5 and 6 are flowcharts showing the operation of the successive adaptive filter 53. Now, let S be the transmission signal (output signal from the stabilizing filter 54) at a certain sampling time j.
(J). The transmission output signal S (j) is stored in the transmission signal storage unit 533. Filter coefficients H (1), H (2), ..., H (n), ..., H (N) of the successive adaptive filter 53.
(However, 1 ≦ n ≦ N) is stored in the coefficient storage unit 535.

The control unit 534 inputs the transmission signal S (j-n) input in the past from the transmission signal storage unit 533 and the filter coefficient H (n) from the coefficient storage unit 535, and calculates the product of both (FIG. 5 S11). The control unit 534 obtains the sum of all products obtained for all n and outputs the sum as an estimated value from the output terminal 536 (S12 in FIG. 5).

At the same time, the control unit 534 corrects the filter coefficient. The correction method is as follows. For the nth coefficient H (n), the output e (j) of the adder 52 is added to the transmission signal storage unit 53.
3 is multiplied by the signal S (j-n) stored in S3 (S2 in FIG. 6).
1), which is further multiplied by a small positive value correction coefficient α (S22 in FIG. 6) and added to the filter coefficient H (n) to update the filter coefficient H (n) (S2 in FIG. 6).
3), the updated filter coefficient H (n) is stored in the coefficient storage unit 53.
Enter in 5 again. By repeating this operation, the filter coefficient of the coefficient storage unit 535 gradually approaches the correct value.

Further, by multiplying the filter coefficient by a value slightly smaller than 1, the past influence can be gradually eliminated. That is, even if the filter is disturbed, the influence can be eliminated.

The present invention will be described in more detail below. The adaptive wraparound canceller apparently has IIR (Info
Although it constitutes a unit impulse response) filter, from the viewpoint of adaptive processing, it results in the problem of optimizing the coefficient of the FIR filter (approaching the characteristic F of the wraparound propagation path) with the transmission signal S of the relay device as an input. Therefore, theoretically, it is possible to cope with a large wraparound amount. However, since it is an adaptive IIR filter in which a loop is formed by the filter H and the adder, the factor that becomes unstable when the amount of wraparound is large is extremely large. or,
Since the master signal X acts as noise, the convergence is extremely limited in terms of system identification.

Now, the sneak signal at the input point of the relay device 5 at a certain sampling time j is y _j , its estimated value is z _j ,
It is assumed that the signal from the master station is x _j , the transmission signal is s _j , the tap K of the filter H and the k-th coefficient at time j are h _kj . And
The total sum of noises in the multipath, sneak path, receiving amplifier, etc. is represented by n _j . Also, in order to simplify the explanation, it is assumed that there is no multipath. Note that the compensation of the multipath characteristic becomes a coefficient estimation problem of a normal IIR adaptive filter.

[0027] Then, the estimated value z _j of the echo signal y _j is,

[0028]

[Equation 1] And the output signal of the relay apparatus, the _{s j = y j -z j +} x j + n j = u j + x j + n j ... (5). Note that u _j is the cancellation residual difference, that is, u _j = y _j −z _j (6).

In the relay device 5, the signal x _j from the master station
Since the wraparound signal y _j cannot be separated from the wraparound signal y _j , the wraparound canceller that does not use decision feedback cannot estimate the signal x _j from the master station. That is, the standard that the transmission signal s _j approaches the signal x _j from the master station cannot be used. So instead of that,
The least squares method is used to modify the filter coefficients so as to minimize the power of the transmission signal s _j (and consequently the cancellation residual difference u _j ). Therefore, the signal x _j from the master station and the total sum n _{j of} noise act as a disturbance.

To obtain the correction amount of the coefficient, the transmission signal s _j
Partial differentiation of the square of by the coefficient h _{k, j} gives

[0031]

[Equation 2] Is obtained. This equation, h _k, the direction to be corrected and _j (negative) is the polarity of the s _j × s _jk, it represents that, therefore, h _k, modified expression of _j is, h _{k, j + It can be written that 1} = h _{k, j} + μs _j s _jk (8). Here, μ is a positive and relatively small value. Various methods have been proposed as an actual correction formula. Here, an adaptive algorithm called a learning identification method that is easy to analyze is adopted. In the learning identification method,

[0032]

[Equation 3] Is. Here, Sj = (s _j-1 , s _j-2 , ..., s _jK ) ^T
… (10),

[0033]

[Equation 4] Is. Here, T represents transposition. Hereinafter, the sampled value of each signal is expressed in lower case and the matrix is expressed in upper case. The convergence condition in general FIR system identification is 0 <α <2 (12). In this case, when the correction coefficient α is 1, the convergence speed is the fastest. However, the smaller the correction coefficient α, the better the final cancellation amount. In the relay device 5, the signal x from the master station
_{Since j} acts as noise, an extremely small value is used as the correction coefficient α.

Further, the OFDM signal is not white noise but a signal with a limited band (although it has a substantially flat frequency characteristic in that band), and the characteristic can be obtained in a band where no signal exists. Can not. Although it does not often become a problem in a normal system identification problem, if a loop is drawn like a sneak-cancellation device and the loop gain is 1 or more, once noise is generated outside the band, it grows. It may start and become unstable.

In the present invention, a slight loss (several dB is sufficient) is given to the out-of-band component in the loop so that the out-of-band component is converged stably. The "providing a slight loss" is the stabilizing filter 54 shown in FIG.
A band pass filter (band pass filter: BPF) is an example of the stabilizing filter 54.

In the iterative adaptive filter 53, there is a probability that the filter coefficient is corrected in the wrong direction, and as described above, the input signal may not be white noise. It is not always moving in the right direction. Therefore, it is actually effective to add a leak effect to the filter coefficient and gradually eliminate the past effect. As a method, each time the filter coefficient is modified, the value of the filter coefficient is slightly smaller than 1 (0.999999).
Etc.) or a certain small positive constant is determined. If the filter coefficient is positive, this constant is subtracted, and if it is negative, this constant is added. In any case, the stability of the adaptive filter can be further increased by gradually eliminating the past influence.

In the wraparound canceller, O
There is a choice as to whether to handle the FDM signal at the intermediate frequency (IF) or at the complex baseband. Basically, both are the same, but the complex base band requires a complex operation, but the sampling frequency decreases, the bandwidth in which carriers exist is relatively wide, and the autocorrelation is small. Therefore, it is easy to handle.

When the IF is used, the correction formula of the filter coefficient can be used as it is in the above formula, but when the complex signal is used, the above-mentioned correction formula needs to be slightly changed. In that case, in addition to replacing the operation with a complex operation, S (j−n) of e (j) × S (j−n) in the description of the method of modifying the filter coefficient described above is converted into a conjugate complex number S (j). -N) * need only be replaced.

More specifically, when h _{k, j} is recalculated as a complex number,

[0040]

[Equation 5] Becomes Here, * represents conjugation. or,

[0041]

[Equation 6] Is.

FIG. 7 is a block diagram of another example of the adaptive wraparound cancel device. In the figure, the same components as those in FIG. 2 are designated by the same reference numerals, and the description thereof will be omitted. Although the stabilizing filter 54 is inserted between the adder 52 and the amplifier 55 in FIG. 2, the present invention is not limited to this, and as shown in FIG. It is also possible to insert it between the adaptive filter 53. However, in this case, the stabilizing filter 54
-2 must be included in the amplifier 55. In FIG. 7, the amplifier is divided into 55-1 and 55-2 and the stabilizing filter 54-2 is inserted between them, but the amplifier 55-1 may be omitted.

The point of the present invention is that it is the first example in which the successive successive adaptive filter is used to cancel the wraparound. When the wraparound loop gain is 0 dB or more (when the loop gain is 1 or more), the successive filter is used. May be unstable, and a stabilization filter is inserted to solve the problem, and stabilization is achieved by leaking the coefficient little by little.

[0044]

According to the present invention, there is provided an adaptive wraparound canceling device for canceling wraparound by using it in a relay device of a transmission signal based on the OFDM system, which device is a successive adaptive filter and an output of the successive adaptive filter. An adaptive wraparound cancel device that has a simple structure and is easy to downsize because it includes an adder that adds a signal and a received signal from a receiving antenna and outputs the addition result to the successive adaptive filter and a transmission signal. It becomes possible to provide a device. Other inventions according to the present invention also have the same effects as the present invention.

More specifically, according to the present invention, a wrap-around canceller that sequentially corrects the filter coefficient each time a sample value is input makes the structure of the device extremely simple, and is applicable regardless of the modulation method. What is possible is that, even if the loop gain is large, the operation is possible.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of a relay system using an adaptive wraparound cancel device according to the present invention.

FIG. 2 is a detailed configuration diagram of an adaptive wraparound cancel device according to the present invention.

FIG. 3 is a flowchart showing the operation of the adaptive wraparound canceller.

FIG. 4 is a configuration diagram of an example of a successive adaptive filter 53.

FIG. 5 is a flowchart showing the operation of a sequential adaptive filter 53.

FIG. 6 is a flowchart showing the operation of the successive adaptive filter 53.

FIG. 7 is a configuration diagram of another example of the adaptive wraparound cancel device.

[Explanation of symbols]

Antenna of 1 master station 2 transmission lines 3 Receiving device's receiving antenna 5 Repeater 7 Transmitter antenna of relay device 8 Roundabout transmission line 9 Next-stage relay station or receiver 51 Input terminal of repeater 52 adder 53 Adaptive filter 54 Stabilizing filter 55 Amplifier 56 Output terminal of repeater 531 1st input terminal 532 2nd input terminal 533 Transmission signal storage unit 534 control unit 535 Coefficient storage unit 536 output terminal

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 5K022 DD01 DD34                 5K046 AA05 BA01 BB00 EE06 EE57                       EF11                 5K072 AA22 BB14 DD11 DD16 GG09                       GG14

Claims (19)

[Claims] 1. An adaptive sneak-cancellation device for canceling sneak in a relay device for a transmission signal based on the OFDM system, comprising: a successive adaptive filter; an output signal of the successive adaptive filter; and a received signal from a receiving antenna. And an adder that outputs the addition result to the successive adaptive filter and outputs it as a transmission signal. 2. The successive adaptive filter is configured so that a difference between a loop-in signal fed back from an output side to an input side of the relay device and a prediction signal output from the successive adaptive filter is minimized. The adaptive wraparound canceller according to claim 1, further comprising filter coefficient modifying means for modifying a coefficient. 3. The filter coefficient correction means includes a transmission signal storage section for storing the transmission signal, a coefficient storage section for storing filter coefficients of the successive adaptive filter, the transmission signal storage section and the coefficient storage section. 3. The adaptive wraparound canceling device according to claim 2, further comprising: a control unit that controls the control signal to calculate a predetermined estimated value from the transmission signal and corrects the filter coefficient. 4. The control unit calculates a product of a transmission signal accumulated in the signal accumulating unit and a filter coefficient accumulated in the coefficient accumulating unit, and outputs a sum of the products as the estimated value. The adaptive wraparound canceling device according to claim 3, which is characterized in that. 5. The control unit multiplies the output of the adder by the transmission signal accumulated in the signal accumulating unit, further multiplies the multiplication result by a correction coefficient, and adds the result to the filter coefficient to add the filter coefficient. The adaptive wraparound cancel device according to claim 3 or 4, characterized in that 6. The adaptive wraparound cancel device according to claim 2, wherein the correction of the filter coefficient is performed by using a least square method. 7. The filter coefficient modifying means, when modifying the filter coefficient, either multiplies the coefficient by a value slightly smaller than 1 (such as 0.999999), or determines a small positive constant, 7. The adaptive wraparound canceling device according to claim 2, wherein if the value is positive, the constant is reduced, and if the value is negative, the constant is added. 8. The adaptive filter according to claim 1, further comprising a stabilizing filter that attenuates an out-of-band component in the output signal of the adder by a certain amount and outputs the attenuated component to the successive adaptive filter. A wraparound cancel device. 9. The adaptive wraparound cancel device according to claim 8, further comprising an amplifier that amplifies the output of the stabilizing filter, and the amplified signal becomes a transmission signal. 10. The adaptive wrap-around cancel device according to claim 1, wherein the relay device is an SFN relay device. 11. An adaptive wraparound canceling method for canceling a sneak in a relay device for a transmission signal based on the OFDM system, wherein the relay device includes a successive adaptive filter and an adder. Is an addition step of obtaining a difference between a reception signal from a reception antenna and a prediction value which is an output of the successive adaptive filter by the adder, an addition result input step of inputting the addition result to the successive adaptive filter, A wraparound signal path characteristic estimating step of estimating a path characteristic of a wraparound signal that is fed back from the output side to the input side of the repeater by a successive adaptive filter. 12. The successive adaptive filter includes a filter coefficient modifying step of modifying a coefficient of the successive adaptive filter so that a difference between the wraparound signal and a prediction signal output from the successive adaptive filter is minimized. The adaptive wraparound canceling method according to claim 11, which is characterized in that. 13. The filter coefficient correction step inputs a transmission signal input in the past and a filter coefficient of the successive adaptive filter, calculates a product of both, and calculates a sum of all products. 13. The adaptive wraparound canceling method according to claim 12, further comprising an estimated value output step of outputting the estimated value. 14. The filter coefficient correction step comprises a signal multiplication step of multiplying an output of the adder by a transmission signal input in the past, a correction coefficient multiplication step of further multiplying the multiplication result by a correction coefficient, and the multiplication thereof. 14. The adaptive wraparound canceling method according to claim 12, further comprising a filter coefficient correction step of adding a result to the filter coefficient and correcting the filter coefficient. 15. A stabilizing filter that attenuates an out-of-band component in an output signal of the adder by a certain amount and outputs the attenuated component to the successive adaptive filter is further included, and the canceling method includes a reception signal from the reception antenna. An addition step of obtaining a difference from a predicted value which is an output of the successive adaptive filter by the adder, an addition result input step of inputting the addition result to the stabilizing filter and the successive adaptive filter, and A stabilizing filter output transmission / input step of transmitting an output and inputting it to the successive adaptive filter, and a sneak path characteristic estimating step of estimating a characteristic of the sneak path by the successive adaptive filter. Item 1
The adaptive wraparound canceling method according to any one of 1 to 14. 16. The method according to claim 12, wherein the modification of the filter coefficient is performed by using a least square method.
Either of the adaptive wraparound canceling methods. 17. The filter coefficient modifying step comprises multiplying the filter coefficient by a value slightly smaller than 1 (such as 0.999999) when modifying the filter coefficient, or by defining a small positive constant. 17. The adaptive wraparound canceling method according to claim 12, wherein if is positive, this constant is subtracted, and if negative, this constant is added. 18. The adaptive wraparound canceling method according to claim 15, further comprising an amplifier that amplifies the output of the stabilizing filter, and the amplified signal becomes a transmission signal. 19. The adaptive wraparound canceling method according to claim 11, wherein the relay device is an SFN relay device.