JPH0856186A - Transmitter and transmission method - Google Patents

Abstract

PURPOSE:To maintain high demodulation efficiency independently of a fading state by generating the same pattern data as known pattern data in transmission data at a receiver side and using an adaptive filter whose tap number is made variable so as to demodulate reception data. CONSTITUTION:A high frequency RF signal processing circuit 3 uses an RF/IF/ detection circuit 3A to apply high frequency amplification to a reception signal and applies orthogonal detection to the signal. Then the signal is converted into a digital signal by an A/D converter circuit 3B and the signal is given to a MODEM/data processing circuit 4. The circuit 4 demodulates received data by signal processing in a demodulation digital signal processor DSP 4A and executes periodic processing such as fetch processing, equalization processing, demodulation processing and error correction processing. The number of taps of the adaptive filter used for equalization processing is made variable and the adaptive filter after becoming variable is used to demodulate reception data. Thus, even when the reception data are received under an environment having lots of distortion due to fading, the data are demodulated at a low bit error rate.

Description

Detailed Description of the Invention

[0001]

[Table of Contents] The present invention will be described in the following order. Field of Industrial Application Conventional Technology Problems to be Solved by the Invention Means for Solving the Problems (FIGS. 1 and 2) Action Example (FIGS. 1 to 7) (1) Overall Configuration (FIGS. 1 and 2) (2) Demodulation process (FIGS. 3 and 4) (3) Transmission path characteristic estimation method (3-1) Overview (3-2) Tap number determination process 1 (FIG. 5) (3-3) Tap number determination Decision processing 2 (FIGS. 6 and 7) (4) Other embodiments Effect of the invention

[0002]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a transmission device and a transmission method, and is preferably applied to, for example, a digital cordless telephone and a reception method adopted in the same.

[0003]

2. Description of the Related Art Today, various systems (digital cordless telephone system and digital car telephone system) are being studied in various countries in order to realize a digital mobile communication system. For example, DECT (Digital European Cordl
ess Telecommunications) and GSM (Groupe Special d
e Mobile) is that.

[0004]

By the way, in a system in which a terminal moves at a high speed such as a digital car telephone, the received signal is easily affected by fading depending on the condition of the transmission line, and the receiving characteristic is easily deteriorated. Moreover, since the fluctuations in the reception characteristics due to fading are not uniform, if the processing method at the receiving terminal is fixed, complicated processing is required to obtain the optimum transmission path characteristics (channel response) and equalize the waveform of the received signal. In some cases, it was necessary and sufficient characteristics could not be obtained.

Under such a situation, the synchronization timing is deviated and the demodulation efficiency is likely to be deteriorated, and the bit error rate may be deteriorated. In addition, if the processing is too complicated, the circuit scale may become large and the power consumption may increase accordingly.

The present invention has been made in consideration of the above points, and an object thereof is to propose a transmission device and a transmission method that are much more resistant to distortion of the transmission line than the conventional ones.

[0007]

In order to solve such a problem, according to the present invention, pattern generating means for generating the same pattern data as the known pattern data (t _S ) included in the transmission data transmitted on the transmission path. and (4A), the correlation value a correlation between the received data (r _k) and the pattern data (t _S) is received via a transmission channel correlation detector means for sequentially detected as (CorrRS _k) (4A), the correlation value variable to the tap number of the adaptive filter used for equalization of the received data (r _k) on the basis of a signal processing means for demodulating the received data (r _k) using the adaptive filter after the variable a (4A) It should be installed in the transmission device.

Further, in the present invention, the data is transmitted to the transmission line.
Known pattern data (t_S)
Generate the same pattern data as
(T _S) And the received data received via the transmission line
(R_k) And the correlation value with
Correlation value (CorrRS_k) Received data based on
(R_k) Number of taps of the adaptive filter used for equalization processing
The received data using the adaptive filter after changing
(R_k) To the transmission device.
It

[0009]

On the receiving side, the same pattern data as the known pattern data (t _S ) included in the transmission data transmitted to the transmission line is generated, and the pattern data (t _S ) is first received through the transmission line. The correlation with the received data (r _k ) is _obtained . Then by varying the tap number of the adaptive filter used for equalization of the received data based on the correlation value _{(CorrRS k) (r k)} , demodulates the received data (r _k) using the adaptive filter after the variable . As a result, data can be demodulated at a low bit error rate even when the received data (r _k ) is received in a situation where there are many distortions due to fading or the like. Further, since the number of taps can be changed as appropriate, the amount of calculation required for signal processing can be reduced without lowering the bit error rate.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings.

(1) Overall Configuration First, a schematic configuration of a digital mobile phone will be described. FIG.
As shown in FIG. 1, the mobile phone 1 is configured by various circuits such as a radio frequency (RF) signal processing circuit 3, a modulation / demodulation / data processing circuit 4, and a voice processing circuit 5 in addition to the transmitting / receiving antenna 2. Then, while the transmitted voice signal is reproduced through the speaker 6, the voice of the caller is captured through the microphone 7 and transmitted from the transmitting / receiving antenna 2.

Here, the processing circuits 3 to 5 are respectively connected via a bus 8 to a central processing unit (CPU: Central Proc).
essing unit 9). The CPU 9 operates based on a program stored in a ROM (Read Only Memory) 10A, and a RAM (Ra
ndom Access Memory) 10B is used as a calculation table. Further, the CPU 9 sends and receives data to and from the display operation unit 11,
It is designed to interface with the caller.

FIG. 2 shows the mobile phone 1 in more detail with a focus on the receiving system. The high frequency (RF) signal processing circuit 3 carries out high frequency amplification of the received signal by the RF / IF / detection circuit 3A and then performs quadrature detection, and converts this into a digital signal by the A / D conversion circuit 3B. Then, the received data is applied to the modulation / demodulation / data processing circuit 4.

The modulation / demodulation / data processing circuit 4 is a digital signal processor (DSP) for demodulation.
In addition to demodulating the received data by the signal processing in 4A, the synchronization processing such as the FETCH (FCH) processing, the equalization processing, the demodulation processing and the error correction processing are executed and the demodulated data is given to the voice DSP 5A. Has been done. Incidentally, the voice DSP 5A is used for demodulating the compressed voice data. Next, the principle of the waveform equalization processing by the demodulation DSP 4A and the demodulation processing executed at this time will be described.

(2) Demodulation process Fading occurs when a plurality of reflected waves with different delay times reflected at various places are received as a composite wave. This state can be modeled by a transversal type filter as shown in FIG. This diagram is expressed by considering up to a delay time of 5 bits, and h _k represents a tap coefficient that determines the magnitude of the signal corresponding to each delay time.

By the way, how much time delay can be actually dealt with by this model is 5 times of one bit.
Up to twice the time. For example, the data bit rate is 270
If it is [kbps (bit per second)], the time per bit is about 3.7 [μs], which is five times that of 18.5 [μs].
s] can be dealt with. Therefore, if there is a reflected wave with a time delay exceeding this, this model cannot handle it.

As can be seen from FIG. 3A, the fading is represented as the sum of the values _obtained by multiplying the data for each bit interval by the tap coefficient h _k . That is, the fading can be said to be the result of the convolution operation of each delayed bit and the tap coefficient h _k . Therefore, the total bit delay amount (day lace plate) (here, 5 bits)
Can be specified, the characteristics of the transmission line can be determined by obtaining the tap coefficient h _k .

In order to remove the influence of this fading on the side of the cellular phone 1, the received signal must be equalized in waveform to return it to a signal without fading. In principle, if the value of the tap coefficient h _k representing the transmission line characteristic is accurately known, a filter having an inverse characteristic with respect to the transmission line characteristic can be constructed, so that the influence of fading can be eliminated. Therefore, how to actually estimate the characteristics of the transmission path accurately is an important technique for improving the reception characteristics.

However, in the case of mobile communication, the characteristics of the transmission line fluctuate considerably with time, and it is generally not easy to accurately estimate this. But GSM
In the car telephone system represented by the system, since the time division multiple access method is adopted as the access method, the unit transmission time (that is, 1 burst) assigned to the transmission data is as short as about 580 [μs], and within each burst The effect of fading can be regarded as almost constant.

Therefore, pattern data (training sequence) is inserted in the center of each burst during transmission,
The tap coefficient h _k for each burst will be estimated by determining the correlation with this training data upon reception. This is based on the fact that the following equations are established between the correlation value and the tap coefficient.

Here, the autocorrelation value AutoCorr _k for the bit patterns that make up the training sequence is

[Equation 1] Select one that satisfies the orthogonal relationship as shown in. Incidentally, ts represents a training sequence.

The received signal sequence r _k is then the transmitted signal sequence s
_{Using k} and tap coefficient h _k ,

[Equation 2] It is represented by. When the correlation value CorrRS _k between the training sequence ts and the received signal r _k is calculated based on these two formulas, the following formula is obtained.

(Equation 3) Becomes

If the bit pattern of the received signal r _{k and} the bit pattern of the training sequence ts are the same, the equation (3) is given by the following equation due to the orthogonality of the training sequence ts.

[Equation 4] Becomes

That is, the correlation value CorrRS _k between the received signal r _k and the training sequence ts becomes the tap coefficient h _k . The transmission line characteristics (channel response) can be estimated by using this relationship. However, in an actual transmission line, the response characteristic may spread to more than 5 bits, so if the bit delay amount is fixed at 5 bits, an error may occur in the estimation of the transmission line characteristic. In a situation in which a particularly large signal delay occurs, such a situation is likely to occur when the surroundings are surrounded by mountains.

In such a case, the transmission path estimation with 5 taps is not sufficient, and the number of taps of 6 bits or more is required. Therefore, Fig. 4 shows the actual measurement results of this situation using fading in the hilly area. Where the broken line (t
ap5) is a 5-tap fading model (Fig. 3 (A)).
Is the reception characteristic when the is adopted, and the solid line (tap6) is 6
It is a reception characteristic when a tap fading model (FIG. 3B) is adopted. As shown in the figure, it was found that the characteristics of the 6-tap model were considerably improved.

However, in most cases, the fading model simulation with a relatively short delay time shows no characteristic change or only a slight characteristic deterioration. Therefore, as far as this result is seen, it is expected that the tap number of 6 improves the characteristics as a whole.
However, an increase in the number of taps directly leads to an increase in the amount of waveform equalization processing, so it cannot be easily increased. For example, in the case of an equalizer using Viterbi decoding, the number of states is 16 when the number of taps is 5, whereas the number of states is 32 when the number of taps is 6, and the processing amount is doubled if simply considered. Will end up.

This increase in the amount of processing may require a processor having a high processing speed in some cases, and shortens the sleep time in the case of shifting to the sleep mode after the processing of the processor is completed, resulting in an increase in current consumption. There was a connecting problem. Furthermore, if the number of taps is increased when the delay time due to fading is short, etc., it may lead to deterioration of the characteristics in some cases, so it is desirable to minimize the number of taps.

(3) Transmission Channel Characteristic Estimation Method (3-1) Outline In this section, the tap number for setting the fading model is determined based on the value of the tap coefficient estimated at the time of channel estimation, and the determined tap number is determined. A method of improving the signal demodulation function of the receiver within a minimum processing time by equalizing the waveform based on the fading model of (3) will be shown.

As described in the previous section, it is possible to estimate the transmission path by calculating the correlation value between the bit sequence of the received signal and the bit sequence of the training sequence. The tap coefficient of is not obtained. Actually, since it is necessary to obtain the reception timing by this processing, the correlation value calculation processing is performed in a wider range to obtain a larger number of correlation values than the required number of taps, and then the required number of taps is calculated from these. Choose.

At this time, the tap number is selected based on the value obtained by weighting and averaging the correlation values. At this time, however, what kind of value it takes becomes a problem. In this case as well, the optimum weighting value differs depending on the fading situation, and therefore it is necessary to select the optimum value according to the situation at that time. For example, the time delay due to multipath does not change so much in a short time unless the receiver moves at a high speed, so the multipath status can be determined to some extent based on the status of the burst signal received up to the previous time.
Therefore, a good reception result can be obtained by applying a weight considered to be optimum from the determination results up to the previous time (past) at the time of selection.

The tap coefficient h _k obtained by the above method is obtained only once per burst regardless of whether the number of taps is 5 or 6, and the ratio thereof is not so large in view of the processing amount of the entire equalization processing. Not big. However, the waveform equalization work such as the Viterbi algorithm processing using this coefficient is performed for each bit, and as described above, the total calculation amount greatly differs when the tap number is 5 or 6. Therefore, in the present embodiment, the number of taps is set to 6 only when necessary, and the number of taps is normally set to 5 to reduce the total amount of processing and realize a process in which the characteristics are not deteriorated so much.

Incidentally, the difference between the processing algorithms of the transmission path estimation processing with the number of taps 5 and the transmission path estimation processing with the number of taps 6 is only the difference in the number of taps finally selected to be 5 or 6, and there is no difference in the algorithm itself. There is no difference, and the throughput is slightly larger when the number of taps is 6, but there is no big difference. In this way, focusing only on the transmission path estimation processing, even if the number of taps is calculated as 6, the processing amount does not increase significantly.
Find the tap coefficient for each. Next, a concrete processing procedure is shown in FIG.
Will be explained.

(3-2) Tap Number Determining Process 1 First, when the tap number selecting process is started from step SP1, the correlation value between the received signal r _k and the training sequence ts is calculated in the following step SP2, and these are calculated. Six tap coefficients are calculated based on the weighted average of Here, the six tap coefficients are respectively h _-2 , h _-1 , h ₀ , h ₁ , and
Let h ₂ and h ₃ .

When the calculation process of the six tap coefficients is completed in this way, the process proceeds to step SP3, and the tap coefficients h _-2 and h ₃ located at both ends of the tap coefficient are compared and the value (energy) is small. Pay attention to the one. This is because if this value is small, the effect on the waveform is considered to be small, and even if it is ignored and the number of taps is selected, the effect on decoding efficiency can be reduced. Incidentally, the description of the energy of the tap coefficient is because the tap coefficient may take a complex value. In that case, the absolute value is used for the judgment.

If it is determined in step SP3 that the tap coefficient h _{-2 having} the smaller delay time is smaller than the tap coefficient h _{3 having} the larger delay time, the tap coefficient h _-2 is next supplied in step SP4. Is a predetermined threshold T
Move to a process smaller or larger than h.
This is because if the value (energy) of the tap coefficient is sufficiently small, it is regarded as 0 and the number of taps is reduced to 5.

That is, the value of the tap coefficient h _-2 is the threshold value T
If h or more, the process goes to step SP5 to set the tap number to 6. Then, the six tap coefficients h _{-2 to} h ₃ including the tap coefficient coefficient h _-2 are set as values to be used in the subsequent processing, and the series of processing is ended in the subsequent step SP7.

On the other hand, if the tap coefficient h _-2 is smaller than the threshold value Th in step SP4, step SP8.
Move to and set the tap number to 5. The following step SP9 5 single value h _-1 ~ excluding tap coefficients h _-2 in
A set of h _{3 is set} to the tap coefficient, and a series of processing is ended.

On the other hand, if it is determined in step SP3 that the tap coefficient h ₃ corresponding to the signal having a long delay time is smaller, the process proceeds to step SP10, but the subsequent processing is similar to that of step SP4 to step SP9. Processing. The reason is as follows. That is, if it is determined in step SP10 that the value of the tap coefficient h ₃ is greater than or equal to the threshold value Th, the number of taps is set to 6 in step SP11, and then six tap coefficients h including this tap coefficient h ₃ are set. _{-2 to} h ₃ are set as the values used in the subsequent processing, while the value of the tap coefficient h ₃ is the threshold value Th.
After excisional the tap number 5 in step SP13 when it is determined that a smaller, and ends the series of processes by excisional five values h _-2 to h _2, except the tap coefficient h ₃ to the tap coefficient It is done like this.

Therefore, for example, if the tap coefficient h _-2 is 20 out of the six tap coefficients h _-2 to h ₃ obtained in step SP2, and the tap coefficient h ₃ is 1000. , The tap coefficient h ₋₂ is smaller than the other tap coefficients h ₃ , the process is executed via step SP3 through step S3.
Move to P4, and at this step SP4, this value (= 2
0) is determined to be a value that is small enough to be ignored. When it is determined that this value (= 20) is sufficiently small, the tap number is set to 5.

By the way, there is a problem that the influence on the decoding process can be reduced even if the value is neglected, but this can be set based on the result of simulation or the like. However, even if set in this way, the optimum value changes depending on the amplitude of the received signal, etc., so in actual processing, the threshold Th itself is normalized by the energy of the received signal, or the energy of the selected tap coefficient is set. It is also important to normalize with. Of course, the result is the same even if the selected tap coefficient other than optimizing the threshold value is normalized by the energy.

As described above, the tap number (5 or 6) required for modeling can be selected by the tap coefficient value. Moreover, the processing method is simple and the increase in the amount of calculation can be reduced. Finally, the change in the reception characteristic in the case of using the method of preliminarily selecting the tap number to 5 or 6 is obtained by simulation, and the result is shown in FIG. The characteristic curve corresponding to this method is a broken line indicated by "variable tap".

Incidentally, this example is also the result obtained by using the fading model in the hilly region similarly to the case where the number of taps is fixed to 5 and the case where the number of taps is fixed to 6. It can be seen that almost the same result as that of Equation 6 is obtained. Actually, the simulation results were obtained for various fading cases, and these characteristics hardly changed even when the tap number was changed.

By the way, comparing the ratio of the number of taps set to 6 and the ratio of the number of taps set to 5 from the simulation result, it is found that the number of taps is 6
It was only [%], and the subsequent processing was a tap number 5. Therefore, it can be seen that the processing amount is smaller than that when the number of taps is fixed to 6 and the processing is performed, and the characteristics can be maintained almost as they are.

According to the above configuration, the number of taps of the fading model (that is, 5) necessary for the transmission path estimation processing is used.
The tap coefficients h _{−2 to} h ₃ of one more tap number (that is, 6) than the number of
Of these, the tap coefficient h _-2 corresponding to the shortest delay time and the tap coefficient h ₃ corresponding to the longest delay time (that is, tap coefficients h _-2 and h _{3 at} both ends) are compared, and the tap coefficient of the smaller one is compared. By determining whether the tap number is set to 5 or 6 according to whether the value (energy) is sufficiently small, it is possible to reduce the signal processing amount without degrading the reception characteristic. It is possible to obtain a receiving device that can.

Further, since it is possible to appropriately obtain an optimum fading model according to the fading situation from the estimation result, the equalization characteristic can be further improved.
As a result, the demodulation efficiency of received data can be improved and the bit error rate can be reduced.

(3-3) Tap Number Determining Process 2 In the embodiment of the preceding paragraph, the six tap coefficients h _{-2 to} h ₃ obtained at the time of estimating the transmission path characteristics are used as they are as tap coefficients, or these tap coefficients h _{-2 to} h ₃ are used. Although one embodiment in which five taps not including either one of both ends is used as the tap coefficient has been described, a method of specifying the tap coefficient by the procedure shown in FIGS. 6 and 7 is also conceivable. The following will be described in order.

Also in the case of this example, the processing from step SP21 to step SP22 is the same as step SP explained in FIG.
This is the same as the processing from 1 to step SP2. The difference is in the subsequent step SP23 and subsequent steps. That is, when the six tap coefficients h _{-2 to} h ₃ are obtained, the largest value (energy) of these 6 tap coefficients h _{-2 to} h ₃ is detected in step SP23, and the corresponding value is detected. The position (delay time) is set as p1.

When this process is completed, the next step is SP.
24, the tap coefficient having the second largest value (energy) among the tap coefficients is detected, and the position (delay time) at which this value is obtained is set as p2. Similarly, when this process is completed, the process proceeds to step SP25,
The tap coefficient having the third largest value (energy) among the tap coefficients is detected, and the position (delay time) at which this value is obtained is set as p3.

When the three positions where a large value (correlation value) is obtained by the processing of these steps SP23 to SP25, the difference between the respective peak positions is obtained as shown in step SP26. Here, the difference | p1-p2 | between the position p1 where the largest value is obtained and the position p2 where the second largest value is obtained is defined as d1, and the position p2 where the second largest value is obtained and the third largest value. The difference | p2-p3 | with respect to the position p3 at which is obtained is d2, and the difference | p1-p3 | between the position p1 at which the largest value is obtained and the position p3 at which the third largest value is obtained is d3.

When these calculations are completed, the maximum value of these three values is obtained in the following step SP27, and this is defined as the maximum position difference maxd. When the maximum position difference maxd is obtained in this way, the process proceeds to step SP28 and the process for determining whether the peak positions are relatively distant or concentrated. Here, the threshold value is determined to be 4.
If it is determined that the maximum positional difference maxd is 4 or more, it is determined that a received signal with a large amplitude is obtained in a relatively wide delay time range, and the number of taps is set to 6 in step SP29.
To set.

Then, the six values obtained above are used for step SP.
At 30, the final tap coefficients h _{-2 to} h ₃ are set, and then the process proceeds to step SP31 to end the series of processes. For example, this processing is performed when the maximum position difference maxd of the peak positions p1 to p3 that gives the first to third largest values are separated to some extent as 3 and 7.

On the other hand, if it is determined in step SP28 that the maximum positional difference maxd is smaller than 4, it is determined that the received signals with large amplitude are concentrated in a relatively narrow range, and the process proceeds to step SP32 and the tap number is set to 5. Set. And in this case, these six tap coefficients h _-2 ~
The coefficients at 5 positions including the 3 peak positions obtained previously in h ₃ are finally set as the values of the 5 tap coefficients. At this time, the other two may be set according to the magnitude of the coefficient value, for example. Incidentally, this processing is performed when the peak positions are concentrated like 5 and 6.

According to the above arrangement, the number of taps of the fading model (that is, 5) necessary for the transmission path estimation processing is used.
The tap coefficients h _{−2 to} h ₃ of one more tap number (that is, 6) than the number of
The number of taps and the tap coefficient are set based on the degree of dispersion of the peak positions where a large value (energy) is obtained, so that the signal processing amount can be reduced without degrading the reception characteristics. Can be obtained.

(4) Other Embodiments In the above embodiment, the case where the number of taps of the fading model that gives the transmission path characteristic is reduced from 6 to 5 has been described, but the present invention is not limited to this. Instead, it can be applied to a case where the number is reduced to 5 or less. For example, if both tap coefficient values at both ends are small, it is possible to ignore all of these values and calculate the tap number as 4. By doing so, the processing time can be further shortened.

Further, in the above-mentioned embodiment, a case has been described in which six tap coefficients are obtained during the transmission path estimation process and the number of taps is set to 5 or 6 according to the value of the tap coefficient. However, the present invention is not limited to this, and by selecting only one of the tap coefficients at both ends, the processing amount may be finally reduced by a fading model with a tap number of 5. This is one of the methods for estimating the transmission path, but according to the simulation, when the transmission path is estimated by such a method, the characteristics can be slightly improved as compared with the conventional method.

Further, in the above-mentioned embodiment, the case of the portable telephone has been described, but the present invention is not limited to this, and can be widely applied to other receiving devices such as a car telephone. Furthermore, although the receiving apparatus has been described in the above-mentioned embodiments, the present invention is not limited to this, and can be widely applied to a reproducing apparatus for reproducing data recorded at high density from a recording medium.

[0057]

As described above, according to the present invention, the same pattern data as the known pattern data included in the transmission data transmitted to the transmission line on the receiving side is generated and received via this and the transmission line. Of the adaptive filter used for equalizing the received data based on this correlation value, and demodulating the received data using the adaptive filter after the change. Data can be demodulated at a low bit error rate even when received data is received in a situation where there is a lot of distortion due to aging or the like.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a transmission device according to the present invention.

FIG. 2 is a schematic diagram illustrating a configuration example of a receiving unit of a transmission device.

FIG. 3 is a schematic diagram showing a fading model.

FIG. 4 is a characteristic curve diagram showing bit error rate characteristics depending on the number of taps.

FIG. 5 is a flowchart showing an example of a tap number selection procedure.

FIG. 6 is a flowchart showing an example of a tap number selection procedure.

FIG. 7 is a flow chart showing an example of a tap number selection procedure.

[Explanation of symbols]

1 ... Cellular phone, 2 ... Antenna, 3 ... RF processing circuit, 3A ... RF / IF / detection circuit, 3B ... A / D conversion circuit, 4 ... Modulation / demodulation / data processing circuit, 4A ... Demodulation DSP, 5 ... Voice processing circuit, 5A ... Voice DS
P, 6 ... speaker, 7 ... microphone, 8 ... bus, 9 ...
... CPU, 10A ... RAM, 10B ... ROM, 11
…… Display operation part.

Claims (7)

[Claims] 1. A pattern generation means for generating the same pattern data as the known pattern data included in the transmission data transmitted to the transmission line, and reception data received via the transmission line and the pattern data. Correlation detection means for sequentially detecting correlation as a correlation value, and the number of taps of the adaptive filter used for equalization processing of the received data based on the correlation value is changed, and the received data is changed using the adaptive filter after the change. And a signal processing unit for demodulating. 2. The signal processing means estimates a plurality of tap coefficients using an adaptive filter having a predetermined tap number, and corresponds to the tap coefficient corresponding to the longest response time and the shortest response time among them. 2. The transmission apparatus according to claim 1, wherein the number of taps of the adaptive filter actually used is determined based on the tap coefficient to be used. 3. The signal processing means compares the tap coefficient corresponding to the longest response time with the tap coefficient corresponding to the shortest response time, finds which coefficient value is smaller, and determines that it is smaller. 3. The transmission apparatus according to claim 2, wherein when the value of the tap coefficient is sufficiently small, the adaptive filter configured by excluding the tap coefficient is set as the actually used adaptive filter, and the tap number is determined. 4. The signal processing means estimates a plurality of tap coefficients using an adaptive filter having a predetermined tap number, and compares the tap coefficients detected to have a relatively large coefficient value among them. 2. The transmission device according to claim 1, wherein an adaptive filter to be actually used is set based on whether the target concentration or the distance is set, and the tap number is determined. 5. The same pattern data as the known pattern data included in the transmission data transmitted to the transmission line is generated, and the correlation value between the pattern data and the reception data received via the transmission line is obtained. Processing, and processing for varying the number of taps of the adaptive filter used for equalization processing of the received data based on the correlation value obtained by the above processing and demodulating the received data using the adaptive filter after the change And a transmission method comprising: 6. A value of a tap coefficient corresponding to the longest response time and a tap coefficient corresponding to the shortest response time, of a plurality of tap coefficients estimated by using an adaptive filter having a predetermined tap number. 6. The transmission method according to claim 5, wherein the number of taps of the adaptive filter actually used is determined based on the above. 7. A plurality of tap coefficients are estimated using an adaptive filter having a predetermined number of taps, and tap coefficients from which a relatively large coefficient value is obtained are relatively concentrated or separated. 6. The transmission method according to claim 5, wherein the number of taps of the adaptive filter actually used is determined based on whether or not the taps are actually used.