JPH11234182A - Adaptive signal processor and adaptive signal processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an accurate filter coefficient for an adaptive filter, converged to an optimum value. SOLUTION: At the time of the input of passing signals 115 of an unknown system, signals before the unknown system passing are estimated in respective plural adaptive filters 102-104 corresponding to a filter coefficient updated by an initial coefficient generated at random, errors 116-118 of the estimated signals and known signals d (i) which are the signals before the unknown system passing are calculated and the filter coefficient converges corresponding to the errors 116-118. Then, a smallest one of the plural errors 116-118 at the time of convergence is detected and the filter coefficient corresponding to the smallest error (116, for instance) is selected from the plural adaptive filters 102-104. Thus, even in the case that the unknown system is provided with plural minimum values, the filter coefficient accurately converged to the optimum value is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to adaptive signal processing in general, and when unknown coefficients in an unknown system such as a radio wave propagation environment, for example, coefficients which affect a propagation signal are sequentially estimated, The present invention relates to an adaptive signal processing device and an adaptive signal processing method capable of accurately estimating a coefficient converged to an optimum value without being limited to a local minimum value even under a condition where a plurality of local minimum values exist in an unknown system.

[0002]

2. Description of the Related Art Conventionally, a filter for sequentially estimating an unknown coefficient in an unknown system from an input signal and an output signal input through the unknown system is called an adaptive filter. This adaptive filter is used for an echo canceller, adaptive equalization, and the like. For example, the adaptive filter estimates an impulse response in a propagation path (unknown system) between a base station and a mobile wireless terminal device in a mobile communication system, and The tap coefficient (filter coefficient) is updated.

FIG. 9 shows a block diagram of a conventional adaptive signal processing device. That is, the adaptive signal processing apparatus shown in FIG. 9 estimates coefficients of an unknown system using the above-described adaptive filter 901.

An input signal x (i) that has passed through an unknown system
The known signal d (i) output from the known signal memory unit 902 and the input signal x (i) in accordance with
Is added to the signal y (i) obtained by
, And sequentially corrects the filter coefficient of the adaptive filter 901 according to the obtained error e (i).

An algorithm for correcting the filter coefficient of the adaptive filter 901 is called an adaptive algorithm.
Various studies have been conducted, but due to the ease of realization, L
The MS (Least Mean Square) algorithm is often used.

[0006] The LMS algorithm will be briefly described. For the input signal x (i) at time i, the state vector of the adaptive filter 901 is represented by the following equation (1). x _N (i) = [x (i), x (i−1),..., x (i−N + 1)] ^T (1) where N is the impulse response length of the adaptive filter 901,
Represents that the vectors of x (i) to x (i-N + 1) are originally arranged vertically, but here are transposed to indicate that they are arranged side by side.

If the coefficient vector of adaptive filter 901 is hN (i), the output signal y (i) of adaptive filter 901 is given by the following equation (2). _{y (i) = h N (} i) x N T (i) ... (2) On the other hand, the impulse response of the unknown system 902 and a w _N output signals d (i) is expressed by the following formula (3) . _{d (i) = w N x} N T (i) ... (3) The error is expressed by the following equation (4). e (i) = d (i ) -y (i) = (w N -h N (i)) x T (i) ... (4) The coefficient of the LMS algorithm update, the following equation (5)
It is represented by h _N (i + 1) = h _N (i) + α × E (i) × x _N (i) (5) where α is a constant or a variable called a step gain. Therefore, in the adaptive signal processing device, in order to update the filter coefficient by applying the algorithm,
First, an unknown system is estimated using the known signal d (i) of a predetermined pattern, and the adder 903 calculates an output error e (i) of the adaptive filter 901 with respect to the unknown system. Then, using the error e (i) and the step gain α, the filter coefficient of the adaptive filter 901 is updated by the above equation (5).

[0008] By repeating this procedure, the error e
(I) becomes smaller, the filter coefficients of the adaptive filter 901 converge, and the unknown system 902 can be accurately estimated.

[0009]

However, in the above conventional adaptive signal processing apparatus, the unknown system 902 to be estimated has a plurality of minimum values 1001, 1002, and 1003 in the LMS algorithm as shown in FIG. In such a case, the convergence of the error to the minimum value (optimum value) 1004 is extremely deteriorated.

This is because the LMS algorithm updates the filter coefficient so as to reduce the error e (i) as described above.
Even if it begins to converge to the minimum values 1001 to 1003 instead of 04, the error e (i) apparently decreases,
This is because the minimum values 1001 to 1003 are determined to be the optimum values 1004.

An object of the present invention is to provide an adaptive signal processing apparatus and an adaptive signal processing method capable of accurately obtaining a filter coefficient of an adaptive filter converged to an optimum value.

[0012]

Means for Solving the Problems The present invention has the following arrangement to solve the above-mentioned problems.

According to a first aspect of the present invention, the adaptive signal processing apparatus estimates a signal before passing through the unknown system according to a filter coefficient updated with a coefficient initial value generated at random when a passing signal of the unknown system is input, An error between the estimated signal and a known signal that is a signal before passing through the unknown system is calculated, and a plurality of adaptive filters that converge the filter coefficients according to the calculated error are obtained. A minimum value detection unit that detects a minimum one of the errors at the time of convergence; and a selection unit that selects a filter coefficient corresponding to the detected minimum error from the plurality of adaptive filters. .

With this configuration, the input signals passing through the unknown system are adaptively processed in parallel by a plurality of adaptive filters using coefficient initial values, which are a plurality of random filter coefficients, and a plurality of adaptive filters obtained by this processing are obtained. Since the filter coefficient corresponding to the minimum value among the errors is selected, even when the unknown system has a plurality of minimum values, a filter coefficient that converges to the optimum value (minimum value) with high accuracy can be obtained.

According to a second aspect of the present invention, in the adaptive signal processing apparatus according to the first aspect, waveform distortion of a signal passing through an unknown system is compensated in accordance with a filter coefficient output from the selection means. And an equalizing means.

With this configuration, even when the unknown system has a plurality of local minima, the unknown coefficient of the unknown system is accurately estimated, and equalization is performed using a filter coefficient corresponding to the unknown coefficient obtained by the estimation. Since the processing is performed, the quality of the decoding result obtained by the equalization processing is improved.

According to a third aspect of the present invention, there is provided a receiving apparatus that optimally controls the directivity of the antenna according to a filter coefficient obtained by processing a received signal of the antenna by the adaptive signal processing apparatus according to the first aspect. That has the function of

With this configuration, even when the propagation path of the received radio wave has a plurality of local minima, the adaptive signal processing device estimates the coefficient that affects the radio wave on the propagation path with high accuracy. Since the directivity of the antenna is controlled by the corresponding filter coefficient to be optimal, the performance of the antenna pointing control is improved.

According to a fourth aspect of the present invention, the mobile radio terminal device optimizes the directivity of the antenna according to a filter coefficient obtained by processing a signal received by the antenna by the adaptive signal processing device according to the first aspect. Control function.

With this configuration, even when the propagation path of the received radio wave has a plurality of local minima, the adaptive signal processor estimates the coefficient that affects the radio wave on the propagation path with high accuracy. Since the directivity of the antenna is controlled by the corresponding filter coefficient to be optimal, the performance of the antenna pointing control is improved.

According to a fifth aspect of the present invention, there is provided a receiving apparatus according to the first aspect, wherein the adaptive signal processing apparatus according to the first aspect is connected to a plurality of antennas and processes the received signal of the antenna. 2. The adaptive signal processing according to claim 1, wherein: antenna directivity control means for optimally controlling the directivity of a plurality of antennas; synthesizing means for synthesizing the reception signals of the respective antennas which are optimally controlled; And an equalizing means for compensating for waveform distortion of a signal passing through the propagation path in accordance with a filter coefficient obtained by processing by the device.

With this configuration, even when the propagation path of the received radio wave has a plurality of local minima, the adaptive signal processing device estimates the coefficient that affects the radio wave on the propagation path with high accuracy. By the corresponding filter coefficients, the directivity of the plurality of antennas is controlled so as to be optimal. Further, the optimum filter coefficient is obtained by the adaptive signal processing device from the signal obtained by combining the reception signals of the respective antennas. Since the equalization processing is performed according to the coefficients, the quality of the decoding result obtained by the equalization processing is improved.

According to a sixth aspect of the present invention, there is provided a base station apparatus according to the filter coefficient obtained by processing a reception signal of an antenna by the adaptive signal processing apparatus of the first aspect connected to a plurality of antennas. 2. The adaptive signal according to claim 1, wherein: antenna directivity control means for optimally controlling the directivity of the plurality of antennas; synthesizing means for synthesizing the reception signals of the respective antennas which are optimally controlled; And an equalizing means for compensating for waveform distortion of the synthesized signal in accordance with a filter coefficient obtained by processing in the processing device.

With this configuration, even when the propagation path of a received radio wave has a plurality of local minima, the adaptive signal processing device estimates the coefficient that affects the radio wave on the propagation path with high accuracy. By the corresponding filter coefficients, the directivity of the plurality of antennas is controlled so as to be optimal. Further, the optimum filter coefficient is obtained by the adaptive signal processing device from the signal obtained by combining the reception signals of the respective antennas. Since the equalization processing is performed according to the coefficients, the quality of the decoding result obtained by the equalization processing is improved.

According to a seventh aspect of the present invention, in the base station apparatus according to the sixth aspect, a channel decoder for channel-decoding the signal compensated by the equalizing means and an audio signal And an audio decoder for decoding in the case of (1).

With this configuration, a voice or non-voice signal is obtained after the equalization process is performed according to the optimum filter coefficient as in the above-described claim 6, so that a high-quality voice or non-voice signal is obtained. be able to.

In the adaptive signal processing method according to the present invention, each of the plurality of adaptive filters has a filter coefficient updated with a coefficient initial value randomly generated when a pass signal of an unknown system is input. Estimate the signal before passing through the unknown system, calculate the error between the estimated signal and the known signal that is the signal before passing through the unknown system, and after converging the filter coefficient according to the calculation error, The smallest one of the plurality of errors at the time of convergence is detected, and a filter coefficient corresponding to the detected smallest error is selected from the plurality of adaptive filters.

According to this method, an adaptive process is performed in parallel using a plurality of random coefficient initial values, and a filter coefficient that minimizes an error is selected. Even in the case of having the filter coefficient, the filter coefficient converged to the optimum value with high accuracy can be obtained.

According to a ninth aspect of the present invention, there is provided the adaptive signal processing method according to the eighth aspect, wherein the waveform distortion of the passing signal of the unknown system is reduced according to a filter coefficient selected from a plurality of adaptive filters. I compensated.

According to this method, adaptive processing is performed in parallel using a plurality of random coefficient initial values. Equalization processing is performed according to a filter coefficient that minimizes an error, and then decoding is performed. As a result, the quality of the decoding result is improved.

According to a tenth aspect of the present invention, in the antenna directivity control method in the receiving apparatus, each of the plurality of adaptive filters to which the received signal of the antenna is inputted is updated with the coefficient initial value generated at random. Estimating a signal before passing through the propagation path corresponding to the received signal according to the filter coefficient, calculating an error between the estimated signal and a known signal that is a signal before passing through the propagation path, After converging the filter coefficients, a minimum one of a plurality of errors at the time of the convergence is detected, and a filter coefficient corresponding to the detected minimum error is selected from the plurality of adaptive filters. The directivity of the antenna is optimally controlled according to the filter coefficient.

According to this method, even when the propagation path of a received radio wave has a plurality of local minima, a coefficient that accurately affects the radio wave of the propagation path can be estimated by adaptive signal processing.
By the filter coefficient corresponding to the estimated coefficient,
Since the directivity of the antenna is controlled to be optimal,
The performance of antenna pointing control is improved.

[0033]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the adaptive signal processing device and the adaptive signal processing method according to the present invention will be specifically described below with reference to the drawings.

(Embodiment 1) FIG. 1 is a block diagram showing an adaptive signal processing apparatus according to Embodiment 1 of the present invention.

The adaptive signal processing apparatus 101 shown in FIG. 1 includes a plurality of adaptive filters, that is, first and second,..., Nth adaptive filters 102, 103, and 104, and a minimum value detector 105.
, Selection unit 106, input terminal 107, output terminal 10
8 is provided. Each adaptive filter 1
02 to 104 have the same configuration. As will be described on behalf of the first adaptive filter 102, the coefficient updating unit 109
, An estimation unit 110 and an error calculation unit 111.

The coefficient updating unit 109 of each of the adaptive filters 102 to 104 has an input signal 115 input to the input terminal 107 and a coefficient initial value 11 randomly generated.
2, 113, 114 are input.
However, it is assumed that the input signal 115 is a signal that has passed through, for example, an unknown system corresponding to a propagation path between a base station and a mobile wireless terminal in a mobile communication system.

In such a configuration, first, an input signal 115 is applied to each adaptive filter 102 through an input terminal 107.
When input to the coefficient updating units 109 to 104, the coefficient updating unit 109 updates the filter coefficients according to the above equation (5) described in the conventional example. Here, in the initial state,
Each coefficient updating unit 109 calculates a coefficient initial value 11
The filter coefficients are updated using 2 to 114, and the updated filter coefficients are output to the estimation unit 110.

The estimating unit 110 estimates the signal passing through the unknown system according to the above equation (2), and outputs the estimated signal to the error calculating unit 111. The error calculator 111 calculates
The error between the estimated signal and the known signal d (i) described with reference to FIG. 9 in the conventional example is calculated using the above equation (4),
The calculated errors 116, 117, and 118 are fed back to the coefficient updating unit 109 and output to the minimum value detecting unit 105.

The coefficient updating unit 109 and the estimating unit 11
0 and the operation of the error calculator 111
This process is repeated until the filter coefficients 2 to 104 converge. When the filter coefficients converge, the minimum value detection unit 105
Detects the smallest one of the errors 116, 117, and 118 at the time of convergence, and outputs the detected smallest error 119 to the selector 106.

The selection unit 106 controls each of the adaptive filters 102 to
A filter coefficient (for example, 120) corresponding to the minimum error 119 is selected from the filter coefficients 120, 121, and 122 output from the coefficient update unit 109 of 104, in other words, a filter coefficient 120 that minimizes the error is selected. , And outputs the selected filter coefficient 120 to the output terminal 108 as an output signal 123.

As described above, according to the first embodiment, the coefficient initial values 112 to 1 which are a plurality of random filter coefficients
A plurality of adaptive filters 102 to 104 using the filter 14 adaptively process the input signal 116 passing through the unknown system in parallel, and obtain errors 116 and 11 obtained by this process.
Since the filter coefficient corresponding to the minimum value among 7, 118 is selected and output, even when the unknown system has a plurality of minimum values, the optimum value (minimum value) is accurately determined.
Can be obtained.

(Embodiment 2) FIG. 2 is a block diagram showing an adaptive signal processing apparatus according to Embodiment 2 of the present invention. However, in the second embodiment shown in FIG. 2, the same parts as those in the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and the description thereof will be omitted.

That is, the adaptive signal processing device shown in FIG. 2 is configured by connecting the equalizer 201 to the output terminal of the adaptive signal processing device 101 described in the first embodiment. In such a configuration, first, the signal 115 that has passed through the unknown system is input to the adaptive signal processing device 101 via the input terminal 107. The adaptive signal processing device 101
By performing the adaptive processing as described in the first embodiment, optimal filter coefficients 123 are output to equalization section 201.

The equalizing section 201 has a filter coefficient 123
, An equalization process for compensating for waveform distortion of the input signal 115 is performed, and the signal after this process is decoded. Then, the decoded signal is output to the output terminal 204 as the output signal 203.

As described above, according to the second embodiment, even when the unknown system has a plurality of minimum values, the unknown coefficient of the unknown system is accurately estimated by the adaptive signal processing device 101, and the unknown coefficient is obtained by this estimation. Since the equalization processing is performed using the filter coefficients corresponding to the unknown coefficients, the quality of the decoding result obtained by the equalization processing can be improved.

(Embodiment 3) FIG. 3 is a block diagram showing a receiving apparatus using an adaptive signal processing apparatus according to Embodiment 3 of the present invention.

The receiving apparatus shown in FIG.
1 and an adaptive signal processing device 101 described in the first embodiment for a receiving device including an antenna directivity control unit 302 for controlling the directivity of the antenna 301.

In such a configuration, the antenna 301
Is input to the adaptive signal processing apparatus 101, the adaptive signal processing apparatus 101 obtains an optimal filter coefficient 304, and the antenna directivity control unit 3
02 is output. However, the filter coefficient 304 corresponds to a coefficient that affects the radio wave of the propagation path (unknown system) through which the received signal 303 has been propagated.

The antenna directivity control unit 302 controls the directivity of the antenna 301 according to the filter coefficient 304 so that the influence of fading due to an interference wave or the like is removed. In addition, as a receiving device in which the directivity of the antenna is controlled as described above, for example, a mobile wireless terminal device is used.

As described above, according to the third embodiment, even when the propagation path of the radio wave received by the receiving apparatus has a plurality of minimum values, adaptive signal processing apparatus 101 can accurately influence the radio wave on the propagation path. The coefficient to be given is estimated, and the filter is controlled so that the directivity of the antenna is optimized by using the filter coefficient corresponding to the coefficient obtained by the estimation, so that the performance of the antenna pointing control can be improved.

(Embodiment 4) FIG. 4 is a block diagram showing a receiving apparatus using a plurality of adaptive signal processing apparatuses according to Embodiment 4 of the present invention.

The receiving apparatus shown in FIG. 4 includes a plurality of antennas 401, 402, 403 and a plurality of antennas 40,
A plurality of antenna directivity control units 404, 405, and 406 for controlling the directivity of the adaptive signal processing units 1 to 403; adaptive signal processing devices 407, 408, and 409 having the same functions as the adaptive signal processing device 101 described in the first embodiment; It comprises a combining unit 410, an adaptive signal processing device 411 having the same function, and an equalizing unit 412.

In such a configuration, each antenna 40
Signals 414, 415, 416 received at 1-403
Is input to each of adaptive signal processing devices 407 to 409, optimal filter coefficients 417, 418, and 419 are obtained in adaptive signal processing devices 407 to 409, and output to antenna directivity control units 404 to 406. Is done. However, the filter coefficients 417 to 419 correspond to the reception signals 414 to 414.
Reference numeral 416 corresponds to a coefficient that affects the radio wave of the propagation path (unknown system) that has been propagated.

Each antenna directivity control section 404 to 406
Controls the directivity of each of the antennas 401 to 403 according to the filter coefficients 417 to 419 such that the influence of fading due to interference waves or the like is removed. Received signals 420 and 42 of antennas 401 to 403 that have performed this control
1, 422 are combined in a combining unit 411, and the combined signal 423 is output to the adaptive signal processing device 411, where
After an optimum filter coefficient 424 is obtained from the combined signal 423, the result is output to the equalizer 412.

The equalizer 412 performs equalization processing for compensating for waveform distortion in a composite signal of a plurality of received signals according to the filter coefficient 424, and then decodes the received signal.
An output signal 425 as a result of the decoding is output to an output terminal 413.

As described above, according to the fourth embodiment, even when the propagation path of the radio wave received by the receiving device has a plurality of local minima, the adaptive signal processing device accurately affects the radio wave on the propagation path. The coefficients are estimated, and the filter coefficients corresponding to the coefficients obtained by the estimation are used to control so that the directivity of the plurality of antennas is optimized, and also adaptive from the signal obtained by combining the received signals of the antennas. Since the optimal filter coefficient is obtained by the signal processing device and the equalization processing is performed according to the filter coefficient, the quality of the decoding result obtained by the equalization processing can be improved.

(Embodiment 5) FIG.5 shows a block diagram of a base station using a plurality of adaptive signal processing devices according to Embodiment 5 of the present invention. However, in the fifth embodiment shown in FIG. 5, the same parts as those in the fourth embodiment shown in FIG. 4 are denoted by the same reference numerals, and the description thereof will be omitted.

The base station 501 shown in FIG. 5 is configured by including a channel decoder 502 and a voice decoder 503 in the components of the receiving apparatus shown in FIG.

That is, the channel decoder 502 decodes the decoded signal 425 output from the equalizer 412 for each channel. If the decoded signal 504 is an audio signal, the decoded signal 504 is decoded by the audio decoder 503 and then output. Output to terminal 505. On the other hand, the decode signal 5
If 04 is a non-speech signal, it is output to the output terminal 505 as it is.

As described above, according to the fifth embodiment, after performing equalization processing in the base station according to the optimum filter coefficient as in the fourth embodiment, a speech or non-speech signal is obtained. Therefore, a high-quality voice or non-voice signal can be obtained.

(Embodiment 6) FIG. 6 is a flowchart for explaining an adaptive signal processing method according to Embodiment 6 of the present invention.

First, at step 601, a plurality of coefficient initial values required for updating the filter coefficients of the adaptive filter for estimating the unknown coefficients in the unknown system are randomly generated.

Next, in steps 602, 603 and 604, updating of the respective coefficient initial values and the filter coefficients using the above equation (5) are performed in parallel. Step 60
In 5,606,607, the signal before passing through the unknown system is estimated in parallel using the above equation (2) according to the updated filter coefficients.

In steps 608, 609 and 610, the error between the estimated signal and the known signal before passing through the unknown system according to the above equation (4) is calculated in parallel. In steps 611, 612, and 613, it is determined whether or not the update of the filter coefficient due to the convergence of the filter coefficient has been completed according to the calculated errors.

If the result of this determination is NO, step 6
Returning to steps 02 to 604, the processing is performed. If YES, the minimum error among the calculated errors is determined and detected in step 614. Finally, in step 615, the filter coefficient corresponding to the detected minimum error is
Select and output from each adaptive filter.

As described above, according to the sixth embodiment, adaptive processing is performed in parallel using a plurality of random coefficient initial values, and a filter coefficient that minimizes an error is selected and output. Has a plurality of minimum values, it is possible to obtain a filter coefficient converged to an optimum value with high accuracy.

(Embodiment 7) FIG. 7 is a flowchart for explaining an adaptive signal processing method according to Embodiment 7 of the present invention. However, in the seventh embodiment shown in FIG. 7, the processing of steps 701 to 715 is the same as that of the sixth embodiment described with reference to FIG.

Accordingly, in step 716, according to the filter coefficient corresponding to the minimum error output in step 715, equalization processing for compensating for waveform distortion of the signal passed through the unknown system is performed. , The signal subjected to the equalization processing is decoded.

As described above, according to the seventh embodiment, adaptive processing is performed in parallel using a plurality of random coefficient initial values, and equalization processing is performed in accordance with a filter coefficient that minimizes an error. After decoding, the quality of the decoding result can be improved.

(Embodiment 8) FIG. 8 is a flowchart illustrating an operation of antenna directivity control in a receiving apparatus to which an adaptive signal processing method according to Embodiment 8 of the present invention is applied.

First, in step 801, a plurality of coefficient initial values required for updating filter coefficients of an adaptive filter for estimating unknown coefficients in a radio wave propagation path are randomly generated.

Next, in steps 802, 803, and 804, updating of each coefficient initial value and updating of the filter coefficient using the above equation (5) are performed in parallel. Step 80
In 5,806 and 807, the signal before passing through the propagation path is estimated in parallel using the above equation (2) according to the updated filter coefficients.

At steps 808, 809 and 810, the error between the estimated signal and the known signal before passing through the propagation path according to the above equation (4) is calculated in parallel. Step 8
In 11, 812, 813, it is determined whether or not the update of the filter coefficient due to the convergence of the filter coefficient is completed according to the calculated error.

If the result of this determination is NO, step 8
The processing returns to 02 to 804, and if YES, in step 814, the smallest one of the calculated errors is determined and detected. In step 815, a filter coefficient corresponding to the detected minimum error is selected from each adaptive filter and output. Finally, the directivity of the antenna is controlled according to the filter coefficient such that the influence of fading due to an interference wave or the like is removed.

As described above, according to the eighth embodiment, even when the propagation path of the received radio wave has a plurality of local minima, the coefficient which affects the radio wave of the propagation path with high accuracy can be estimated by the adaptive signal processing. Since the control is performed so that the directivity of the antenna is optimized using the filter coefficient corresponding to the coefficient obtained by this estimation, the performance of the antenna pointing control can be improved.

[0076]

As is apparent from the above description, according to the present invention, even when the unknown system has a plurality of local minima, the filter coefficients of the adaptive filter converged to the optimum value (minimum value) with high accuracy can be obtained. Obtainable.

Even when the unknown system has a plurality of minimum values, the unknown coefficients of the unknown system are accurately estimated, and the equalization process is performed using the filter coefficients corresponding to the unknown coefficients obtained by the estimation. Since it is performed, the quality of the decoding result obtained by the equalization processing can be improved.

Further, even when the propagation path of the received radio wave has a plurality of local minima, the adaptive signal processor estimates the coefficient which affects the radio wave on the propagation path with high accuracy, and corresponds to the estimated coefficient. Since the directivity of the antenna is controlled by the filter coefficient to be optimal, the performance of the antenna directivity control can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram of an adaptive signal processing device according to a first embodiment of the present invention.

FIG. 2 is a block diagram of an adaptive signal processing device according to a second embodiment of the present invention.

FIG. 3 is a block diagram of a receiving device using an adaptive signal processing device according to Embodiment 3 of the present invention.

FIG. 4 is a block diagram of a receiving apparatus using a plurality of adaptive signal processing apparatuses according to Embodiment 4 of the present invention.

FIG. 5 is a block diagram of a base station using a plurality of adaptive signal processing devices according to Embodiment 5 of the present invention.

FIG. 6 is a flowchart for explaining an adaptive signal processing method according to Embodiment 6 of the present invention.

FIG. 7 is a flowchart for explaining an adaptive signal processing method according to Embodiment 7 of the present invention.

FIG. 8 is a flowchart for explaining an operation of antenna directivity control in a receiving apparatus to which an adaptive signal processing method according to Embodiment 8 of the present invention is applied.

FIG. 9 is a block diagram of a conventional adaptive signal processing device.

FIG. 10 is an explanatory diagram of a plurality of local minimum values of an estimated unknown system.

[Explanation of symbols]

 Reference Signs List 101 adaptive signal processing device 102 to 104 adaptive filter 105 minimum value detection unit 106 selection unit 112 to 114 coefficient initial value 115 input signal 117 to 119 error

Claims (10)

[Claims] When a signal passing through an unknown system is input, a signal before passing through the unknown system is estimated according to a filter coefficient updated with a coefficient initial value generated at random, and the estimated signal and the signal passing through the unknown system are estimated. A plurality of adaptive filters that calculate an error from a known signal that is a previous signal and converge the filter coefficient in accordance with the calculated error, and a minimum of the convergence errors obtained by the plurality of adaptive filters. An adaptive signal processing apparatus, comprising: a minimum value detecting means for detecting a minimum value; and a selecting means for selecting a filter coefficient corresponding to the detected minimum error from the plurality of adaptive filters. 2. The adaptive signal processing device according to claim 1, further comprising: an equalizing unit that compensates for waveform distortion of a signal passing through the unknown system according to the filter coefficient output from the selecting unit. 3. A function for optimally controlling the directivity of the antenna according to a filter coefficient obtained by processing a received signal of the antenna by the adaptive signal processing device according to claim 1. Receiving device. 4. A function for optimally controlling the directivity of the antenna according to a filter coefficient obtained by processing a received signal of the antenna by the adaptive signal processing device according to claim 1. Mobile radio terminal device. 5. The directivity of the plurality of antennas is optimized according to a filter coefficient obtained by processing a reception signal of the antenna by the adaptive signal processing device according to claim 1, which is connected to each of the plurality of antennas. An antenna directivity control means for controlling, a synthesis means for synthesizing the reception signals of the respective optimally controlled antennas, and a filter obtained by processing the synthesized signal by the adaptive signal processing device according to claim 1. An equalizing means for compensating for waveform distortion of a signal passing through a propagation path in accordance with the coefficient. 6. The directivity of the plurality of antennas is optimized according to a filter coefficient obtained by processing a reception signal of the antenna by the adaptive signal processing device according to claim 1 connected to each of the plurality of antennas. An antenna directivity control means for controlling, a synthesis means for synthesizing the reception signals of the respective optimally controlled antennas, and a filter obtained by processing the synthesized signal by the adaptive signal processing device according to claim 1. A base station apparatus comprising: equalizing means for compensating for waveform distortion of the combined signal according to a coefficient. 7. The apparatus according to claim 6, further comprising a channel decoder for channel-decoding the signal compensated by the equalizing means, and an audio decoder for decoding when the channel-decoded signal is an audio signal. The base station device as described in the above. 8. In each of a plurality of adaptive filters, when a pass signal of an unknown system is input, a signal before passing through the unknown system is estimated in accordance with a filter coefficient updated with a coefficient initial value generated at random. Calculating an error between the estimated signal and a known signal that is a signal before passing through the unknown system, converging the filter coefficient according to the calculated error, and then calculating a minimum of a plurality of errors during the convergence. , And selecting a filter coefficient corresponding to the detected minimum error from the plurality of adaptive filters. 9. The adaptive signal processing method according to claim 8, wherein waveform distortion of a passing signal of the unknown system is compensated according to a filter coefficient selected from a plurality of adaptive filters. 10. In each of a plurality of adaptive filters to which a reception signal of an antenna is input, before passing through a propagation path corresponding to the reception signal according to a filter coefficient updated with a randomly generated coefficient initial value. Is estimated, an error between the estimated signal and a known signal that is a signal before passing through the propagation path is calculated, and after converging the filter coefficient according to the calculated error, a plurality of Detecting a minimum error, selecting a filter coefficient corresponding to the detected minimum error from the plurality of adaptive filters, and optimally controlling the directivity of the antenna according to the selected filter coefficient; An antenna directivity control method in a receiving apparatus, comprising: