JP4545683B2 - Interference wave removing apparatus and interference wave removing method - Google Patents

Description

  The present invention relates to an interference wave removal apparatus and an interference wave removal method using a complex tap selection adaptive filter that transmits a broadband signal in advance and estimates a delay time in which an impulse response value of an unknown system (an unknown propagation system) is localized. .

  As examples of applying unknown system identification by an adaptive filter, an interference wave canceller, an echo canceller, an adaptive equalizer, and the like are known. Here, a tap selection adaptive filter used as one method of an interference wave removing apparatus will be described (for example, see Patent Document 1). The tap selection adaptive filter, when identifying an unknown system having an impulse response localized on the time axis, operates the sub-adaptive filter having M taps only in the delay time where the impulse response is localized. Is.

  This tap selection adaptive filter is composed of a plurality of sub-adaptive filters having a small number of tap stages arranged in cascade.

  In the unknown system identification process, first, the time for which the impulse response is localized is estimated. At this time, a delay time evaluation function is calculated using two types of loads, ie, a complex value and a load whose phase is changed, to obtain a delay time. The delay time here is normalized by the sample interval. Next, a load corresponding to the impulse response value is obtained. The load is obtained by operating an adaptive filter with M taps in the delay time estimated by the connected delay time estimation devices. The delay time estimation and load estimation processes are sequentially performed from the first-stage sub-adaptive filter group to the K-th-stage sub-adaptive filter group. The system identification of an unknown system is performed with such a configuration.

  When the output of the unknown system is an impulse response localized on the time axis, the number of tap stages of the conventional adaptive filter is required only to match the impulse response waveform on the time axis. If the number of tap stages of the adaptive filter increases due to the wide band of interference waves or the spread of the time axis, etc., the computation load increases and processing takes time. Therefore, it is considered that a tap selection adaptive filter that operates the adaptive filter only in the delay time in which the impulse response is localized is advantageous from the viewpoint of reducing the calculation load.

  When the transmission signal has a narrow band and periodicity, the delay time evaluation function of the tap selection adaptive filter also has periodicity, so that the delay time estimation error may occur and the accuracy of system identification of unknown systems may deteriorate. It was seen. This cause will be described with reference to FIGS. 11 and 12 according to the theory of the delay time evaluation function. In order to simplify the explanation, it is assumed that the plant noise is sufficiently small and the number of delay times in which the impulse response values of the unknown system are localized matches the number of sub-adaptive filter groups.

The delay time evaluation function performs the following processing. Here, a case where the delay time of the k-th stage is estimated will be described. In FIG. 11, the signals obtained by delaying the transmission signal x (t) by the delay circuit 90 are multiplied by weights g and g ^* having appropriate complex values by multipliers 91 and 92, respectively. Subtractors 93 and 94 respectively subtract from e _k−1 (t). These are obtained as mean square values by means of mean square value calculation devices 95 and 96 (steps 901 and 902 in FIG. 12). The load g ^* is a complex conjugate of the load g. The operation so far is performed so that the maximum search value d of the delay time is satisfied so as to satisfy the time axis of the impulse response of the unknown system, and the delay circuit 90 changes the d delay times to obtain two mean square values. These two mean square values are added by the adder 97 for each delay time and input to the mean square value evaluation circuit 98 (step 903 in FIG. 12).

Deviation between adding the mean square value and its primary approximate line calculated delay time becomes maximum, and outputs it as the delay time d _k. By performing the above operation, the delay time d _k where the impulse response value is localized is estimated.

An equation for obtaining this mean square error is shown in Equation (1). The equation (1) is an expression showing the delay time evaluation function obtained by adding the mean square value of the two error signal as a principle for determining the delay time d _k.

However, when k = 1, the error signal e _k−1 (t) is the output f (t) of the unknown system. Here, d is a delay time searched in the delay circuit 90, e _k (t) is an error signal, t is a time factor of the digital signal, g is a weight of the adaptive filter, and g ^* is a complex conjugate with the weight g of the adaptive filter. N ′ is the number of samples using the adaptive filter weight g, NN ′ is the number of samples using the adaptive filter weight g ^* , and x is the transmission signal.

Next, a first-order approximate line of the mean square value of the obtained error signal is obtained. The search delay time d at which the deviation between the mean square value of the straight line and the error signal becomes the maximum value is the delay time (^) d _k (where (^) d indicates that ^ is on d). . FIG. 13 schematically shows this delay time estimation method. In FIG. 13, the solid line represents the mean square value of the error signal, and the dotted line represents the linear approximation line. The horizontal axis shows the delay time. In the delay time indicated by the arrow in the figure, the deviation between the mean square value and the linearly approximated straight line takes the maximum value, and this delay time is set as the k-th delay time. The principle that the delay time can be estimated by the maximum value of the deviation can be expressed by the following equation by modifying the equation (1).

  Here, the first term on the right side of Equation (2) is as follows.

Here, _ck is an impulse response value of an unknown system, r _xx is an autocorrelation function of a transmission signal having N ′ samples, r ′ _xx is an autocorrelation function of a transmission signal having N−N ′ samples, and r _nx , R _xn is a cross-correlation function between a transmission signal and plant noise, K is the number of sub-adaptive filter groups, and * is a complex conjugate.

  In Expression (2), the first term on the right side is a constant value. This is because, from equation (3), since white noise is considered in the transmission signal, the autocorrelation value of the transmission signal is 0, the cross-correlation value of the transmission signal and plant noise is also 0, and the remaining terms are constant values. Because it takes.

The second term and the third term on the right side of Equation (2) are such that when the delay time becomes d = d _k during the delay time search using d as a parameter, r _xx (d−d _k ) has a maximum value, The deviation from the approximate line is maximized. Search delay time d at this time (^) impulse response value by the d _k can be used to estimate the delay time d _k localized (step 904 in FIG. 12).

JP 2001-308751 A

  However, in the interference wave canceller using the complex tap selection adaptive filter described so far, when the transmission signal has a narrow band and periodicity, the delay time evaluation function also has periodicity and estimates a non-true delay time. Value was seen. Due to this incorrect delay time estimate, it was impossible to estimate the impulse response value of the unknown propagation system, and the suppression performance was degraded.

The problem that the amplitude of the evaluation function Φ ′ (d _k ) for estimating the delay time oscillates when the transmission signal is a signal having periodicity such as an unmodulated sine wave will be described.

When attention is paid to the second term and the third term on the right side of the delay time evaluation function Φ ′ (d) shown in Expression (2), r _xx and r ′ _xx take the maximum values when d = d _k .

Hereinafter, a case where the transmission signal is a complex sine wave will be considered. Sub adaptive filter number set for the sake of simplicity K = 1, the impulse response also c ₁ only one type of the unknown system. The autocorrelation functions r _xx and r ′ _xx are considered to be the same although the number of samples is different. Taking these into account, the second term and the third term on the right side of Equation (2) are transformed as follows.

Here, if the transmission signal is a complex sine wave e ^−j2πft , Equation (4) is as follows.

Here, the inside of <> represents an expected value calculation. In formula (5), since periodicity is generated by e ^{−j2πf (d−d1)} regardless of the search for d, it can be seen that the amplitude of Φ ′ (d) oscillates during the delay time search. The ^{superscript (d−d1)} in e ^{−j2πf (d−} d1) is (d−d ₁ ). From these facts, it can be understood that when the transmission signal is a periodic function, the evaluation function has periodicity, and therefore the estimation of the delay time may be erroneous. FIG. 14 schematically shows the delay time estimation method at this time.

  The present invention has been made to solve the above-described problems, and its purpose is to use a complex tap selection adaptive filter having a low amount of computation when a transmission signal has a narrow band and periodicity. Thus, an interference wave removing apparatus and an interference wave removing method capable of removing an interference wave from an unknown system in correspondence with signal processing of a narrowband signal are obtained.

  An interference wave canceller according to the present invention includes a wideband signal generator that generates a wideband signal, a transmission signal generator that generates a transmission signal, and a broadband signal generator that transmits the wideband signal to an unknown system. A switch that outputs a first control signal when switched, and outputs a second control signal when switched to the transmission signal generator in order to transmit the transmission signal to the unknown system, and an estimated delay time When the first control signal is input to the memory for storing the delay time, the delay time is estimated based on the delay time evaluation function, the estimated delay time is stored in the memory, and the second control signal is When input, the estimated delay time is called from the memory, a weight at the estimated delay time is estimated using an adaptive algorithm, and an error signal is obtained from the estimated delay time and the estimated weight. And flop selection adaptive filter, in which the transmission signal is provided a subtracter reduce interference signals by subtracting the output of the complex tap selection adaptive filter from the received signal that has passed through the unknown system.

  The interference wave elimination method according to the present invention includes a step of transmitting a wideband signal to an unknown system, estimating a delay time based on a delay time evaluation function of the wideband signal, and storing the estimated delay time in a memory; Transmitting to the unknown system, calling the estimated delay time from the memory, estimating a load at the estimated delay time using an adaptive algorithm, obtaining an error signal from the estimated delay time and the estimated weight, and transmitting the transmitted signal to the unknown And a step of subtracting the error signal from the received signal that has passed through the system to suppress the interference wave.

  The interference wave canceling apparatus and method according to the present invention is adapted to signal processing of a narrowband signal in order to use a complex tap selection adaptive filter having a low calculation amount when the transmission signal has a periodicity in a narrowband, There is an effect that interference waves from unknown systems can be removed.

Embodiment 1 FIG.
An interference wave canceling apparatus according to Embodiment 1 of the present invention will be described with reference to FIGS. 1 is a diagram showing a configuration of an interference wave canceling apparatus according to Embodiment 1 of the present invention. FIG. 2 is a diagram showing the configuration of the complex tap selection adaptive filter of the interference wave canceller according to Embodiment 1 of the present invention. In addition, in each figure, the same code | symbol shows the same or equivalent part.

In FIG. 1, an interference wave canceling apparatus according to Embodiment 1 of the present invention includes a wideband signal generator 1 that generates a wideband signal, a transmission signal generator 2 that generates a transmission signal, a wideband signal generator 1, and a transmission. A switch 3 for switching the signal generator 2, a transmission antenna 4 for transmitting a broadband signal or a transmission signal x (t) to an unknown system, and an estimation of the delay time when the impulse response of the unknown system is localized A complex tap selection adaptive filter 5 adapted to adaptively operate only taps corresponding to the delay time, and a complex number, and a memory 6 for storing the estimated delay time, etc.
A receiving antenna 7 for receiving a desired wave, an adder 8 for adding the received signal f (t) and plant noise n (t), desired wave e _k by subtracting the output of the adaptive filter 5 from the output of the adder 8 Subtractors 9 (9 ₁ ,..., 9 _k ,..., 9 _K ) that output (t) are provided. Note that transmission means such as a modulator is not shown between the switch 3 and the transmission antenna 4, and reception means such as a demodulator is not shown between the reception antenna 7 and the adder 8. . An arrow from the transmission antenna 4 to the reception antenna 7 represents a leakage interference wave, that is, a transmission signal that has passed through an unknown system.

Further, in FIG. 2, the complex tap selection adaptive filter 5, the delay time estimating unit ₅₁ 1, _..., 51 k, ..., and 51 _K, the adaptive filter ₅₂ 1 M stage, _..., 52 k, ..., and 52 _K Is provided. 1 and 2, the dotted line represents a control signal.

  Next, the operation of the interference wave canceller according to Embodiment 1 will be described with reference to the drawings. FIG. 3 is a flowchart showing the operation of the interference wave canceller according to Embodiment 1 of the present invention.

  Here, for the sake of simplicity, the description will be made based on a situation where the plant noise is sufficiently small. In the interference wave canceling apparatus according to the first embodiment, when the number of sub-adaptive filter groups is K = 1, the delay time is estimated in advance with a broadband signal, and the load is applied when each interference wave is incident. Ask.

Consider a case where a broadband signal (white noise or the like) is transmitted from the broadband signal generator 1 in advance. At this time, the switch 3 switches to the wideband signal generator 1 side. Further, the switch 3 uses the first control signal (1) so that the delay time estimation device 51 ₁ , the M-stage adaptive filter 52 ₁ , and the memory 6 can distinguish that the delay time is estimated using the wideband signal. Send a dotted line). The operation of the complex tap selection adaptive filter 5 is shown in FIG. Here, the “processing performed in advance with a broadband signal” in FIG. 3 is performed.

When a wideband signal is transmitted in steps 101, 102, and 103, the autocorrelation function _rxx of the transmission signal in equation (2) has a correlation value peak only at d = d _i . In this case, the delay time evaluation function Φ ′ (d) has only one peak.

Next, in step 104, and the only time delay estimate the search parameters _{d 1} having a peak _{(^) d 1-pre.} (The subscript _1-pre represents 1_pre.) This delay time estimated value (^) d _1-pre is stored in the memory 6. Thus, the preprocessing ends.

When the interference wave is actually suppressed, the “process when suppressing interference wave” in FIG. 3 is performed. At this time, the switch 3 switches to the transmission signal generator 2 side and sends a second control signal (dotted line) to the delay time estimation device 51 ₁ , the M-stage adaptive filter 52 ₁ , and the memory 6.

In step 151, the delay time estimation processing of the complex tap selection adaptive filter 5 is not performed, but the estimated delay time (^) d _1-pre obtained by transmitting a broadband signal in advance is called from the memory 6 and directly obtained by preprocessing. This value is assumed to be a delay time estimated value (^) d ₁ = (^) d _1-pre . In this delay time _{(^) d 1 = (^} ) d 1-pre, adaptive filter 52 ₁ is operated by using an adaptive algorithm to calculate the load _{g 1,} load estimation value _{g 1} after calculation ( ^) and _{g 1.} The type of adaptive algorithm is not particularly limited, such as a normalized LMS algorithm.

Next, in step 152, the error signal e ₁ (t) is obtained using the delay time (^) d ₁ = (^) d _1-pre used in step 151 and the obtained load g ₁ . This process can be expressed by the following equation.

  Here, H indicates complex conjugate transpose.

  When these processes are completed, the delay time in which the impulse response of the unknown system is localized and the magnitude thereof can be estimated (step 153).

  The unknown system is simulated from the estimated delay time of the impulse response of the unknown system and its magnitude, and the unknown system output signal is subtracted from the received signal to suppress the interference wave (step 154).

That is, when the switch 3 is switched to the wideband signal generator 1 side, the switch 3 sends a first control signal to the delay time estimation device 51 ₁ , the M-stage adaptive filter 52 ₁ , and the memory 6. Based on this first control signal, the adaptive filter 52 _1, a load g from the memory 6, and set by calling the g ^*, the delay time estimating unit 51 _1, the load g, g ^* of each error signal by using A mean square value is obtained, a delay time evaluation function is obtained from the two mean square values, a delay time is estimated from the delay time evaluation function, and the estimated delay time is stored in the memory 6. These pre-processing are steps 101-104.

When the interference wave is actually suppressed, the switch 3 is switched to the transmission signal generator 2 side, and the second control signal is sent to the delay time estimation device 51 ₁ , the M-stage adaptive filter 52 ₁ , and the memory 6. Send. Based on this second control signal, the delay time estimating unit 51 ₁ functions as a delay circuit by setting calls the estimated delay time from the memory 6, the adaptive filter 52 _1, a load g using an adaptive algorithm ₁ determined, subtractor 9 ₁ obtains an error signal from the estimated delay time and the load g _1, which interference wave is suppressed by. These interference wave suppression processes are steps 151 to 154.

Embodiment 2. FIG.
An interference wave removing apparatus according to Embodiment 2 of the present invention will be described with reference to FIG. FIG. 4 is a flowchart showing the operation of the interference wave canceller according to Embodiment 2 of the present invention. The configuration of the interference wave canceller according to Embodiment 2 of the present invention is the same as that of Embodiment 1 above.

  Here, for the sake of simplicity, the description will be made based on a situation where the plant noise is sufficiently small. In the second embodiment, when the number of sub-adaptive filter groups is K ≧ 2, a wideband signal is transmitted in advance, 1,..., K-stage delay time estimation values and respective sub-adaptive filter weights are obtained. Of these, the estimated delay time values of 1,..., K stages are used for actual interference wave suppression.

Consider a case where a broadband signal (white noise or the like) is transmitted from the broadband signal generator 1 in advance. At this time, the switch 3 switches to the wideband signal generator 1 side. Further, the switch 3 performs the first control so that the delay time estimation device 51 _k , the M-tap adaptive filter 52 _k , and the memory 6 can distinguish that the delay time is estimated using the wideband signal. Send a signal (dotted line). The operation of the complex tap selection adaptive filter 5 is shown in FIG. Here, the “processing to be performed in advance with a broadband signal” in FIG. 4 is performed.

In step 201, 202 and 203, when transmitting broadband signals, the autocorrelation function _{r xx} of the transmitted signal in the formula (2) is the peak of the correlation value appears only in d = _{d k.} In this case, the delay time evaluation function Φ ′ (d) has only one peak.

Next, in step 204, the search parameter d _i having the only peak is set as a delay time estimated value (^) d _k-pre . This estimated delay time value (^) d _k-pre is stored in the memory 6.

Next, in step 205, operates the adaptive filter 52 _k of small number of tap stages in the delay time estimation value obtained in step 204, using the adaptive algorithm, and calculates the load g _k-pre, after calculation g load estimation value of the _{k-pre (^)} and _{g k-pre.}

Next, in step 206, an error signal e _k-pre (t) is _obtained from the delay time estimated value (^) d _k-pre and the load (^) g _k-pre .

  In step 207, if the minimum value of the error signal obtained in step 206 is smaller than a preset threshold value, this repetitive processing is stopped. In other cases, k = k + 1 is set, and steps 201 to 206 are repeated until k = K. Thus, the preprocessing ends.

When the interference wave is actually suppressed, the “process when suppressing interference wave” in FIG. 4 is performed. At this time, the switch 3 switches to the transmission signal generator 2 side and sends the second control signal (dotted line) to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6.

In step 251, the delay time estimation process of the complex tap selection adaptive filter 5 is not performed, and the estimated delay time (^) d _k-pre obtained by transmitting a broadband signal in advance is called from the memory 6 and obtained by preprocessing. Let (^) d _k-pre be the k-th stage delay time estimate (^) d _k = (^) d _k-pre . In this delay time _{(^) d k = (^} ) d k-pre, operates the adaptive filter 52 _k taps M stage, and the load value _{g k} after calculation load estimation value (^) _{g k,} this The load value is assumed to be an impulse response estimation value of an unknown system. It doesn't matter about the type of adaptive algorithm.

Next, in step 252, the error signal e _k (t) is obtained using the delay time (^) d _k = (^) d _k-pre used in step 251 and the obtained load (^) g _k. . This process can be expressed by the following equation.

However, when k = 1, e _k−1 (t) is the output f (t) of the unknown system.

  In step 253, if the minimum value of the error signal obtained in step 252 is smaller than a preset threshold value, the iterative process is stopped. In other cases, k = k + 1 is set, and the processes of steps 251 and 252 are repeated until k = K.

  When these processes are completed, it is possible to estimate the delay time in which the impulse responses of all unknown systems are localized and the magnitude thereof.

  The unknown system is simulated from the estimated delay times and the magnitudes of the impulse responses of all unknown systems, and interference waves are removed (step 254).

That is, when the switch 3 is switched to the wideband signal generator 1 side, the switch 3 sends a first control signal to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Based on the first control signal, the adaptive filter 52 _k calls and sets the loads g and g ^* from the memory 6, and the delay time estimation device 51 _k uses the loads g and g ^* , respectively. A mean square value is obtained, a delay time evaluation function is obtained from the two mean square values, a delay time is estimated from the delay time evaluation function, and the estimated delay time is stored in the memory 6. Further, the delay time estimation device 51 _k estimates a load at the estimated delay time using an adaptive algorithm, and obtains an error signal from the estimated delay time and the estimated load. Until the minimum value of the error signal reaches the threshold value, the process from obtaining the first mean square value to the process obtaining the last error signal is repeated. These pre-processing are steps 201-207.

When the interference wave is actually suppressed, the switch 3 is switched to the transmission signal generator 2 side, and the second control signal is sent to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Send. Based on the second control signal, the delay time estimation device 51 _k functions as a delay circuit by calling and setting the estimated delay time from the memory 6, and the adaptive filter 52 _k uses a weight ( ^) G _k is obtained, and the subtractor 9 _k obtains an error signal from the estimated delay time and the estimated weight (^) g _k . The process from obtaining the first load to the process obtaining the last error signal is repeated until the minimum value of the error signal reaches the threshold value. Thereby, the interference wave is suppressed. These interference wave suppression processes are steps 251 to 254.

Embodiment 3 FIG.
An interference wave removing apparatus according to Embodiment 3 of the present invention will be described with reference to FIG. FIG. 5 is a flowchart showing the operation of the interference wave canceller according to Embodiment 3 of the present invention. Note that the configuration of the interference wave canceller according to Embodiment 3 of the present invention is the same as that of Embodiment 1 above.

  Here, for the sake of simplicity, the description will be made based on a situation where the plant noise is sufficiently small. In the third embodiment, when the number of sub-adaptive filter groups is K ≧ 1, a wideband signal is transmitted in advance, and a delay time estimated value of 1,..., K stages and a weight of each sub-adaptive filter are obtained. These are used during actual interference wave processing.

Consider a case where a broadband signal (white noise or the like) is transmitted from the broadband signal generator 1 in advance. At this time, the switch 3 switches to the wideband signal generator 1 side. Further, the switch 3 performs the first control so that the delay time estimation device 51 _k , the M-tap adaptive filter 52 _k , and the memory 6 can distinguish that the delay time is estimated using the wideband signal. Send a signal (dotted line). The operation of the complex tap selection adaptive filter 5 is shown in FIG. Here, the “processing performed in advance with a broadband signal” in FIG. 5 is performed.

In Steps 301, 302, and 303, when a broadband signal is transmitted, a correlation value peak appears only in d _k = d _{i in the} autocorrelation function r _xx of the transmission signal in Equation (2). In this case, the delay time evaluation function Φ ′ (d) has only one peak.

Next, in step 304, the search parameter d _i having the only peak is set as a delay time estimated value (^) d _k-pre . This estimated delay time value (^) d _k-pre is stored in the memory 6.

Next, in step 305, operates the adaptive filter 52 _k of small number of tap stages in the delay time estimation value obtained in step 304, using the adaptive algorithm, and calculates the load g _k-pre, after calculation g _k-pre is assumed to be a load estimated value (^) g _k-pre and is stored in the memory 6.

Next, in step 306, an error signal e _k-pre (t) is _obtained from the delay time (^) d _k-pre and the load (^) g _k-pre .

  In step 307, if the minimum value of the error signal obtained in step 306 is smaller than a preset threshold value, the iterative process is stopped. In other cases, k = k + 1 is set, and steps 301 to 306 are repeated until k = K. Thus, the preprocessing ends.

When the interference wave is actually suppressed, the “processing when suppressing the interference wave” in FIG. 5 is performed. At this time, the switch 3 switches to the transmission signal generator 2 side and sends the second control signal (dotted line) to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6.

In step 351, the delay time estimation process of the complex tap selection adaptive filter 5 is not performed, and the estimated delay time (^) d _k-pre and the weight (^) g _k-pre obtained by transmitting a broadband signal in advance are stored in the memory. 6 and (^) d _k-pre obtained in the pre-processing is set as the k-th stage delay time estimated value (^) d _k = (^) d _k-pre . In this delay time (^) _dk = (^) _dk-pre , the adaptive filter _52k with M taps is operated, the load _gk is calculated, and the calculated load value is (^) _gk. And an unknown impulse response estimate. Here, the estimated load (^) g _k-pre is used as the initial value of the load update. Thereby, the convergence time of the adaptive algorithm can be shortened. It doesn't matter about the type of adaptive algorithm.

Next, in step 352, the error signal e _k (t) is obtained using the delay time (^) d _k = (^) d _k-pre used in step 351 and the obtained load (^) g _k. . This process can be expressed by the following equation.

However, when k = 1, e _k−1 (t) is the output f (t) of the unknown system.

  In step 353, if the minimum value of the error signal obtained in step 352 is smaller than a preset threshold value, the iterative process is stopped. In other cases, k = k + 1 is set, and steps 351 and 352 are repeated until k = K.

  When these processes are completed, it is possible to estimate the delay time and the size of the impulse responses of all unknown systems.

  The unknown system is simulated from the estimated delay times and the magnitudes of the impulse responses of all unknown systems, and interference waves are removed (step 354).

That is, when the switch 3 is switched to the wideband signal generator 1 side, the switch 3 sends a first control signal to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Based on the first control signal, the adaptive filter 52 _k calls and sets the loads g and g ^* from the memory 6, and the delay time estimation device 51 _k uses the loads g and g ^* , respectively. A mean square value is obtained, a delay time evaluation function is obtained from the two mean square values, a delay time is estimated from the delay time evaluation function, and the estimated delay time is stored in the memory 6. Further, the delay time estimation device 51 _k estimates the load at the estimated delay time using an adaptive algorithm and stores the estimated load in the memory 6. Next, an error signal is obtained from the estimated delay time and estimated load. Until the minimum value of the error signal reaches the threshold value, the process from obtaining the first mean square value to the process obtaining the last error signal is repeated. These preprocessing are steps 301 to 307.

When the interference wave is actually suppressed, the switch 3 is switched to the transmission signal generator 2 side, and the second control signal is sent to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Send. Based on the second control signal, the delay time estimation device 51 _k functions as a delay circuit by calling and setting the estimated delay time from the memory 6, and the adaptive filter 52 _k calls the estimated weight from the memory 6. As an initial value, the load g _k is estimated using an adaptive algorithm, and the subtractor 9 _k obtains an error signal from the estimated delay time and the estimated load (^) g _k . The process from obtaining the first load to the process obtaining the last error signal is repeated until the minimum value of the error signal reaches the threshold value. Thereby, the interference wave is suppressed. These interference wave suppression processes are steps 351 to 354.

Embodiment 4 FIG.
An interference wave removing apparatus according to Embodiment 4 of the present invention will be described with reference to FIG. FIG. 6 is a flowchart showing the operation of the interference wave canceller according to Embodiment 4 of the present invention. Note that the configuration of the interference wave canceller according to Embodiment 4 of the present invention is the same as that of Embodiment 1 above.

  Here, for the sake of simplicity, the description will be made based on a situation where the plant noise is sufficiently small. In the third embodiment, a wideband signal is processed in advance to obtain 1,..., K stages of delay time estimates, and the load of the 1,. Although stored in the memory 6 and used as an initial value at the time of calculating the k-th stage load when suppressing the interference wave, in the fourth embodiment, the load value obtained in advance during the processing of the broadband signal is used to suppress the interference. Sometimes used as a load value.

Consider a case where a broadband signal (white noise or the like) is transmitted from the broadband signal generator 1 in advance. At this time, the switch 3 switches to the wideband signal generator 1 side. In addition, the switch 3 uses the first control signal ( _k ) so that the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6 can distinguish that the delay time is estimated using the wideband signal. Send a dotted line). Here, the “processing performed in advance with a broadband signal” in FIG. 6 is performed.

In step 401, when transmitting broadband signals, the autocorrelation function _{r xx} of the transmitted signal in the formula (2) is the peak of the correlation value appears only in d = _{d k.} In this case, the delay time evaluation function Φ ′ (d) has only one peak.

Next, in step 404, the search parameter d _i having the only peak is set as a delay time estimated value (^) d _k-pre . This estimated delay time value (^) d _k-pre is stored in the memory 6.

Next, in step 405, operates the adaptive filter 52 _k of small number of tap stages in the delay time estimation value obtained in step 404, using the adaptive algorithm, and calculates the load g _k-pre, after calculation g _k-pre is assumed to be a load estimated value (^) g _k-pre and is stored in the memory 6.

Next, in step 406, an error signal e _k-pre (t) is _obtained from the delay time (^) d _k-pre and the load (^) g _k-pre .

  In step 407, if the minimum value of the error signal obtained in step 406 is smaller than a preset threshold value, the iterative process is stopped. In other cases, k = k + 1 is set, and steps 401 to 406 are repeated until k = K. Thus, the preprocessing ends.

When the interference wave is actually suppressed, the “processing when suppressing the interference wave” in FIG. 6 is performed. At this time, the switch 3 switches to the transmission signal generator 2 side and sends the second control signal (dotted line) to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6.

In step 451, the processing of the delay time estimation of the complex tap selection adaptive filter 5, without load calculation of the adaptive filter 52 _k taps M stage, estimated delay time obtained by sending a pre-wideband signal (^) d _k-pre and the estimated load (^) g _k-pre are called from the memory 6, and the error signal e _k (by using the delay time estimated value (^) d _k-pre and the load value (^) g _k-pre t). This eliminates the need for adaptive algorithm operations. This process can be expressed by the following equation.

However, when k = 1, e _k−1 (t) is the output f (t) of the unknown system.

  When these processes are completed, it is possible to estimate the delay time and the size of the impulse responses of all unknown systems.

  The unknown system is simulated from the estimated delay times and the magnitudes of the impulse responses of all unknown systems, and interference waves are removed (step 452).

That is, when the switch 3 is switched to the wideband signal generator 1 side, the switch 3 sends a first control signal to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Based on the first control signal, the adaptive filter 52 _k calls and sets the loads g and g ^* from the memory 6, and the delay time estimation device 51 _k uses the loads g and g ^* , respectively. A mean square value is obtained, a delay time evaluation function is obtained from the two mean square values, a delay time is estimated from the delay time evaluation function, and the estimated delay time is stored in the memory 6. Further, the delay time estimation device 51 _k estimates the load at the estimated delay time using an adaptive algorithm and stores the estimated load in the memory 6. Next, an error signal is obtained from the estimated delay time and estimated load. Until the minimum value of the error signal reaches the threshold value, the process from obtaining the first mean square value to the process obtaining the last error signal is repeated. These pre-processing are steps 401-407.

When the interference wave is actually suppressed, the switch 3 is switched to the transmission signal generator 2 side, and the second control signal is sent to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Send. Based on the second control signal, the delay time estimation device 51 _k functions as a delay circuit by calling and setting the estimated delay time from the memory 6, and the adaptive filter 52 _k calls the estimated weight from the memory 6. To set. The subtractor 9 _k obtains an error signal from the estimated delay time and the estimated load (^) g _k . Thereby, the interference wave is suppressed. These interference wave suppression processes are steps 451 to 452.

Embodiment 5 FIG.
An interference wave removing apparatus according to Embodiment 5 of the present invention will be described with reference to FIG. FIG. 7 is a flowchart showing the operation of the interference wave canceller according to Embodiment 5 of the present invention. Note that the configuration of the interference wave canceller according to Embodiment 5 of the present invention is the same as that of Embodiment 1 above.

  Here, for the sake of simplicity, the description will be made based on a situation where the plant noise is sufficiently small. In the fifth embodiment, when the number of sub-adaptive filter groups is K = 1, the delay time is estimated in advance with a wideband signal, and the load is obtained when each interference wave is incident.

Consider a case where a broadband signal (white noise or the like) is transmitted from the broadband signal generator 1 in advance. At this time, the switch 3 switches to the wideband signal generator 1 side. Further, the switch 3 uses the first control signal (1) so that the delay time estimation device 51 ₁ , the M-stage adaptive filter 52 ₁ , and the memory 6 can distinguish that the delay time is estimated using the wideband signal. Send a dotted line). Here, the “processing performed in advance with a broadband signal” in FIG. 7 is performed.

In step 501, when a broadband signal is transmitted, a cross-correlation function r _xf between the broadband signal x (t) and the received signal f (t) through which the broadband signal has passed through the unknown system is calculated. This can be expressed as: d is a delay time search parameter and is given so as to satisfy the time axis of the impulse response of the unknown system.

  Here, <> represents an expected value calculation, and * represents a complex conjugate.

Next, in step 502, the peak of the correlation value appears only in d ₁ = d _i in equation (10). Let the search parameter d _i giving this unique peak be the delay time estimate (^) d _{1 -pre} . This estimated delay time value (^) d _1-pre is stored in the memory 6. Thus, the preprocessing ends.

  Hereinafter, the processing in steps 551 to 554 is the same as that in steps 151 to 154 in the first embodiment, and a description thereof will be omitted.

That is, when the switch 3 is switched to the wideband signal generator 1 side, the switch 3 sends a first control signal to the delay time estimation device 51 ₁ , the M-stage adaptive filter 52 ₁ , and the memory 6. Based on this first control signal, the delay time estimating unit 51 ₁ calculates the wideband signal x (t), the cross-correlation function of the broadband signal is a received signal f that has passed through the unknown system (t), The delay time that maximizes the cross-correlation value is stored in the memory 6 as the estimated delay time. These pre-processing are steps 501 to 502.

When the interference wave is actually suppressed, the switch 3 is switched to the transmission signal generator 2 side, and the second control signal is sent to the delay time estimation device 51 ₁ , the M-stage adaptive filter 52 ₁ , and the memory 6. Send. Based on this second control signal, the delay time estimating unit 51 ₁ functions as a delay circuit by setting calls the estimated delay time from the memory 6, the adaptive filter 52 _1, a load g using an adaptive algorithm ₁ determined, subtractor 9 ₁ obtains an error signal from the estimated delay time and the load g _1, which interference wave is suppressed by. These interference wave suppression processes are steps 551 to 554.

Embodiment 6 FIG.
An interference wave canceller according to Embodiment 6 of the present invention will be described with reference to FIG. FIG. 8 is a flowchart showing the operation of the interference wave canceller according to Embodiment 6 of the present invention. Note that the configuration of the interference wave canceller according to Embodiment 6 of the present invention is the same as that of Embodiment 1 above.

  Here, for the sake of simplicity, the description will be made based on a situation where the plant noise is sufficiently small. In the sixth embodiment, when the number of sub-adaptive filter groups is K ≧ 2, wideband signals are processed in advance, 1,..., K-stage delay time estimation values and respective sub-adaptive filter weights are obtained, Of these, the estimated delay time values of 1,..., K stages are used during actual interference wave processing.

Consider a case where a broadband signal (white noise or the like) is transmitted from the broadband signal generator 1 in advance. At this time, the switch 3 switches to the wideband signal generator 1 side. In addition, the switch 3 uses the first control signal ( _k ) so that the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6 can distinguish that the delay time is estimated using the wideband signal. Send a dotted line). In this pre-processing, “processing performed in advance with a broadband signal” in FIG. 8 is performed.

In step 601, a wideband signal is transmitted, and a cross-correlation function r _xf between the wideband signal and a received signal that has passed through the unknown system is calculated.

Next, in step 602, the cross-correlation function r _xf has a correlation value peak only at d = d _k . The search parameter d _k having the only peak is assumed to be a delay time estimated value (^) d _k-pre . This estimated delay time value (^) d _k-pre is stored in the memory 6.

Next, in step 603, the adaptive filter 52 _k having a small number of taps is operated with the delay time estimated value (^) d _k-pre obtained in step 602, and the weight g _k-pre is calculated using an adaptive algorithm. The calculated g _k-pre is defined as a load estimated value (^) g _k-pre .

Next, in step 604, the error signal e _k-pre (t) is calculated by the following equation using the delay time estimated value (^) d _k-pre and the load (^) g _k-pre .

However, when k = 1, e _k−1 (t) is the output f (t) of the unknown system.

  In step 605, if the minimum value of the error signal obtained in step 604 is smaller than a preset threshold value, the iterative process is stopped. In other cases, k = k + 1 is set, and steps 601 to 604 are repeated until k = K. Thus, the preprocessing ends.

  Hereinafter, the processing in steps 651 to 654 is the same as that in steps 251 to 254 in the second embodiment, and a description thereof will be omitted.

That is, when the switch 3 is switched to the wideband signal generator 1 side, the switch 3 sends a first control signal to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Based on the first control signal, the delay time estimation device 51 _k calculates a cross-correlation function between the wideband signal x (t) and the received signal f (t) through which the wideband signal has passed through the unknown system, The delay time that maximizes the cross-correlation value is stored in the memory 6 as the estimated delay time. Further, the delay time estimation device 51 _k estimates a load at the estimated delay time using an adaptive algorithm, and obtains an error signal from the estimated delay time and the estimated load. The process from obtaining the first cross-correlation function to the process obtaining the last error signal is repeated until the minimum value of the error signal reaches the threshold value. These pre-processing are steps 601 to 605.

When the interference wave is actually suppressed, the switch 3 is switched to the transmission signal generator 2 side, and the second control signal is sent to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Send. Based on the second control signal, the delay time estimation device 51 _k functions as a delay circuit by calling and setting the estimated delay time from the memory 6, and the adaptive filter 52 _k uses the adaptive algorithm to load g _k is obtained, and the subtractor 9 _k obtains an error signal from the estimated delay time and the estimated weight (^) g _k . The process from obtaining the first load to the process obtaining the last error signal is repeated until the minimum value of the error signal reaches the threshold value. Thereby, the interference wave is suppressed. These interference wave suppression processes are steps 651 to 654.

Embodiment 7 FIG.
An interference wave canceling apparatus according to Embodiment 7 of the present invention will be described with reference to FIG. FIG. 9 is a flowchart showing the operation of the interference wave canceller according to Embodiment 7 of the present invention. Note that the configuration of the interference wave canceller according to Embodiment 7 of the present invention is the same as that of Embodiment 1 above.

  Here, for the sake of simplicity, the description will be made based on a situation where the plant noise is sufficiently small. In the seventh embodiment, when the number of sub-adaptive filter groups is K ≧ 1, wideband signals are processed in advance, 1,..., K stages of delay time estimation values and respective sub-adaptive filter weights are obtained, Is used when actual interference waves are suppressed.

Consider a case where a broadband signal (white noise or the like) is transmitted from the broadband signal generator 1 in advance. At this time, the switch 3 switches to the wideband signal generator 1 side. In addition, the switch 3 uses the first control signal ( _k ) so that the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6 can distinguish that the delay time is estimated using the wideband signal. Send a dotted line). In the pre-processing, “processing performed in advance with a broadband signal” in FIG. 9 is performed.

In step 701, a wideband signal is transmitted, and a cross-correlation function r _xf of the wideband signal and a received signal that has passed through the unknown system is calculated.

Next, in step 702, the cross-correlation function r _xf has a correlation value peak only at d = d _k . The search parameter d having the only peak is assumed to be a delay time estimated value (^) d _k-pre . This estimated delay time value (^) d _k-pre is stored in the memory 6.

Next, in step 703, an adaptive filter with a small number of taps is operated with the delay time estimated value (^) d _k-pre obtained in step 702, and the load g _k-pre is calculated using an adaptive algorithm. Then, g _k-pre after the calculation is set as a load estimated value (^) g _k-pre and stored in the memory 6.

Next, in step 704, the error signal e _k-pre (t) is calculated by the following equation using the delay time estimated value (^) d _k-pre and the load (^) g _k-pre .

However, when k = 1, _ek-1-pre (t) is an unknown output f (t).

  In step 705, if the minimum value of the error signal obtained in step 704 is smaller than a preset threshold value, the iterative process is stopped. In other cases, k = k + 1 is set, and steps 701 to 704 are repeated until k = K. Thus, the preprocessing ends.

  Hereinafter, the processing in steps 751 to 754 is the same as that in steps 351 to 354 in the third embodiment, and a description thereof will be omitted.

That is, when the switch 3 is switched to the wideband signal generator 1 side, the switch 3 sends a first control signal to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Based on the first control signal, the delay time estimation device 51 _k calculates a cross-correlation function between the wideband signal x (t) and the received signal f (t) through which the wideband signal has passed through the unknown system, The delay time that maximizes the cross-correlation value is stored in the memory 6 as the estimated delay time. Further, the delay time estimation device 51 _k estimates the load at the estimated delay time using an adaptive algorithm and stores the estimated load in the memory 6. Next, an error signal is obtained from the estimated delay time and estimated load. The process from obtaining the first cross-correlation function to the process obtaining the last error signal is repeated until the minimum value of the error signal reaches the threshold value. These preprocessing are steps 701 to 705.

When the interference wave is actually suppressed, the switch 3 is switched to the transmission signal generator 2 side, and the second control signal is sent to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Send. Based on the second control signal, the delay time estimation device 51 _k functions as a delay circuit by calling and setting the estimated delay time from the memory 6, and the adaptive filter 52 _k calls the estimated weight from the memory 6. As an initial value, the load g _k is estimated using an adaptive algorithm, and the subtractor 9 _k obtains an error signal from the estimated delay time and the estimated load (^) g _k . The process from obtaining the first load to the process obtaining the last error signal is repeated until the minimum value of the error signal reaches the threshold value. Thereby, the interference wave is suppressed. These interference wave suppression processes are steps 751 to 754.

Embodiment 8 FIG.
An interference wave canceller according to Embodiment 8 of the present invention will be described with reference to FIG. FIG. 10 is a flowchart showing the operation of the interference wave canceller according to Embodiment 8 of the present invention. Note that the configuration of the interference wave canceller according to Embodiment 8 of the present invention is the same as that of Embodiment 1 above.

  Here, for the sake of simplicity, the description will be made based on a situation where the plant noise is sufficiently small. In the seventh embodiment, a wideband signal is processed in advance to obtain 1,..., K-stage delay time estimates, and the 1,. Although stored in the memory 6 and used as an initial value at the time of calculating the k-th stage load when suppressing the interference wave, in the eighth embodiment, the load value obtained in advance during the processing of the wideband signal is used for interference suppression. Sometimes used as a load value.

Consider a case where a broadband signal (white noise or the like) is transmitted from the broadband signal generator 1 in advance. At this time, the switch 3 switches to the wideband signal generator 1 side. In addition, the switch 3 uses the first control signal ( _k ) so that the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6 can distinguish that the delay time is estimated using the wideband signal. Send a dotted line). In the pre-processing, “processing performed in advance with a broadband signal” in FIG. 10 is performed.

In step 801, a wideband signal is transmitted, and a cross-correlation function r _xf between the wideband signal and a received signal that has passed through the unknown system is calculated.

Next, in step 802, the cross-correlation function r _xf has a correlation value peak only at d = d _k . The search parameter d having the only peak is assumed to be a delay time estimated value (^) d _k-pre . This estimated delay time value (^) d _k-pre is stored in the memory 6.

Next, in step 803, the adaptive filter 52 _k having a small number of tap stages is operated with the delay time estimation value (^) d _k-pre obtained in step 802, and the weight g _k-pre is calculated using an adaptive algorithm. The calculated g _k-pre is set as a load estimated value (^) g _k-pre and stored in the memory 6.

Next, in step 804, the error signal e _k-pre (t) is calculated by the following equation using the delay time estimated value (^) d _k-pre and the load (^) g _k-pre .

However, when k = 1, _ek-1-pre (t) is an unknown output f (t).

  In step 805, if the minimum value of the error signal obtained in step 804 is smaller than a preset threshold value, the iterative process is stopped. In other cases, k = k + 1 is set, and steps 801 to 804 are repeated until k = K. Thus, the preprocessing ends.

  Hereinafter, the processing in steps 851 to 852 is the same as that in steps 451 to 452 of the above-described fourth embodiment, and thus description thereof is omitted.

That is, when the switch 3 is switched to the wideband signal generator 1 side, the switch 3 sends a first control signal to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Based on the first control signal, the delay time estimation device 51 _k calculates a cross-correlation function between the wideband signal x (t) and the received signal f (t) through which the wideband signal has passed through the unknown system, The delay time that maximizes the cross-correlation value is stored in the memory 6 as the estimated delay time. Further, the delay time estimation device 51 _k estimates the load at the estimated delay time using an adaptive algorithm and stores the estimated load in the memory 6. Next, an error signal is obtained from the estimated delay time and estimated load. The process from obtaining the first cross-correlation function to the process obtaining the last error signal is repeated until the minimum value of the error signal reaches the threshold value. These pre-processing are steps 801 to 805.

When the interference wave is actually suppressed, the switch 3 is switched to the transmission signal generator 2 side, and the second control signal is sent to the delay time estimation device 51 _k , the M-stage adaptive filter 52 _k , and the memory 6. Send. Based on the second control signal, the delay time estimation device 51 _k functions as a delay circuit by calling and setting the estimated delay time from the memory 6, and the adaptive filter 52 _k calls the estimated weight from the memory 6. To set. The subtractor 9 _k obtains an error signal from the estimated delay time and the estimated load (^) g _k . Thereby, the interference wave is suppressed. These interference wave suppression processes are steps 851 to 852.

It is a figure which shows the structure of the interference wave removal apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the structure of the complex tap selection adaptive filter of the interference wave removal apparatus which concerns on Embodiment 1 of this invention. It is a flowchart which shows operation | movement of the interference wave removal apparatus which concerns on Embodiment 1 of this invention. It is a flowchart which shows operation | movement of the interference wave removal apparatus which concerns on Embodiment 2 of this invention. It is a flowchart which shows operation | movement of the interference wave removal apparatus which concerns on Embodiment 3 of this invention. It is a flowchart which shows operation | movement of the interference wave removal apparatus which concerns on Embodiment 4 of this invention. It is a flowchart which shows operation | movement of the interference wave removal apparatus which concerns on Embodiment 5 of this invention. It is a flowchart which shows operation | movement of the interference wave removal apparatus which concerns on Embodiment 6 of this invention. It is a flowchart which shows operation | movement of the interference wave removal apparatus which concerns on Embodiment 7 of this invention. It is a flowchart which shows operation | movement of the interference wave removal apparatus which concerns on Embodiment 8 of this invention. It is a conceptual diagram which shows the fundamental structure of the conventional delay time estimation apparatus. It is a flowchart which shows operation | movement of the conventional tap selection adaptive filter. It is a conceptual diagram which shows the method of delay time estimation in a wideband signal. It is a conceptual diagram which shows the method of delay time estimation in the narrow-band signal with periodicity.

Explanation of symbols

1-wideband signal generator, second transmission signal generator, 3 switches, 4 transmit antennas, 5 complex tap selection adaptive filter, 6 memory, 7 receiving antenna, 8 adders, _{9, 9} 1, _{9 k} subtractor ₅₁ 1, 51 _k delay time estimation device, 52 ₁ , 52 _k adaptive filter.

Claims (12)

A broadband signal generator for generating a broadband signal;
A transmission signal generator for generating a transmission signal;
When switched to the broadband signal generator side to transmit the broadband signal to an unknown system, a first control signal is output, and switched to the transmission signal generator side to transmit the transmission signal to the unknown system A switch that outputs a second control signal when received,
A memory for storing the estimated delay time;
When the first control signal is input, a delay time is estimated based on a delay time evaluation function and the estimated delay time is stored in the memory. When the second control signal is input, the memory A complex tap selection adaptive filter that calls an estimated delay time, estimates a weight at the estimated delay time using an adaptive algorithm, and obtains an error signal from the estimated delay time and the estimated weight;
An interference wave canceling apparatus comprising: a subtractor that suppresses an interference wave by subtracting an output of the complex tap selection adaptive filter from a reception signal in which the transmission signal passes through the unknown system. A broadband signal generator for generating a broadband signal;
A transmission signal generator for generating a transmission signal;
When switched to the broadband signal generator side to transmit the broadband signal to an unknown system, a first control signal is output, and switched to the transmission signal generator side to transmit the transmission signal to the unknown system A switch that outputs a second control signal when received,
A memory for storing estimated delay times and weights;
When the first control signal is input, the delay time is estimated based on the delay time evaluation function, the estimated delay time is stored in the memory, and the load at the estimated delay time is estimated using an adaptive algorithm. When the estimated load is stored in the memory and the second control signal is input, the estimated delay time and the estimated load are called from the memory, and the estimated load is used as an initial value to determine the estimated delay time using the adaptive algorithm. A complex tap selection adaptive filter that obtains an error signal from an estimated delay time and an estimated weight at the time of input of the second control signal;
An interference wave canceling apparatus comprising: a subtractor that suppresses an interference wave by subtracting an output of the complex tap selection adaptive filter from a reception signal in which the transmission signal passes through the unknown system. A broadband signal generator for generating a broadband signal;
A transmission signal generator for generating a transmission signal;
When switched to the broadband signal generator side to transmit the broadband signal to an unknown system, a first control signal is output, and switched to the transmission signal generator side to transmit the transmission signal to the unknown system A switch that outputs a second control signal when received,
A memory for storing estimated delay times and weights;
When the first control signal is input, the delay time is estimated based on the delay time evaluation function, the estimated delay time is stored in the memory, and the load at the estimated delay time is estimated using an adaptive algorithm. A complex tap selection adaptive filter that stores an estimated weight in the memory and, when the second control signal is input, calls the estimated delay time and estimated weight from the memory and obtains an error signal from the estimated delay time and estimated weight. When,
An interference wave canceling apparatus comprising: a subtractor that suppresses an interference wave by subtracting an output of the complex tap selection adaptive filter from a reception signal in which the transmission signal passes through the unknown system. A broadband signal generator for generating a broadband signal;
A transmission signal generator for generating a transmission signal;
When switched to the broadband signal generator side to transmit the broadband signal to an unknown system, a first control signal is output, and switched to the transmission signal generator side to transmit the transmission signal to the unknown system A switch that outputs a second control signal when received,
A memory for storing the estimated delay time;
When the first control signal is input, a delay time is estimated based on a cross-correlation function between the broadband signal and the received signal through which the broadband signal has passed through the unknown system, and the estimated delay time is stored in the memory. In addition, when the second control signal is input, the estimated delay time is called from the memory, the load at the estimated delay time is estimated using an adaptive algorithm, and an error signal is obtained from the estimated delay time and the estimated load. A complex tap selection adaptive filter;
An interference wave canceling apparatus comprising: a subtractor that suppresses an interference wave by subtracting an output of the complex tap selection adaptive filter from a reception signal in which the transmission signal passes through the unknown system. A broadband signal generator for generating a broadband signal;
A transmission signal generator for generating a transmission signal;
When switched to the broadband signal generator side to transmit the broadband signal to an unknown system, a first control signal is output, and switched to the transmission signal generator side to transmit the transmission signal to the unknown system A switch that outputs a second control signal when received,
A memory for storing estimated delay times and weights;
When the first control signal is input, a delay time is estimated based on a cross-correlation function between the broadband signal and the received signal through which the broadband signal has passed through the unknown system, and the estimated delay time is stored in the memory. Then, the load at the estimated delay time is estimated using an adaptive algorithm and the estimated load is stored in the memory. When the second control signal is input, the estimated delay time and the estimated load are read from the memory. A complex tap selection adaptive filter that estimates a load at the estimated delay time using an adaptive algorithm with the estimated weight as an initial value and obtains an error signal from the estimated delay time and the estimated weight at the time of input of the second control signal When,
An interference wave canceling apparatus comprising: a subtractor that suppresses an interference wave by subtracting an output of the complex tap selection adaptive filter from a reception signal in which the transmission signal passes through the unknown system. A broadband signal generator for generating a broadband signal;
A transmission signal generator for generating a transmission signal;
When switched to the broadband signal generator side to transmit the broadband signal to an unknown system, a first control signal is output, and switched to the transmission signal generator side to transmit the transmission signal to the unknown system A switch that outputs a second control signal when received,
A memory for storing estimated delay times and weights;
When the first control signal is input, a delay time is estimated based on a cross-correlation function between the broadband signal and the received signal through which the broadband signal has passed through the unknown system, and the estimated delay time is stored in the memory. Then, the load at the estimated delay time is estimated using an adaptive algorithm and the estimated load is stored in the memory. When the second control signal is input, the estimated delay time and the estimated load are read from the memory. A complex tap selection adaptive filter that obtains an error signal from the call, estimated delay time and estimated weight;
An interference wave canceling apparatus comprising: a subtractor that suppresses an interference wave by subtracting an output of the complex tap selection adaptive filter from a reception signal in which the transmission signal passes through the unknown system. Transmitting a broadband signal to an unknown system, estimating a delay time based on a delay time evaluation function of the broadband signal, and storing the estimated delay time in a memory;
Transmitting a transmission signal to the unknown system, calling an estimated delay time from the memory, estimating a load at the estimated delay time using an adaptive algorithm, and obtaining an error signal from the estimated delay time and the estimated weight;
A method of suppressing interference waves by subtracting the error signal from a reception signal in which the transmission signal passes through the unknown system. A wideband signal is transmitted to an unknown system, a delay time is estimated based on a delay time evaluation function of the wideband signal, and the estimated delay time is stored in a memory, and a weight at the estimated delay time is estimated using an adaptive algorithm. Storing the estimated load in the memory;
A transmission signal is transmitted to the unknown system, an estimated delay time and an estimated weight are called from the memory, an estimated algorithm is used to estimate a load at the estimated delay time using an estimated algorithm, and the estimated delay time and the transmission Obtaining an error signal from an estimated load at the time of signal transmission;
A method of suppressing interference waves by subtracting the error signal from a reception signal in which the transmission signal passes through the unknown system. A wideband signal is transmitted to an unknown system, a delay time is estimated based on a delay time evaluation function of the wideband signal, and the estimated delay time is stored in a memory, and a weight at the estimated delay time is estimated using an adaptive algorithm. Storing the estimated load in the memory;
Transmitting a transmission signal to the unknown system, calling an estimated delay time and estimated load from the memory, and obtaining an error signal from the estimated delay time and estimated load;
A method of suppressing interference waves by subtracting the error signal from a reception signal in which the transmission signal passes through the unknown system. Transmitting a broadband signal to an unknown system, estimating a delay time based on a cross-correlation function between the broadband signal and a received signal through which the broadband signal has passed through the unknown system, and storing the estimated delay time in a memory;
Transmitting a transmission signal to the unknown system, calling an estimated delay time from the memory, estimating a load at the estimated delay time using an adaptive algorithm, and obtaining an error signal from the estimated delay time and the estimated weight;
A method of suppressing interference waves by subtracting the error signal from a reception signal in which the transmission signal passes through the unknown system. A wideband signal is transmitted to an unknown system, a delay time is estimated based on a cross-correlation function between the wideband signal and a received signal that has passed through the unknown system, and the estimated delay time is stored in a memory and adaptively Estimating a load at the estimated delay time using an algorithm and storing the estimated load in the memory;
A transmission signal is transmitted to the unknown system, an estimated delay time and an estimated weight are called from the memory, an estimated algorithm is used to estimate a load at the estimated delay time using an estimated algorithm, and the estimated delay time and the transmission Obtaining an error signal from an estimated load at the time of signal transmission;
A method of suppressing interference waves by subtracting the error signal from a reception signal in which the transmission signal passes through the unknown system. A wideband signal is transmitted to an unknown system, a delay time is estimated based on a cross-correlation function between the wideband signal and a received signal that has passed through the unknown system, and the estimated delay time is stored in a memory and adaptively Estimating a load at the estimated delay time using an algorithm and storing the estimated load in the memory;
Transmitting a transmission signal to the unknown system, calling an estimated delay time and estimated load from the memory, and obtaining an error signal from the estimated delay time and estimated load;
A method of suppressing interference waves by subtracting the error signal from a reception signal in which the transmission signal passes through the unknown system.

Images