JP2006067334A - Noise reduction device and receiver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a noise reduction device and a receiver which can properly reject a noise mixed in a signal. <P>SOLUTION: When a noise detection circuit 13 detects a pulse noise, a noise frequency detection circuit 14 detects a frequency of occurrence of the detected pulse noise. A control signal showing a frequency detection result switches between an ordinary processing circuit 16 and a selective processing circuit 17 in a noise reduction circuit 15. For example, the ordinary processing circuit 16 carries out a noise cut processing. The selective processing circuit 17 carries out a noise suppression processing such as a high cut processing or a mute processing different from a noise cut processing. Therefore, the noise reduction can be aimed by a proper processing in accordance with frequency of occurrence of noise. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a noise removing device that removes noise mixed in a signal to be processed, and a receiver that includes a noise removing device for a received signal.

  2. Description of the Related Art Conventionally, a receiver or the like that receives a radio signal may include noise in a received signal, and is provided with a noise removing device. Among noises, irregular pulse noise is difficult to remove without affecting the received signal. For example, a receiver mounted on an automobile generates ignition noise from an engine of the own vehicle, discharge noise received from a transmission line while traveling near a transmission line, or a factory device receiving while traveling near a factory. Noise and noise generated from the railway while traveling near the railway track are mixed in the received radio signal.

  The present applicant uses an AM broadcast receiver to detect pulse noise mixed in the received signal using a filter or the like, and then removes the section of the signal in which noise is detected by cutting it in time. The noise cancellation technique which performs the process to perform is disclosed (for example, refer patent documents 1-3). There is also known a noise canceling technique for removing a long pulse width or a short pulsed noise using an FM receiving unit that is vacant when receiving an AM broadcast in a radio receiver having an FM broadcast receiving function. (For example, see Patent Document 4).

  Note that when various digital signal processing is performed for reception of broadcast radio waves, the digital signal includes harmonic components, and thus the digital signal processing itself may become a noise generation source. There has also been proposed a technique for avoiding interference due to harmonics by combining filter processing or the like when a broadcast wave signal is received by a pulse count type demodulation circuit (see, for example, Patent Document 5).

Japanese Patent No. 2755497 Japanese Patent Publication No. 7-71014 JP 2000-91932 A JP-A-8-274663 JP-A-9-284163

  In the conventional noise cancellation technique, the pulsed noise is removed by temporally cutting a section of the signal including the detected pulsed noise. The detection of the pulse noise is performed on the upstream side of reception where the reception signal is not delayed, and the cut of the pulse noise is performed on the downstream side of reception where the reception signal is delayed. On the rear stage of reception, the pulse noise is also delayed due to the delay in the band selection filter processing for the signal. Therefore, if at least the rise is detected on the front stage, it can be cut off from the rise of the pulse noise on the rear stage. . Since the signal is delayed on the rear stage side and the time width of the pulse noise is increased, the pulse noise is cut over a time width longer than the time width of the pulse noise detected on the front stage side.

  However, if the repetition rate of the pulse noise increases and the pulse width decreases, there is a risk that the signal cut at the subsequent stage will not be in time and noise cancellation becomes invalid. In addition, even if the delay of the pulse noise is made in time with the addition of a delay circuit, etc., the pulse width of the pulse noise increases, so the proportion of time that the signal is cut increases and is included in the signal. Information may be lost. There is no omnipotently effective removal of noise mixed in the signal, and if the selection of the process is wrong, the removal of the noise is insufficient, and only the loss of information from the signal results. There is a possibility of becoming.

  The objective of this invention is providing the noise removal apparatus and receiver which can remove appropriately the noise mixed in a signal.

The present invention includes a noise detection means for detecting noise mixed in a signal,
Noise removal means that can select different noise removal processing depending on the nature of the noise for the signal;
The noise removing device includes a noise discriminating unit that discriminates a property of noise detected by the noise detecting unit and selects a noise removing process in the noise removing unit according to a discrimination result.

  In the present invention, the noise removing means can select at least a normal process selected as an initial state and a selection process selected when a predetermined noise characteristic is determined as the noise removal process. It is characterized by that.

  The present invention further includes selection changing means for controlling the change of the selection of the noise removal processing in the noise removal means based on the determination result of the noise determination means to be performed with a change over a predetermined time. And

In the present invention, the noise removal means can select at least a normal process selected as an initial state and a selection process selected when a predetermined noise characteristic is determined as the noise removal process. ,
The selection changing means is characterized in that the time required for the change from the selection process to the normal process is longer than the time required for the change from the normal process to the selection process.

Further, in the present invention, the noise removing means can remove the pulse noise by a plurality of different processes including the normal process and the selection process according to the repetition frequency,
The noise discrimination means is
Pulseity determination means for determining whether or not the noise detected by the noise detection means is pulsed noise;
A frequency detecting means for detecting a repetition frequency of noise determined as pulsed noise by the pulse determining means,
The selection process is selected when the repetition frequency detected by the frequency detection unit exceeds a preset reference as the nature of the noise.

  Also, in the present invention, the frequency detection means detects the frequency from the generation speed of the pulse noise.

  Further, in the present invention, the frequency detection means detects the frequency from the number of detections of the pulse noise within a certain section.

  Also, in the present invention, the frequency detection means detects the frequency from the pulse noise detection interval.

In the present invention, the frequency detection means may be
Addition / subtraction means for performing a predetermined addition process or subtraction process depending on whether or not the noise detection means detects pulse noise at predetermined time intervals with respect to a preset initial value,
The frequency is detected based on the processing result of the addition / subtraction means at a predetermined time.

In the present invention, the frequency detection means may be
Including noise extraction means for extracting only noise components from the signal;
The frequency is detected based on the signal level of the noise component extracted by the noise extraction means.

In the present invention, the frequency detection means may be
Further comprising direct current means for directing the noise component detected by the noise extraction means,
The DC level output from the DC converting means is the signal level.

  In the present invention, the noise removing unit may perform a noise cut process for temporally blocking only a noise detection section in the signal as the normal process.

  In the present invention, the noise removing unit may perform a high cut process for suppressing a high frequency component in the signal as the selection process.

  In the present invention, the noise removing unit may perform a mute process for suppressing the amplitude level of the signal as the selection process.

  In the present invention, the noise removing unit may change a cut-off frequency for performing the high-cut processing according to the intensity of the signal.

  In the invention, it is preferable that the noise detecting unit changes a degree of suppression of the high frequency component in the high cut processing according to a repetition frequency of noise detected by the frequency detecting unit.

  In the present invention, the noise removing unit may change the degree of suppression according to the intensity of the signal.

In the present invention, the signal transmission path is provided with variable means for changing the transmission amount of the signal,
The noise removing unit changes the degree of suppression according to the transmission amount of the variable unit.

  In the present invention, the noise removing unit may perform the selection process on a frequency band component corresponding to a repetition rate of noise detected by the noise detecting unit.

  In the present invention, the noise removing unit may perform the selection process together with the normal process by adding a weight corresponding to a repetition rate of noise detected by the noise detecting unit.

  The present invention is further characterized in that it further comprises signal filter means for applying a filter process for preventing false detection to a signal that is subject to noise detection by the noise detection means.

  In the present invention, the signal filter means performs an averaging filter process on the signal.

Furthermore, the present invention includes the noise removal device according to any one of the above,
A receiver characterized in that a received signal is the signal.

The present invention also includes a noise removing device that changes the degree of suppression according to the signal strength,
The receiver is characterized in that the received signal is the signal and the received electric field strength is the strength of the signal.

  According to the present invention, noise mixed in the signal is detected by the noise removing unit, the nature of the detected noise is determined by the noise determining unit, and the noise removing process in the noise removing unit is performed according to the determination result. Therefore, noise mixed in the signal can be appropriately removed.

  Further, according to the present invention, for example, the normal processing is determined corresponding to the nature of noise that is highly likely to be mixed in the signal, and the selection processing is determined corresponding to the nature of noise that is likely to be mixed next. Therefore, noise can be appropriately removed from a signal mixed with noise having different properties.

  Further, according to the present invention, the selection changing unit controls the selection of the noise removal processing in the noise removing unit so as to be performed with a change over a predetermined time. For example, a signal from which noise is removed is output as an acoustic signal. In this case, the auditory performance can be improved.

  Further, according to the present invention, since the time required for changing from the selection process to the normal process is longer than the time required for changing from the normal process to the selection process, it is necessary to switch to the selection process. Transition to a short time, quickly avoid the impact on the signal due to changes in the nature of the noise, and if switching to selection processing is no longer necessary, return to normal processing over time, and improve hearing performance Can do.

  Further, according to the present invention, the noise removing means can remove the pulse noise by a plurality of different processes including the normal process and the selection process according to the repetition frequency, and according to the repetition frequency with respect to the pulse noise. The removal process can be performed appropriately.

  Further, according to the present invention, since the frequency is detected from the generation speed of pulse noise, it is possible to appropriately select a process for noise removal according to the speed at which pulse noise is generated.

  Further, according to the present invention, since the frequency is detected from the number of detections of pulse noise within a certain interval, processing for noise removal is appropriately performed according to the number of detections of pulse noise detected within a certain interval. Can be selected.

  Further, according to the present invention, since the frequency is detected from the pulse noise detection interval, it is possible to appropriately select the process for noise removal according to the time interval at which the pulse noise is detected.

  According to the present invention, a predetermined addition process or subtraction process is performed on a predetermined initial value at predetermined time intervals depending on whether or not the noise detecting means detects pulse noise. Depending on the result of the addition / subtraction process, a process for noise removal can be appropriately selected.

  Further, according to the present invention, it is possible to extract only the noise component from the signal and appropriately select the process for noise removal according to the signal level of the noise component extracted from the signal.

  In addition, according to the present invention, the noise component is converted into a direct current to obtain a signal level for frequency detection, so that it is possible to appropriately select a process for noise removal according to the direct current level.

  In addition, according to the present invention, the noise cut processing for temporally blocking only the noise detection section in the signal is performed as the normal processing, and for example, single-shot or low-frequency pulse noise can be reliably removed. it can.

  Further, according to the present invention, the high cut processing for suppressing the high frequency component in the signal is performed as the selection processing, and in the case where a lot of pulse noise having a high frequency component is mixed in the signal, Also, the noise component can be relatively greatly suppressed. For example, when performing noise cut processing, the ratio of blocking the signal for noise removal increases according to the high frequency component, and information is lost from the signal. It is possible to suppress only the component and relatively increase the signal-to-noise ratio so as not to damage the information contained in the signal.

  Further, according to the present invention, the mute processing for suppressing the amplitude level of the signal is performed as the selection processing, the peak of pulse noise and the like is reduced, the signal-to-noise ratio is relatively increased, and the information contained in the signal is lost. You can avoid it.

  Further, according to the present invention, the cutoff frequency for performing the high cut processing is changed according to the signal strength. For example, if the signal strength is weak, the cutoff frequency is lowered to suppress many high frequency components of noise. Therefore, the signal-to-noise ratio can be relatively increased, and the information contained in the signal can be easily extracted. If the strength of the signal is strong, the cutoff frequency can be increased to reduce the influence on the information that the signal has.

  Further, according to the present invention, the degree of suppression of the high frequency component in the high cut processing is changed according to the noise repetition frequency. For example, if the frequency of noise repetition is large, the suppression of the high frequency component in the high cut processing is performed. The signal level can be increased to relatively increase the signal-to-noise ratio, thereby facilitating extraction of information contained in the signal. If the repetition frequency is small, the degree of suppression of the high frequency component can be lowered to reduce the influence on the information held by the signal.

  Further, according to the present invention, the degree of suppression in the mute process and the degree of suppression of the high frequency component in the high cut process are changed according to the signal intensity. For example, the degree of suppression is reduced when the signal intensity is reduced. To increase the signal-to-noise ratio relatively, making it easier to extract information contained in the signal. Further, if the signal strength is high, the degree of suppression can be reduced to reduce the influence on the information held by the signal.

  Further, according to the present invention, since the degree of suppression is changed according to the transmission amount of the variable means that can change the transmission amount of the signal, the degree of suppression can be appropriately adjusted according to the transmission amount.

  In addition, according to the present invention, since the selection processing is performed on the frequency band component corresponding to the noise repetition rate, the cutoff frequency in the high cut processing according to the noise repetition rate mixed at the time of noise detection. In addition, the degree of suppression in the mute process can be changed appropriately.

  Further, according to the present invention, the selection process is performed together with the normal process by adding a weight corresponding to the repetition rate of the noise detected by the noise detection means. Therefore, if the difference in noise characteristics increases, the weight is increased. Thus, the degree of selection processing can be increased to perform appropriate noise removal.

  Further, according to the present invention, since the filter processing for preventing erroneous detection is performed on the signal that is the target of noise detection, it is possible to appropriately perform noise detection.

  Further, according to the present invention, since the averaging filter process is performed on the signal, for example, noise accompanied by a plurality of vibrations can be detected as one noise, and erroneous detection can be avoided.

  Furthermore, according to the present invention, it is possible to provide a noise removing device that selects an appropriate process according to the nature of noise to be mixed in with respect to the received signal of the receiver.

  Further, according to the present invention, the degree of suppression in noise removal by the noise removal device can be changed according to the received electric field strength of the receiver indicated by, for example, an S meter signal.

  Hereinafter, a plurality of embodiments for carrying out the present invention will be described with reference to the drawings. In addition, the same reference mark is attached | subjected to the part corresponding to the matter demonstrated with the form preceded by each form, and the overlapping description may be abbreviate | omitted. In addition, when only a part of the configuration is described, the other parts of the configuration are the same as those described in advance. In addition to the combination of parts specifically described in each embodiment, the embodiments may be partially combined as long as the combination is not particularly troublesome.

FIG. 1 shows a schematic electrical configuration of a receiver 1 as an embodiment of the present invention. The receiver 1 includes a tuner 2, an intermediate frequency (hereinafter abbreviated as “IF”) amplifier circuit 3, an analog-digital (hereinafter abbreviated as “A / D”) conversion circuit 4, a detection circuit 5, and a low frequency. A circuit 6, a speaker 7, an S meter circuit 8, an automatic gain control (hereinafter abbreviated as “AGC”) circuit 9 and a noise removing device 10 are included. The tuner 2 amplifies a high frequency signal received by a radio signal such as a broadcast radio wave with an antenna, converts the frequency, and derives an IF signal having a constant frequency. For example, when the receiver 1 receives an AM broadcast, the tuner 2 receives and amplifies a high-frequency signal in the medium wave (MW) band of several hundreds kHz to several hundreds of kHz from an antenna, and converts it into an IF signal of about 450 kz. . The IF amplifier circuit 3 selectively amplifies the IF signal output from the tuner 2. In the receiver 1, the output of the IF amplifier circuit 3 is converted from an analog signal to a digital signal by the A / D conversion circuit 4, and the subsequent signal processing is performed by one or more DSPs (Digital Signals).
The digital signal processing is performed using a processor.

  The IF digital signal converted by the A / D conversion circuit 4 is input to the detection circuit 5 and subjected to detection processing for extracting an audio signal and the like. From the detection circuit 5, a DETOUT signal such as an audio signal is derived from a portion corresponding to the amplitude modulation component of the IF signal as a signal after detection processing. The DETOUT signal is amplified by the low frequency circuit 6, and the speaker 7 and the like are driven to obtain an acoustic output. Of the signal after detection processing by the detection circuit 5, a portion corresponding to the carrier component is input to, for example, the S meter circuit 8, and an S meter signal corresponding to the received electric field strength is generated. The S meter signal is input to, for example, the AGC circuit 9, and controls the gains of IF amplification and high frequency amplification in the IF amplifier circuit 3 and the tuner 2, respectively, and the fading phenomenon in which the radio wave propagation state changes and the received electric field strength changes. Is adjusted so that the volume change of the sound output from the speaker 7 is reduced. The noise removal device 10 removes or suppresses noise mixed in the DETOUT signal output from the detection circuit 5 and sends the noise to the low frequency circuit 6.

  FIG. 2 shows a schematic electrical configuration of the noise removing apparatus 10 of FIG. The noise removing device 10 includes a low-pass filter (hereinafter abbreviated as “LPF”) 11, abbreviated as a high-pass filter (hereinafter “HPF”) 12, a noise detection circuit 13, and a noise frequency. A detection circuit 14 and a noise removal circuit 15 are included. The noise removal circuit 15 includes a normal processing circuit 16, a selection processing circuit 17, and a NOT circuit 18. The IF signal converted into a digital signal by the A / D conversion circuit 4 in FIG. 1 is input to the LPF 11. The LPF 11 selectively passes only the low frequency component. For example, the cut-off frequency Fc is set to 50 kHz, and a signal component in a high frequency band exceeding the cut-off frequency Fc, for example, an IF signal component of about 450 kHz is attenuated greatly by the LPF 11 so as not to pass through the LPF 11 substantially. If the IF signal includes a low frequency component by amplitude modulation, the LPF 11 selectively passes only the low frequency component from the IF signal. The LPF 11 and the HPF 12 can be realized by digital signal processing, but can also be subjected to equivalent filter processing by analog signals.

  When noise is mixed in the received signal, the low frequency component passing through the LPF 11 also includes noise. Among noises, pulse noise in particular has a component over a wide frequency range although it lasts only for a short time. Among the low-frequency components passing through the LPF 11, the audio frequency component of AM broadcasting is up to about several kHz. Therefore, for example, a component of 10 kHz or higher is highly likely to correspond to pulse noise. The HPF 12 selectively passes only a frequency component higher than the cutoff frequency Fc, for example, with 15 kHz as a cutoff frequency Fc. This high frequency component is input to the noise detection circuit 13. The noise detection circuit 13 detects pulse noise.

  When pulse noise is detected by the noise detection circuit 13, the noise frequency detection circuit 14 detects the occurrence frequency of the detected pulse noise. The control signal representing the frequency detection result by the noise frequency detection circuit 14 is a digital signal indicating that the logical value is “0” and the frequency is lower than the reference value, and the logical value “1” is that the frequency is higher than the reference value. With this control signal, the normal processing circuit 16 or the selection processing circuit 17 in the noise removal circuit 15 is switched. The normal processing circuit 16 performs a noise cut process for cutting a signal in a portion where, for example, pulse noise is generated. The selection processing circuit 17 performs noise suppression processing different from the noise cut processing. The control signal is supplied to the normal processing circuit 16 via a NOT circuit 18 that inverts the logic, and is directly supplied to the selection processing circuit 17 from the noise frequency detection circuit 14. The normal processing circuit 16 and the selection processing circuit 17 perform noise removal processing from the signal when the logical value of the given control signal becomes “1”, and simply allow the signal to pass when the logical value becomes “0”. It becomes. The DETOUT signal from the detection circuit 5 in FIG. 1 is input to the noise removal circuit 15 as a signal to be subjected to noise removal.

  That is, the noise detection apparatus 10 of the present embodiment includes a noise detection circuit 13 as noise detection means, a noise removal circuit 15 as noise removal means, and a noise frequency detection circuit 14 as noise discrimination means. The noise detection circuit 13 detects noise mixed in the signal. The noise removal circuit 15 can select different noise removal processing for the signal depending on the nature of the noise. Since the noise frequency detection circuit 14 determines the nature of the noise detected by the noise detection circuit 13 and selects the noise removal processing in the noise removal circuit 15 according to the determination result, the noise mixed in the signal is appropriately removed. can do. Note that the noise removal circuit 15 may be provided with a plurality of processing circuits as the selection processing circuit 17 and switched according to the nature of the noise.

  Further, the noise removal processing in the noise removal circuit 15 as noise removal means is selected when at least normal processing in the normal processing circuit 16 selected as the initial state and predetermined noise characteristics are discriminated. Since the selection processing in the selection processing circuit 17 can be selected, for example, normal processing is determined corresponding to the nature of noise that is highly likely to be mixed in the signal, and the nature of noise that is likely to be next mixed is handled. If the selection process is determined, noises having different properties can be appropriately removed.

  Further, the noise removing circuit 15 as noise removing means can remove the pulse noise by a plurality of different processes including a normal process by the normal processing circuit 16 and a selection process by the selection processing circuit 17 according to the repetition frequency. . The noise frequency detection circuit 14 as the noise discrimination means is discriminated as pulse noise by the pulse nature discrimination means for judging whether or not the noise detected by the noise detection circuit 13 is pulse noise. And a frequency detection means for detecting the repetition frequency of the noise, and the selection process is selected when the repetition frequency detected by the frequency detection means exceeds a preset standard as the nature of the noise. The removal process can be appropriately performed according to the repetition frequency.

  Regarding detection and removal of pulse noise, for example, paragraphs [0002] to [0006] of Japanese Patent No. 2755497 (Patent Document 1) and FIGS. 16 to 19 show a conventional pulse noise detection circuit in an AM receiver. Configuration and operation are described. This configuration is the same as that in FIG. 1 in that pulse noise is detected from the IF signal and pulse noise is removed from the output of the detector. In paragraphs [0011] to [0018] and FIGS. 1 to 4 of Patent Document 1, a configuration and operation for detecting and removing pulse noise from the output of the AM tuner on the upstream side of the prior art are described. ing. Paragraphs [0019] to [0034] and FIGS. 5 to 15 describe a configuration and operation for detecting and removing pulse noise from the output of the detector on the downstream side of the prior art. Also in the present invention, the detection and removal of the pulse noise can be performed on the front stage side or the rear stage side as compared with FIG.

  In the present invention, the detection and discrimination of pulse noise and the removal thereof may basically be performed in accordance with the techniques disclosed in the aforementioned Patent Documents 1 to 3. The techniques disclosed in these documents can be replaced by digital signal processing. The determination of pulse noise can be made, for example, based on whether or not the instantaneous value of the signal level is higher than the average level. Note that the order of the serial configuration such as the normal processing circuit 16 and the selection processing circuit 17 can be changed. The same applies to the forms described below.

FIG. 3 shows another embodiment of the present invention, which is a signal filter means for applying a filter process for preventing false detection to a signal subjected to noise detection by a noise detection circuit 13 which is a noise detection means. A schematic electrical configuration of a main part of the noise removing device 20 further including a filter circuit 21 is shown. Since the signal filter circuit 21 performs filter processing for preventing erroneous detection, the noise detection circuit 13 can appropriately perform noise detection. The signal filter circuit 21 as the signal filter means is an absolute value (hereinafter abbreviated as “ABS”) circuit 21.
Averaging filter circuit 23 is included.

  4 shows the signal filter circuit 21 shown in FIG. 3 in which the noise waveform input to the ABS circuit 22 is A, the noise waveform output from the ABS circuit 22 and input to the Averaging filter circuit 23 is B, and the Averaging filter circuit 23. The output noise waveform is indicated by C. Even if it is pulse noise, the noise waveform may be accompanied by a plurality of vibrations as shown in A, and if the noise is detected faithfully by the pulse detection circuit 13, it may be erroneously detected as a plurality of pulse noise. There is sex. Since the ABS circuit 22 converts the noise waveform A into a noise waveform B indicating an absolute value, and further performs averaging filter processing in the averaging filter circuit 23, it can be detected as one noise like the noise waveform C. Therefore, the noise detection circuit 13 after the signal filter circuit 21 can reliably generate one noise detection flag for one pulse noise, and the noise frequency detection circuit 14 and the like can also prevent malfunction. be able to.

  FIG. 5 shows a schematic electrical configuration of a main part of the noise removing device 30 as still another embodiment of the present invention. In the noise removing device 30, the rising edge of the clock signal having a fixed period generated by the clock (CLOCK) generating circuit 31 is converted into a falling edge by the NOT circuit, and the noise detection result is obtained by the noise detecting circuit 33 for each period. A noise frequency detection circuit 34 for adding a noise detection flag representing a logical value if there is a rise. The noise detection circuit 33 is the same as the noise detection circuit 13 shown in FIG. 1 and the like except that a pulse signal indicating the rise of the noise detection flag is derived.

  The noise frequency detection circuit 34 includes a counter 35, a multiplication circuit 36, a delay circuit 37, a multiplication circuit 38, a NOT circuit 39, a sample hold circuit 40, a multiplication circuit 41, a delay circuit 42, a comparison circuit 43, and a reference value setting circuit 44. Including. Such a configuration can be easily realized by a digital signal processing program using a DSP.

  6 and 7 show signal waveforms at the respective parts A to I in FIG. 6 and FIG. 7, the horizontal axis indicates the change over time, but FIG. 7 shows the change over a long period of time compared to FIG.

  6A shows an example of a noise detection flag in the noise detection circuit 33. FIG. FIG. 6B shows a signal obtained by extracting one sample of the rising edge of the noise detection flag. C in FIG. 6 shows a signal obtained by adding the B signal in the counter 35 in FIG. 6D shows a signal obtained by inverting the rising edge of the clock signal supplied from the NOT path 32 of FIG. 5 to the multiplication circuit 36. This signal is a reset signal for the counter 35. 5 receives the signal C obtained by sending the output of the counter 35 by the delay circuit 37 for one period and the signal B from the NOT circuit 32, and supplies a reset signal to the counter 35. The reset signal is given to the counter 35 based on the signal B when the signal C is 1 or more. One period in which the delay circuit 37 delays the signal C corresponds to a sample period of noise detection in the noise detection circuit 33. 6E shows an output waveform of the delay circuit 37 when the counter 35 is reset at the falling edge of the reset signal.

  E of FIG. 7 shows the count signal from the counter 35 which is the output of the delay circuit 37 of FIG. 5 with the time axis reduced compared to FIG. Since the time axis is reduced, the stepwise change as shown in E of FIG. 6 is shown as a continuous change. The count value immediately before reset corresponds to the frequency of pulse noise detection. F in FIG. 7 shows a reset signal supplied from the NOT circuit 35 in FIG. 5 to the multiplier 38. G in FIG. 7 indicates a signal output from the multiplier 38 in FIG. This signal G becomes the value of the count signal E when the F reset signal rises. H in FIG. 7 indicates an output signal of the sample hold circuit 40 that holds the value of the G signal by the rising stripe of the next reset signal F. Since the value of the signal H represents the frequency of pulse noise detection, the reference value Vth with the magnitude of the frequency is set in the reference value setting circuit 44 in FIG. 5, and the H signal is discriminated by the reference value Vth in the comparison circuit 43. A control signal as indicated by I in FIG. 7 can be obtained.

  In the present embodiment, the pulse frequency detection circuit 34 serving as a frequency detection unit detects the frequency from the number of detections of pulse noise within a certain interval, so that it corresponds to the number of detections of pulse noise detected within the certain interval. Thus, it is possible to appropriately select processing for noise removal. That is, the noise detection flag as shown in A of FIG. 6 is converted into a signal E that represents the number of noises in a certain interval that is the period of the reset signal shown in D, and the maximum value of this signal E is converted to the reset signal of D. The time until it rises is held as indicated by H in FIG. Thereby, it is possible to know the repetition rate of the currently mixed noise almost in real time. Finally, a comparison is made with the speed (Vth) at which the signal separation of H in FIG. 7 is desired, and it is determined whether or not the currently mixed noise is faster or slower than Vth, and is shown in I of FIG. Generate a control signal. Thereafter, when the control signal I is at a high level with a logical value “1”, selection processing is performed, and when the control signal I is at a low level with a logical value “0”, optimal noise cancellation processing is performed according to the noise repetition rate. It becomes possible.

FIG. 8 shows the concept of the frequency detection means in still another embodiment of the present invention. In this embodiment, since the frequency detection means detects the frequency from the interval of the noise detection flag indicating the pulse noise detection, the processing for noise removal is appropriately selected according to the time interval at which the pulse noise is detected. can do. For example, if the sample frequency fs is 48 kHz, the count value of the counter between the noise detection flags is 48, and the count value increases by 1 for one sample, the time can be obtained as follows.
Time = count value of counter / sample frequency = 48/48000
= 0.001 (s)
= 1 (ms)

  Since 1 ms corresponds to a frequency of 1 kHz, 1 kHz can be obtained as a frequency corresponding to the noise generation speed from the time interval of 1 ms of the noise detection flag. By calculating the exact time (frequency) between noises, it is possible to control the noise removal circuit in accordance with the noise repetition rate.

  FIG. 9 schematically shows a partial electrical configuration of a noise removing device 50 to which this concept is applied. The noise frequency detection circuit 51 as the frequency detection means includes a counter 52 and a clock circuit 53. The clock circuit 53 generates a clock signal having a sample frequency fs at a cycle shorter than the time assumed as the interval of the noise detection flag. The counter is reset, for example, every time the noise detection flag rises, and counts the clock signal during a non-reset period. Comparing the count value of the counter 52 with the reference value, when the count value is larger than the reference value, it can be determined that the interval between the noise detection flags is large and the frequency of occurrence of the pulse noise is low. When the count value of the counter 52 is smaller than the reference value, it can be determined that the interval between the noise detection flags is small and the occurrence frequency of the pulse noise is high.

  FIG. 10 shows the concept of the frequency detection means in still another embodiment of the present invention. In this embodiment, the noise frequency detection circuit calculates the result of a signal to be subjected to a calculation of + A when noise is detected by the noise detection circuit and −B (A and B are arbitrarily set positive constants) when no noise is detected. Is compared with a reference value to determine the magnitude of the noise frequency.

  The signal in FIG. 10 operates so that the level is high when the noise frequency is high and the level is low when the frequency is low. Therefore, information on the noise frequency can be obtained using this signal. Further, this signal can be realized simply by setting the high level of the noise detection flag to A and the low level to -B, and then continuing to add the signal. Compared with the method shown in FIG. It can be configured easily and easily.

  FIG. 11 schematically shows a partial electrical configuration of a noise removal device 60 to which the concept of FIG. 10 is applied. The noise frequency detection circuit 61 includes a clock circuit 62, an initial value setting circuit 63, an addition circuit 64, a subtraction circuit 65, and a comparison circuit 66. For example, the clock circuit 62 generates a clock signal in accordance with the shortest cycle in which noise is detected by the noise detection circuit 13. The initial value setting circuit 63 generates a certain initial value, for example, every certain time. The adder circuit 64 and the subtractor circuit 65 add + A to the initial value or the calculated value based on the initial value in the adder circuit 64 if noise is detected in accordance with the noise detection flag of the noise detection circuit 13 for each cycle of the clock signal. If no noise is detected, -B is subtracted.

  That is, the noise frequency detection circuit 61 as frequency detection means includes an addition circuit 64 and a subtraction circuit 65 as addition / subtraction means. The adding / subtracting unit performs a predetermined addition process or an initial value set in the initial value setting circuit 63 depending on whether or not the noise detection circuit 13 detects pulse noise at every predetermined time. Perform subtraction. Since the noise frequency detection circuit 61 detects the frequency based on the processing results of the addition circuit 64 and the subtraction circuit 65 at a predetermined time, the noise frequency detection circuit 61 appropriately selects a process for noise removal according to the addition / subtraction process result. be able to.

  FIG. 12 schematically shows a partial electrical configuration of a noise removing device 70 as still another embodiment of the present invention. The noise frequency detection circuit 71 of this embodiment includes an LPF 72. The LPF 72 performs low-pass frequency filtering on the signal A averaged by the signal filter circuit 21, extracts a direct current (DC) component, and derives it as a signal B.

  FIG. 13 (a) shows the signals A and B of FIG. 12 superimposed on the same graph. The LPF 72 integrates the signal A averaged by the signal filter circuit 21 and converts the signal A into a slowly changing signal such as the signal B. FIG. 13B shows a control signal obtained by discriminating the output signal B of the LPF 72 shown in FIG. 13A with the reference value Vth. As shown in FIG. 12, the IF signal is extracted through the LPF 11, HPF 12, ABS circuit 21 and Averaging filter circuit 23, and the signal A having only the noise component is extracted, and the signal B having only the DC component is generated by passing through the LPF 72. The B signal generates a control signal by utilizing the fact that the level becomes large when the noise frequency is high. Although the operation itself of this embodiment is the same as that of FIGS. 10 and 11, in this embodiment, since it is affected by the noise peak value, when the noise level is high and the frequency is high, the selection process is switched. It can be operated reliably.

  As described above, when the frequency detection means includes the LPF 72 as the noise extraction means, only the noise component can be extracted from the signal. Since the frequency detection means detects the frequency based on the signal level of the noise component extracted by the noise extraction means, processing for noise removal is appropriately performed according to the signal level of the noise component extracted from the signal. You can choose.

  The frequency detection means includes an LPF 72 as a direct current means. The direct current means converts the noise component detected by the noise extraction means into direct current. Since the frequency detection means uses the direct current level output from the direct current means as the signal level, processing for noise removal can be appropriately selected according to the direct current level.

  As described above, the noise frequency detection circuit 71 as the frequency detection unit of the noise determination unit can detect the frequency from the generation speed of the pulse noise. When the frequency is detected from the generation speed, it is possible to appropriately select a process for noise removal according to the speed at which the pulse noise is generated.

  FIG. 14 shows a concept in still another embodiment of the present invention. In each of the above-described embodiments, the noise removal processing in the noise removal means is switched between the normal processing and the selection processing according to the logic level of the control signal as shown in FIG. Then, as shown in FIG. 14B, a time constant is added at the time of switching the processing so that the switching is performed slowly. As shown in FIG. 14 (c), by adding a time constant to the control signal of (a) as shown in (b), the auditory sensation after sounding the signal when switching from normal processing to selection processing is performed. The above sudden change can be alleviated and auditory performance can be improved. In the portion surrounded by the broken line in (b), the DETOUT signal of FIG. 1 which is the original signal and the signal after selection processing are weighted and mixed, and the weight distribution is selected from 1 to 0 for the original signal. The processed signal is temporally changed from 0 to 1. This time is, for example, about 1 second.

  In this way, the selection changing means for controlling the change of the selection of the noise removal processing in the noise removal means based on the determination result of the noise detection circuit 71 as the noise determination means to be performed with a change over a predetermined time, If included, for example, when a signal from which noise is removed is output as an acoustic signal, it is possible to avoid an abrupt change in audibility at the time of switching signal processing and to improve the audibility performance.

  FIG. 15 shows a schematic electrical configuration of the main part of the noise removal device 80 that switches the noise removal processing by adding a time constant as shown in FIG. The control signal output from the noise frequency detection circuit 14 is input to the normal processing circuit 16 and the selection processing circuit 17 via the time constant circuit 81. The time constant circuit 81 includes a time constant setting circuit 82 that sets a time constant in the control signal to the normal processing circuit 16 and a time constant setting circuit 83 that sets a time constant in the control signal to the selection processing circuit 17. In FIG. 2, the normal processing circuit 16 and the selection processing circuit 17 are connected in series. However, in the noise removal circuit 85 of the present embodiment, the normal processing circuit 16 and the selection processing circuit 17 are mixed in order to be weighted. Are connected in parallel, and the output side is mixed by the mixing circuit 86.

  FIG. 16 shows a concept in still another embodiment of the present invention. In FIG. 14, the time constant setting circuits 82 and 83 set equivalent time constants for the normal process and the selection process, but rather than the change A at the start of switching from the normal process to the selection process, By setting a longer time constant to the change B at the fall of the switching from the selection process to the normal process, the audibility performance can be further improved. This embodiment can also be realized by the noise removal circuit 85 shown in FIG.

  That is, the noise removal circuit 85 as a noise removal unit can select at least a normal process selected as an initial state and a selection process selected when a predetermined noise characteristic is determined as the noise removal process. is there. The time constant circuit 81 as selection changing means switches the time constant set by each of the time constant setting circuits 82 and 83 between the rising edge and the falling edge of the control signal, and the time required for changing from the normal process to the selection process. Rather, the time required for changing from the selection process to the normal process is set to a long time.

  For example, if the nature of the noise that selects the selection process is greater than the nature of the noise that selects the normal process, and the degree of interference with the signal is greater, it will take less time to switch to the selection process. It is possible to quickly avoid the influence of disturbance on the signal due to the difference in the nature of noise. If switching to the selection process becomes unnecessary, it is possible to return to the normal process over time, avoid a sudden change in audibility due to the process switching, and further improve the audibility performance.

  FIG. 17 shows a schematic electrical configuration of a noise removing device 80 as still another embodiment of the present invention. In the present embodiment, the noise removal circuit 91 includes a noise cut processing circuit 92 for normal processing and a high cut processing circuit 93 for selection processing. The noise removal circuit 91 performs high-cut processing to suppress high-band components on the detection signal (DETOUT) when the noise frequency is high, and noise that cuts only the noise generation section for low-frequency single-shot noise. Cut processing is performed as normal processing.

  The noise cut processing circuit 92 can be configured in the same manner as a noise canceller as disclosed in, for example, the above-described Patent Documents 1 to 3. In the noise canceller, the noise generation interval is interpolated with the signal value at the start of the noise generation interval. Since the noise cut processing circuit 92 performs the noise cut processing that cuts off only the noise detection section in the signal as a normal process, it is surely removed if, for example, pulse noise is single-shot or low in frequency. can do.

  Further, in the noise removal circuit 91, the high cut processing circuit 93 performs high cut processing for suppressing high frequency components in the signal as selection processing, so that a lot of pulse noise with many high frequency components is mixed in the signal as noise. In such a case, it is possible to suppress the noise component relatively larger than the signal. As a normal process, for example, when a noise cut process is performed, the rate at which a signal is cut off for noise removal increases according to a high frequency component, and information is lost from the signal. If the high frequency components are suppressed, even if the high frequency components of the signal are suppressed, the suppression is relatively less than noise, and the signal-to-noise ratio is relatively increased, thereby damaging the information contained in the signal. You can avoid it. As the high cut processing, for example, signal composition in a frequency band higher than 1.5 kHz may be suppressed.

  FIG. 18 shows the processing of the main part when the noise removal apparatus 90 of FIG. This processing is repeatedly performed by being activated by, for example, a timer interrupt or the like while the noise removing device 90 is operating. The process is started from step s0, and in step s1, it is determined whether or not noise is detected by the noise detection circuit 13. If noise is detected, the process proceeds to step s2, and the noise frequency detection circuit 14 determines whether or not the noise repetition speed is slower than the reference speed Vth. If it is determined in step s2 that the process is late, the process proceeds to step s3, and the noise cut process by the noise cut process circuit 92 is selected. If it is determined that the process is not late, the process proceeds to step s4 and the high cut process circuit 93 performs a high cut. Select a process. When no noise is detected in step s1, or when the processing in steps s3 and s4 is completed, the entire processing is temporarily ended in step s5. Note that the process of FIG. 18 can be executed as a repeated loop process instead of an interrupt process. In the case of loop processing, it is only necessary to return to step s1 instead of once ending in step s5.

  In general, when noise with a high repetition rate is mixed, a high-frequency harsh sound is output to the detection signal. This annoying sound cannot be sufficiently removed by the noise cut processing. In the present embodiment, high-cut processing is performed in which high-frequency noise is suppressed when noise with a high repetition rate is mixed, and main information of signals such as broadcast contents is left. For noise with a slow repetition rate, such as ignition (IG) noise, which is likely to be mixed in in-vehicle receivers, etc., perform pre-value interpolation processing with noise cut processing that cuts only the noise generation part of the signal Thus, it is possible to perform an optimum noise removal process according to the noise repetition rate.

  FIG. 19 shows a schematic electrical configuration of a noise removal apparatus 100 as still another embodiment of the present invention. In this embodiment, the noise removal circuit 101 includes a noise cut processing circuit 92 for normal processing and a mute processing circuit 103 for selection processing. The noise removal circuit 101 performs a mute process that suppresses the level of the output signal as a selection process when the frequency of noise is high, and a normal process of a noise cut process that cuts only the noise generation section for single noise with a low frequency. Do as.

  20A shows an example of the waveform of the control signal output from the noise frequency detection circuit 14 in FIG. 19A, and FIG. 20B shows the mute control signal corresponding to the change in the mute amount in the mute processing circuit 103. An example of a waveform is shown. In the mute processing circuit 103, the control signal is inverted and a mute level is set in a portion where the logical value is “0”. The mute level is set to an intermediate value between 0 and 1, for example, 0.7. The section in which the logical value of the control signal is “0” is a section without mute, and the mute processing circuit 103 passes the output from the noise cut processing circuit 92 on which the noise cut process is performed without performing the mute process. The section in which the logical value of the control signal is “1” is a mute section, and the mute processing circuit 103 corresponds to the mute level with respect to the output from the noise cut processing circuit 92 that has not been subjected to noise cut processing. Do suppression.

  Since the noise removal circuit 101 as noise removal means performs mute processing for suppressing the amplitude level of the signal as selection processing, the peak of pulse noise and the like protruding beyond the signal amplitude is reduced, and signal-to-noise is relatively reduced. The ratio can be increased so that the information contained in the signal is not impaired. When the noise frequency is high (when the control signal is a logical value “1”), it is possible to suppress annoying sound due to noise by muting the level of the detection signal to a certain level. In the mute processing circuit 103, a mute control signal for controlling the mute amount can be easily generated by using a control signal from the noise frequency detection circuit 14, as shown in FIG.

  FIG. 21 shows a schematic electrical configuration of a noise removing apparatus 110 as still another embodiment of the present invention. In the present embodiment, the noise removal circuit 111 includes a noise cut processing circuit 92 for normal processing, a high cut processing circuit 93 for selection processing, and a mute processing circuit 103. In the noise removal circuit 111, when the frequency of noise is high, the high cut processing by the high cut processing circuit 93 and the mute processing for suppressing the level of the output signal by the mute processing circuit 103 are simultaneously performed as selection processing, so that auditory performance can be improved. . For single noise with a low frequency, a noise cut process for cutting only the noise occurrence section is performed as a normal process.

  FIG. 22 shows a schematic electrical configuration of a noise removing device 120 as still another embodiment of the present invention. In the present embodiment, the noise removal circuit 121 includes a noise cut processing circuit 92 for normal processing, a high cut processing circuit 122 for selection processing, and a mute processing circuit 123, which are selected by a selector circuit 124. Use. The high cut processing circuit 122 varies the cutoff frequency Fc for performing the high cut processing in accordance with the S meter signal corresponding to the electric field strength detection result of the receiver. Also, the mute processing circuit 123 varies the mute amount, which is the signal suppression amount in the mute processing, according to the S meter signal. The S meter signal can be obtained from the S meter circuit 8 of FIG. In broadcast receivers, etc., it is difficult to understand the contents of the broadcast when the electric field is weak. Lowering the cut-off frequency Fc or increasing the amount of mute reduces the annoying effects of noise even if the contents of the broadcast become more difficult to understand. The effect is greater. In the noise cut processing, the broadcast content is cut, and the broadcast content becomes more difficult to understand.

  That is, the noise removal circuit 121 as the noise removal unit changes the degree of suppression in the mute process and the degree of suppression of the high frequency component in the high cut process according to the signal intensity. If the signal becomes weaker, the degree of suppression can be increased, the signal-to-noise ratio can be relatively increased, and the information contained in the signal can be easily extracted. In addition, if the intensity of the signal is high, the degree of suppression can be reduced to reduce the influence on the information included in the signal, and the information included in the signal can be prevented from being damaged. In the receiver including the noise removal device 120, for example, the degree of suppression in noise removal can be appropriately changed according to the received electric field strength indicated by the S meter signal or the like.

  FIG. 23 shows a schematic electrical configuration of a noise removing device 130 as still another embodiment of the present invention. In this embodiment, the noise removal circuit 131 includes a noise cut processing circuit 92 for normal processing, a high cut processing circuit 132 for selection processing, and a mute processing circuit 133, and is selected by a selector circuit 124. Use. In the high cut processing circuit 132, the cutoff frequency Fc for performing the high cut processing is varied according to the volume amount (Vol level) of the variable volume adjuster provided in the low frequency circuit 6 or the like in the receiver of FIG. Further, the mute amount of the mute processing circuit 133 is also varied according to the volume amount.

  Generally, in a signal processing system such as a receiver, variable means for changing a signal transmission amount such as a volume in low frequency processing is provided in a signal transmission path. Since the noise removal circuit 131 as the noise removal means changes the degree of suppression according to the transmission amount of the variable means, if the transmission amount is small, the degree of suppression is increased and the signal-to-noise ratio is relatively increased. The information contained in the signal can be easily extracted. Also, if the amount of transmission is large, the degree of suppression can be reduced to reduce the influence on the information held by the signal so that the information held by the signal is not impaired.

  FIG. 24 shows a schematic electrical configuration of a noise removing device 140 as still another embodiment of the present invention. In the present embodiment, the noise removal circuit 141 includes a noise cut processing circuit 92 for normal processing, a high cut processing circuit 142 for selection processing, and a mute processing circuit 143, which are selected by a selector circuit 124. Use. Furthermore, a noise speed calculation circuit 144 that calculates the noise generation speed based on the noise frequency detected by the noise frequency detection circuit 14 is provided. The noise speed calculation circuit 144 calculates a noise repetition speed and derives an output according to the speed. The high cut processing circuit 142 linearly changes the cutoff frequency Fc for performing the high cut processing according to the output derived from the noise speed calculation circuit. Further, the mute amount of the mute processing circuit 143 is also changed linearly according to the output derived from the noise speed calculation circuit. This change can be changed in some functional relationship such as stepwise.

  In addition to the control signal for switching the processing in the noise removal circuit 141 from the noise frequency detection circuit 14, a signal related to the noise speed is output, the noise speed calculation circuit 144 calculates the noise speed, and the high cut processing circuit 142 is calculated based on the calculation result. Since the mute amount is made variable by the cut-off frequency Fc and the mute processing circuit 143, the cut-off frequency in the high cut processing and the mute processing in the mute processing according to the repetition rate of the noise mixed at the time of noise detection. The degree of suppression can be changed appropriately.

  The noise speed calculation circuit 144 changes the degree of suppression of the high frequency component in the high cut processing by the high cut processing circuit 142 according to the repetition frequency of the noise detected by the noise frequency detection circuit 14 as the frequency detection means. It may be. For example, if the frequency of noise repetition is high, the effect of noise increases, so the degree of suppression of high-frequency components in high-cut processing is increased, the signal-to-noise ratio is relatively increased, and the information contained in the signal is extracted. It can be made easier. Also, if the repetition frequency is small, the degree of suppression of the high frequency component can be lowered, the influence on the information that the signal has can be reduced, and the information that the signal has can be prevented from being damaged.

  FIG. 25 shows a schematic electrical configuration of a noise removing device 150 as still another embodiment of the present invention. In the present embodiment, the noise removal circuit 151 includes a noise cut processing circuit 92 for normal processing and a band stop filter (BSF) 152 for selection processing, which is selected by a selector circuit 124. The band stop filter 152 includes a noise speed calculation circuit 144 that calculates a noise generation speed based on the noise frequency detected by the noise frequency detection circuit 14 only for the cutoff frequency Fc. The noise speed calculation circuit 144 calculates a noise repetition speed and derives an output according to the speed. The high cut processing circuit 142 cuts off only the signal component near the cutoff frequency Fc for performing the high cut processing. In the BSF 152, the cutoff frequency Fc is variable and is changed in accordance with the frequency derived from the noise speed calculation circuit.

  The noise frequency detection circuit 14 outputs a signal related to the speed of noise separately from the control signal for switching the noise removal processing, and the noise frequency calculation circuit 144 calculates the repetition frequency of the noise. According to the result, the noise removal circuit 151 as the noise removal means makes the cut-off frequency Fc of the BSF 152 variable, so that only the noise component can be cut off.

  FIG. 26 shows a schematic electrical configuration of a noise removal device 160 as still another embodiment of the present invention. In the present embodiment, the noise removal circuit 161 includes a noise cut processing circuit 92 for normal processing, a high cut processing circuit 93 for selection processing, and a mute processing circuit 103. The noise removal circuit 161 further includes variable gain amplifiers 162 and 163 and a mixing circuit 164. The amplifiers 162 and 163 amplify the outputs from the noise cut processing circuit 92 and the mute processing circuit 103, respectively, and the amplification gains GAIN1 and GAIN2 are determined according to the noise speed calculated by the noise speed calculation circuit 165, respectively. The noise frequency detection circuit 14 outputs a signal related to the noise speed, and the noise speed calculation circuit 144 calculates a noise repetition speed and generates a GAIN1 control signal and a GAIN2 control signal according to the speed. When the frequency of noise is high, the GAIN2 weight of the amplifier 163 is increased, and when the frequency is low, the GAIN1 weight of the amplifier 162 is increased. Deriving the output. The high cut processing circuit 142 blocks only the signal component near the cutoff frequency Fc for performing the high cut processing.

  The noise removal circuit 161 as the noise removal unit performs the selection process together with the normal process by adding a weight corresponding to the repetition rate of the noise detected by the noise detection circuit 13 as the noise detection unit. If the difference between the two increases, the weight can be increased to increase the degree of selection processing, and appropriate noise removal can be performed.

  As described above, if the noise removal apparatus of each form is provided, it is possible to select an appropriate process according to the nature of the mixed noise with respect to the reception signal of the receiver. As a receiver, not only AM broadcast reception but also radio communication reception by various amplitude modulation methods, FM broadcast reception, etc., according to each signal, and processing can be switched according to the nature of noise. Appropriate noise removal can be performed. The signal that is subject to noise removal is not only a signal that is modulated with a low-frequency signal such as audio, but even a video or data signal, noise can be removed to improve display quality and transmission quality. .

1 is a block diagram showing a schematic electrical configuration of a receiver 1 as an embodiment of the present invention. It is a block diagram which shows schematic electrical structure of the noise removal apparatus 10 of FIG. It is a block diagram which shows schematic electrical structure of the principal part of the noise removal apparatus 20 as other embodiment of this invention. It is a wave form diagram which shows the change of the noise waveform in the signal filter circuit 21 of FIG. It is a block diagram which shows schematic electrical structure about the principal part of the noise removal apparatus 30 as further another form of implementation of this invention. It is a wave form diagram which shows the signal waveform in each part of A-E of Drawing 5, respectively. FIG. 6 is a waveform diagram showing signal waveforms at respective parts E to I in FIG. 5. It is a wave form diagram which shows the view of the frequency detection means in the further another form of implementation of this invention. FIG. 9 is a block diagram schematically showing a partial electrical configuration of a noise removal device 50 to which the concept of FIG. 8 is applied. It is a wave form diagram which shows the view of the frequency detection means in the further another form of implementation of this invention. It is a block diagram which shows roughly the partial electrical structure of the noise removal apparatus 60 to which the view of FIG. 10 is applied. It is a block diagram which shows roughly the partial electrical constitution of the noise removal apparatus 70 as further another form of implementation of this invention. It is a wave form diagram which shows the signal waveform in each part of the noise frequency detection circuit 71 of FIG. It is a wave form diagram which shows the view in further another form of implementation of this invention. FIG. 15 is a block diagram showing a schematic electrical configuration of a main part of a noise removal device 80 that switches a noise removal process by adding a time constant as shown in FIG. 14. It is a wave form diagram which shows the view in further another form of implementation of this invention. It is a block diagram which shows schematic electrical structure of the noise removal apparatus 80 as further another form of implementation of this invention. It is a flowchart which shows the process of the principal part in the case of implement | achieving the noise removal apparatus 90 of FIG. 17 by the program operation | movement etc. of DSP. It is a block diagram which shows schematic electrical structure of the noise removal apparatus 100 as further another form of implementation of this invention. FIG. 20 is a waveform diagram illustrating a relationship between the control signal and the mute control signal in FIG. 19.

It is a block diagram which shows schematic electrical structure of the noise removal apparatus 110 as further another form of implementation of this invention. It is a block diagram which shows schematic electrical structure of the noise removal apparatus 120 as further another form of implementation of this invention. It is a block diagram which shows schematic electrical structure of the noise removal apparatus 130 as further another form of implementation of this invention. It is a block diagram which shows schematic electrical structure of the noise removal apparatus 140 as further another form of implementation of this invention. It is a block diagram which shows schematic electrical structure of the noise removal apparatus 150 as further another form of implementation of this invention. It is a block diagram which shows schematic electrical structure of the noise removal apparatus 160 as further another form of implementation of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Receiver 3 IF amplifier circuit 5 Detection circuit 8 S meter circuit 10, 20, 30, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160 Noise removal apparatus 11, 72 LPF
13, 33 Noise detection circuit 14, 34, 51, 61, 71 Noise frequency detection circuit 15, 85, 91, 101, 111, 121, 131, 141, 151, 161 Noise removal circuit 16 Normal processing circuit 17 Selection processing circuit 21 Signal filter circuit 23 Averaging filter circuit 35, 52 Counter 40 Sample hold circuit 43 Comparison circuit 63 Addition circuit 64 Subtraction circuit 81 Time constant circuit 82, 83 Time constant setting circuit 92, 142 Noise cut processing circuit 93, 122, 132, 142 High cut Processing circuit 103, 123, 133, 143 Mute processing circuit 144, 165 Noise speed calculation circuit 162, 163 Amplifier

Claims (24)

Noise detection means for detecting noise mixed in the signal;
Noise removal means that can select different noise removal processing depending on the nature of the noise for the signal;
A noise removing device comprising: noise determining means for determining the nature of noise detected by the noise detecting means and selecting a noise removing process in the noise removing means in accordance with the determination result.   The noise removing means can select at least a normal process selected as an initial state and a selection process selected when a predetermined noise characteristic is determined as the noise removal process. The noise removal device according to claim 1.   2. The selection change means for controlling the selection of the noise removal processing in the noise removal means based on the determination result of the noise determination means to be performed with a change over a predetermined time. The noise removal apparatus as described. The noise removal means can select a normal process selected as at least an initial state and a selection process selected when a predetermined noise property is determined as the noise removal process,
4. The selection change means, wherein the time required for the change from the selection process to the normal process is longer than the time required for the change from the normal process to the selection process. Noise removal device. The noise removing means can remove the pulse noise by a plurality of different processes including the normal process and the selection process according to the repetition frequency,
The noise discrimination means is
Pulseity determination means for determining whether or not the noise detected by the noise detection means is pulsed noise;
A frequency detecting means for detecting a repetition frequency of noise determined as pulsed noise by the pulse determining means,
5. The noise removal apparatus according to claim 2, wherein the selection process is selected when the repetition frequency detected by the frequency detection unit exceeds a preset reference as the nature of the noise.   6. The noise removing apparatus according to claim 5, wherein the frequency detecting means detects the frequency from the generation speed of the pulse noise.   6. The noise removing apparatus according to claim 5, wherein the frequency detecting means detects the frequency from the number of detections of the pulse noise within a certain section.   6. The noise removing apparatus according to claim 2, wherein the frequency detecting unit detects the frequency from the pulse noise detection interval. The frequency detection means includes
Addition / subtraction means for performing a predetermined addition process or subtraction process depending on whether or not the noise detection means detects pulse noise at predetermined time intervals with respect to a preset initial value,
6. The noise removing apparatus according to claim 5, wherein the frequency is detected based on a processing result of the adding / subtracting means at a predetermined time. The frequency detection means includes
Including noise extraction means for extracting only noise components from the signal;
6. The noise removing apparatus according to claim 5, wherein the frequency is detected based on a signal level of a noise component extracted by a noise extracting unit. The frequency detection means includes
Further comprising direct current means for directing the noise component detected by the noise extraction means,
11. The noise removing apparatus according to claim 10, wherein a DC level output from the DC unit is the signal level.   The noise according to any one of claims 5 to 11, wherein the noise removing unit performs a noise cut process for temporally blocking only a noise detection section in the signal as the normal process. Removal device.   The noise removal device according to any one of claims 5 to 12, wherein the noise removal unit performs high cut processing for suppressing high frequency components in the signal as the selection processing.   The noise removal apparatus according to claim 5, wherein the noise removal unit performs a mute process for suppressing an amplitude level of the signal as the selection process.   14. The noise removing apparatus according to claim 13, wherein the noise removing unit changes a cut-off frequency for performing the high cut processing in accordance with the intensity of the signal.   16. The noise detection unit according to claim 13, wherein the noise detection unit changes a degree of suppression of a high frequency component in the high cut processing according to a repetition frequency of noise detected by the frequency detection unit. Noise removal device.   16. The noise removing apparatus according to claim 13, wherein the noise removing unit changes the degree of suppression according to the intensity of the signal. The signal transmission path is provided with variable means for changing the transmission amount of the signal,
16. The noise removing apparatus according to claim 13, wherein the noise removing unit changes the degree of suppression in accordance with a transmission amount of the variable unit.   The noise removal means performs the selection processing on a component in a frequency band corresponding to a repetition rate of noise detected by the noise detection means. The noise removal apparatus as described.   The said noise removal means adds the weight corresponding to the repetition speed of the noise detected by the said noise detection means, and performs the said selection process with the said normal process. The noise removal apparatus as described in one.   The noise according to any one of claims 1 to 10, further comprising a signal filter unit that performs a filter process for preventing erroneous detection on a signal that is a target of noise detection by the noise detection unit. Removal device.   The noise removing apparatus according to claim 21, wherein the signal filter means performs an averaging filter process on the signal. A noise removing device according to any one of claims 1 to 22, comprising:
A receiver characterized in that a received signal is the signal. A noise removing device according to claim 15 or 16,
A receiver characterized in that the received signal is the signal and the received electric field strength is the signal strength.

Images