JP2012222661A - Fm radio demodulation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow multi-pass interference to be precisely detected even for an excessive modulation signal, thereby detecting and suppressing the occurrence of white noise due to FM demodulation obstruction by a relatively simple algorithm.SOLUTION: An FM radio demodulation system includes an FM detector (14) to detect an FM receiving signal; a stereo demodulation section (16) to perform stereo demodulation of a detection signal; and a control section (15) to control the stereo demodulation section (16) on the basis of the detection signal. The control section (15) includes a pilot signal strength detection section (HC22) to detect signal strength of a pilot signal contained in the detection signal; and a sideband signal strength detection section (HL21, HH23) to detect the signal strength of a sideband signal that is a signal of a neighborhood frequency band of the pilot signal. The control section (15) controls the stereo demodulation section (16) on the basis of a value of a ratio of the signal strength of the pilot signal and the signal strength of the sideband signal.

Description

  The present invention relates to an FM radio demodulation system that detects an FM demodulation failure during FM (frequency modulation) radio reception and suppresses the generation of noise.

  When FM demodulation fails due to multipath, jamming waves, excessive or insufficient radio wave intensity, circuit noise, etc. when receiving FM radio, large noise with whiteness that is uncomfortable to hearing is generated in the generated audio signal. . As a typical example, when FM radio broadcasting is received in an automobile moving in a building street, a phenomenon in which burst noise occurs due to local spatial power attenuation of radio waves accompanying multipath is known.

  Currently, as a noise generation detection means focusing particularly on multipath interference, a method for detecting a temporary decrease in the energy of a desired wave (a signal having a desired frequency), a method for detecting a decrease in the energy of a pilot signal, and a temporary decrease in the energy of the desired wave There is known a detection method in which a method for detecting a decrease is combined with a method for detecting an increase in the sideband energy of a pilot signal (for example, Patent Document 1).

Further, as a noise suppression method at the time of detecting the occurrence of noise, a suppression method by temporarily stopping an audio output signal, forced monauralization, or volume reduction is generally known.
By the way, as a disadvantage of the multipath interference detection method based on the temporary decrease of the desired wave energy, when the desired wave is overmodulated, the bandpass filter is usually used to detect the desired wave energy, so the output is instantaneous. There is a problem that it is lowered and it is erroneously determined as multipath interference. For such a problem, Patent Document 2 discloses a method of switching the detection threshold according to the degree of modulation.

JP 2002-271219 A JP 2009-278525 A

However, in the method of switching the detection threshold according to the degree of modulation as in Patent Document 2, the estimated value of the degree of modulation changes depending on the dispersion state of the modulated signal, so that it is difficult to control appropriately. is there.
On the other hand, in Patent Document 1, the occurrence of multipath interference is detected by combining the temporary decrease of the desired wave energy and the sideband energy of the pilot signal being equal to or greater than the threshold, but the inclusion of the L + R monaural signal component The accuracy of noise detection is low due to reasons such as level fluctuations. In addition, since noise generation detection means is not applied when it is not overmodulation, it is not possible to suppress noise generated by causes other than multipath interference.

In addition to multipath interference, FM demodulation failures may occur due to unexpected causes such as interference waves, excessive or excessive radio wave power, and noise on the circuit, which cause noise generation.
Therefore, in the present invention, in addition to accurately detecting multipath interference even with an overmodulated signal, the occurrence of white noise due to FM demodulation failure is compared with the level of noise regardless of the cause. An object of the present invention is to provide an FM radio demodulation system that can be detected and suppressed with a simple algorithm.

One aspect of the present invention for solving the above problems includes an FM detector that detects a frequency-modulated (FM) received signal, and a stereo demodulator that stereo-demodulates the detection signal output from the FM detector. A control unit that controls the stereo demodulation unit based on a detection signal output from the FM detector, wherein the control unit outputs the detection signal output from the FM detector. A pilot signal strength detection unit that detects the signal strength of a pilot signal included in the signal, and a sideband signal strength detection unit that detects the signal strength of a sideband signal that is a signal in the vicinity frequency band of the pilot signal, The stereo demodulator is controlled based on a ratio value between the detected signal strength of the pilot signal and the detected signal strength of the sideband signal. It is a FM radio demodulation system according to claim Rukoto.
According to this configuration, it is possible to accurately detect multipath interference even with an overmodulated signal, and to detect and suppress the occurrence of white noise due to FM demodulation failure with a relatively simple algorithm. be able to.

In another aspect of the present invention, the sideband signal strength detection unit is a signal of a high-frequency sideband signal that is a signal in the vicinity of the pilot signal and included in a higher frequency band than the pilot signal. A high-frequency sideband signal strength detection unit for detecting the strength, and a low-frequency sideband signal strength signal for detecting a signal strength of a low-frequency sideband signal that is a signal in the vicinity of the pilot signal and included in a frequency band lower than the pilot signal. A frequency sideband signal strength detection unit, wherein the control unit detects the detected signal strength of the pilot signal, the detected signal strength of the high frequency sideband signal strength, and the signal of the low frequency sideband signal strength. The FM radio demodulation system is characterized in that the stereo demodulation unit is controlled based on a ratio of the sum to the intensity.
According to this configuration, it is possible to accurately detect multipath interference even with an overmodulated signal, and to detect and suppress the occurrence of white noise due to FM demodulation failure with a relatively simple algorithm. be able to.

  According to the present invention, it is possible to accurately detect multipath interference even with an overmodulated signal, and to detect and suppress the occurrence of white noise due to FM demodulation failure with a relatively simple algorithm. be able to.

It is a block diagram of the FM radio receiver which concerns on one Embodiment of this invention. It is a figure which shows an example of a detection signal spectrum in case FM demodulation failure has not occurred. It is a figure which shows an example of the detection signal spectrum at the time of FM demodulation failure having generate | occur | produced. It is a figure which shows the structure of a secondary IIR type peaking filter. It is a figure which shows the frequency response of each peaking filter which implement | achieves bandpass filters HL21, HC22, and HH23. It is a figure which shows an example of a noise suppression parameter | index and noise suppression control. It is a figure which shows the relationship between a noise suppression parameter | index and an output gain. It is a figure which shows an example of noise likelihood detection and noise suppression. It is a figure which shows an example of a detection signal spectrum in case burst noise exists. It is a figure which shows an example of a detection signal spectrum in case white noise exists uniformly shallowly.

Hereinafter, embodiments according to the present invention will be described with reference to the drawings. In each drawing referred to in the following description, the same parts as those in the other drawings are denoted by the same reference numerals.
(Configuration of FM radio receiver)
FIG. 1 is a diagram illustrating a configuration example of an FM radio receiving apparatus according to the present embodiment. An FM radio receiving apparatus 1 shown in FIG. 1 is a receiving apparatus capable of receiving FM radio and outputting in stereo, and adopts a sum / difference method in stereo demodulation. In the present embodiment, the case where the FM radio receiving apparatus 1 is a digital system will be described, but an analog system may be used.

The FM radio receiving apparatus 1 includes an antenna 10, a low noise amplifier 11, an RF (Radio Frequency) front end unit 12, an analog-digital converter (A / D converter) 13, and an FM radio demodulation system 2. . The FM radio demodulation system 2 includes an FM detector 14, a noise generation detection unit 15, and a stereo demodulation unit 16.
The low noise amplifier 11 amplifies a weak signal received by the antenna 10. The RF front end unit 12 is not required for a mixer process for down-converting a signal of a specific frequency to an intermediate frequency by convolving a sine wave with the signal amplified by the low noise amplifier 11 and a band-pass filter such as a ceramic filter. An IF (Intermediate Frequency) signal is output through a process that removes a significant frequency component.

The A / D converter 13 converts the IF signal output from the RF front end unit 12 from analog to digital. The FM detector 14 performs a detection (demodulation) process on the digitally converted signal.
By the way, the detection signal detected by the FM detector 14 has a spectrum as shown in FIG. 9 when burst noise is instantaneously generated due to typical multipath interference, while the noise is continuously generated. Has a spectrum as shown in FIG.

The spectrum of the detection signal at the moment when there is no FM demodulation failure and at a certain moment are shown in detail as shown in FIGS. In each case, the vertical axis represents the signal magnitude, and the horizontal axis represents the signal frequency. The frequency of the pilot signal is 19 kHz, and the frequency of the difference signal is 38 kHz.
Comparing the spectra of FIG. 2 and FIG. 3, the latter has a lower energy peak at 19 kHz of the pilot signal component than the former, while the energy in the vicinity thereof (for example, around 18-20 kHz) has increased.

  That is, as the FM demodulation disturbance becomes stronger, the spectrum of the detection signal approaches a uniform distribution with no frequency deviation. Although the degree of disturbance varies, it is preferable to detect even slight disturbance with high sensitivity. Therefore, in the FM radio receiving apparatus according to the present embodiment, as a detection method, noise detection is detected by capturing the phenomenon that the energy of the 19 kHz pilot signal included in the detection signal decreases and the energy in the vicinity increases. For this purpose, a noise generation detection unit 15 is provided.

(Configuration of noise generation detection unit 15)
The noise generation detection unit 15 includes pilot signal energy (signal strength) PC, signal strength of a low-frequency sideband signal that is a signal included in a frequency band lower than the pilot signal (hereinafter referred to as “low-frequency side vicinity energy”). Each value of the signal strength (hereinafter referred to as “high-frequency side vicinity energy”) PH of a high-frequency sideband signal that is a signal included in a higher frequency band than the pilot and pilot signals is detected. Then, based on the value of the ratio of the detected pilot signal energy PC and the sum of the detected signal strength of the high frequency sideband signal strength and the signal strength of the low frequency sideband signal strength, the stereo demodulator 16 is Control.

Specifically, noise likelihood (PL + PH) / PC is obtained, and noise suppression is performed according to the value. As an example, when the noise likelihood becomes a certain threshold value or more, there is a case where noise suppression is realized by sequentially increasing forced monauralization, high-frequency suppression, and volume reduction according to the excess amount.
The noise generation detection unit 15 includes a pilot signal intensity detection unit (HC22) that detects the energy of the pilot signal, a low-frequency sideband signal intensity detection unit (HL21) that detects low-frequency side vicinity energy, and a high-frequency side vicinity energy. A high-frequency sideband signal intensity detection unit (HH23) for detecting the energy and an energy comparison unit 24. In the present embodiment, as shown in FIG. 1, the noise occurrence detection unit 15 has narrow band-pass filters HL21 and HC22 having a center frequency of 18 to 18.5 kHz, 19 kHz, and 19.5 to 20 kHz, respectively. , HH23 and an energy comparison unit 24. These band-pass filters HL21, HC22, and HH23 accept input of detection signals output from the FM detector 14. In addition, signals obtained by passing through the bandpass filters HL21, HC22, and HH23 are input to the energy comparison unit 24.

(Band pass filters HL21, HC22, HH23)
The bandpass filters HL21, HC22, and HH23 included in the noise generation detection unit 15 can be realized by a secondary IIR filter as shown in FIG. The transfer function of the second-order IIR filter shown in FIG. 4 is expressed by the following (Equation 1).

  More specifically, the bandpass filters HL21, HC22, and HH23 are realized by secondary IIR filters called three peaking filters having a center frequency around 19 kHz and the periphery thereof. Further, as described above, in the present embodiment, the center frequencies of the peaking filters are respectively set to 18.25 kHz, 19 kHz, and 19.75 kHz from the bottom. Each peaking filter is a filter having frequency characteristics as shown in FIG.

(Energy comparison unit 24)
The energy comparison unit 24 uses the values of energy PL, PC, and PH obtained by squaring each signal obtained by passing through the bandpass filters HL21, HC22, and HH23, and r = (PL + PH) / PC Is calculated. Then, the calculation result is output to the stereo demodulation unit 16 as a noise likelihood signal.
The energy PL, PC, and PH may be energy values obtained by squaring each signal obtained by passing through the bandpass filters HL21, HC22, and HH23 and performing primary smoothing. Here, the primary smoothing is a process of calculating the values of energy PL, PC, and PH used for calculating the noise likelihood based on the acquired output signal of the filter and the output signal of the filter for the past. This primary smoothing can suppress false detection due to sudden fluctuations in the energy output value.

  The stereo demodulator 16 outputs a stereo signal by adding or subtracting the difference signal to the monaural signal output as a detection signal from the FM detector 14. Further, the stereo demodulator 16 has a noise suppression control demodulator 31. When performing stereo demodulation, the noise suppression control demodulator 31 controls separation, high frequency suppression, and volume based on the noise likelihood signal calculated by the energy comparison unit 24 to suppress noise components, Output audio output signal.

  Here, when comparing the FM radio receiving apparatus according to the present embodiment with the FM radio receiving apparatus disclosed in Patent Document 1, in Patent Document 1, the low-frequency vicinity energy PL of the pilot signal is simply output in noise detection. As another general method, there is a method of outputting 1 / PC using the signal level of the pilot signal as an index. However, each method does not take a ratio between the signal strength (energy PC) of the pilot signal and the signal strength (energy PL, PH) of the sideband signal that is a signal in the vicinity frequency band of the pilot signal. Is low and is affected by fluctuations in signal level.

On the other hand, in the FM radio receiving apparatus according to the present embodiment, the energy comparison unit 24 takes the ratio of the energy PC and the energy PL and PH, so that the noise detection accuracy is higher than that of the conventional noise detection method. In addition, the noise suppression performance is improved because it is not easily affected by fluctuations in the signal level.
Note that, as a simplified method of calculating the noise likelihood r described above, a method using (PL + PH) * G-PC (G is a predetermined constant) is also considered as an equivalent threshold determination. Also, PL / PC or PH / PC may be used. Although the accuracy is slightly lower than the method of calculating (PL + PH) / PC, the number of filters can be reduced by one, so that the configuration becomes simpler. Even if these are applied, the performance of the FM radio receiver according to the present embodiment is not impaired, and performance comparable to that can be expected.

(Configuration of stereo demodulator 16)
(Noise suppression control type demodulator 31)
The stereo demodulation unit 16 stereo-demodulates the detection signal output from the FM detector 14. In the present embodiment, the stereo demodulator 16 adds and subtracts the L + R monaural signal of less than 19 kHz and the LR difference signal obtained by frequency shifting the signal spreading up and down centered on 38 kHz to obtain the difference between L and R. Obtain a stereo audio output signal.

  By the way, when there is a lot of noise due to FM demodulation failure, that is, when the value of the noise likelihood r is large, “separation control” that intentionally reduces the amount of LR difference signal to be added or subtracted, high frequency components with large noise on hearing Noise can be suppressed by applying “high-frequency suppression control” for reducing the sound volume, “volume control” for limiting the sound volume, and the like. In the present embodiment, the noise suppression control demodulator 31 applies these noise suppression controls stepwise. Hereinafter, specific examples of each noise suppression control will be described.

(Separation control)
In the separation control, the L + R monaural signal and the LR difference signal are input, and the stereo audio outputs Lout and Rout are derived by the following equations using the separation parameter α.
Lout = {(L + R) + α (LR)} / 2
Rout = {(L + R) -α (LR)} / 2
At this time, when FM demodulation is good, α = 1 and Lout = L and Rout = R are completely separated from the signal. However, when there is a lot of noise and the FM demodulation is bad, the difference signal of the above equation is used. Since the white noise component increases with the addition / subtraction of, by setting α = 0, Lout = Rout = (L + R) / 2 (that is, a complete monaural signal), and noise is reduced. Note that whether the FM demodulation state is good or bad may be determined based on, for example, whether or not the value of the noise likelihood r calculated by the energy comparison unit 24 is greater than or equal to a predetermined value.

(High frequency suppression control)
In high-frequency suppression control, noise is generally more perceptible at higher frequencies. Therefore, when a FIR (Finite Impulse Response) type low-pass filter with a cutoff frequency of about 10 kHz is applied, abnormal noise accompanying the on / off of the low-pass filter is applied. Listening noise can be reduced while suppressing the occurrence of noise.

(Volume control)
In volume control, noise in the sense of hearing is reduced by suppressing the volume of the noise generation part. Specifically, the volume is reduced by multiplying the stereo audio signal demodulated based on the noise likelihood by a gain. The gain g decreases with increasing noise likelihood r starting from 1 at normal time. Specifically, Lout and Rout are derived by the following equations.
Lout = g × {(L + R) + α (LR)} / 2
Rout = g × {(L + R) −α (LR)} / 2
Further, when a lower limit value of 0 to 0.5 is provided for the gain g, it is easy to obtain a sense of stability of the output sound.
When integrating the above three noise suppression control functions, for example, the noise suppression index d is obtained from the noise likelihood r by the following formula, and the noise suppression index d is smoothed in the time direction as necessary.
d = (r−r _th ) ^p

FIG. 6 shows noise suppression control to be executed based on the value of the noise suppression index d when p = 0.25. Depending on the value of the noise suppression index d after smoothing, FIG. 6 to determine the control to be executed. In FIG. 6, it is determined that separation control, high-frequency suppression control, and volume control are sequentially executed as the value of the noise suppression index d increases.
An example of the relationship between the noise suppression index d and the output gain g is shown in FIG. In FIG. 7, the vertical axis is the output gain g, and the horizontal axis is the noise suppression index d.

(Example)
FIG. 8 shows an example of an audio output waveform and its spectrum when the above noise suppression processing is actually applied. FIG. 8A shows the spectrum of the detection signal output from the FM detector 14. FIG. 8 (2) shows a spectrum when the detection signal of FIG. 8 (1) is stereo demodulated without performing noise suppression processing. FIG. 8 (3) shows a spectrum when stereo demodulation is performed by performing noise suppression processing on the detection signal of FIG. 8 (1). In the example of FIG. 8, FM demodulation disturbances are generally weak, and noise with remarkable burstiness is generated in the arrow portion described as (a portion where burst noise is remarkable) in FIG. As shown in FIG. 8 (2), the noise suppression index d rises in a portion where significant noise is generated in the sense of hearing. When FIG. 8 (3) and FIG. 8 (2) are compared, the output gain It can be seen that burst noise is effectively suppressed by g and the like. The effectiveness has been confirmed by hearing experiments.

  In the RF front end unit 12 and the FM detector 14, noise suppression control is performed when there is a separate means for detecting an abrupt decrease in the energy of a desired wave, the influence of an interference wave, an FM demodulation failure due to an equalizer or the like. Information regarding the FM demodulation failure detected by these means may be added to the type demodulator 31 for comprehensive judgment. Thereby, further better noise suppression control performance can be expected.

  By the way, in the method of determining the noise likelihood by binary as in Patent Document 1, the sound is basically muted at the time of detecting the multipath interference, so that the bad influence on the erroneous determination is great. For example, although a certain level of effect can be expected for burst noise caused by multipath interference as shown in FIG. 9, FM demodulation is disturbed shallowly and temporally as shown in FIG. In situations where noise can occur, the sound is interrupted and becomes very difficult to hear.

  On the other hand, in the FM radio receiving apparatus according to the above-described embodiment, different noise suppression processes are applied in four stages according to the value of the noise suppression index d. Furthermore, noise suppression control is applied in the order of separation control, high-frequency suppression control, and volume control, so even if white noise that occurs uniformly shallowly exists, Can be reduced from being interrupted and becoming difficult to hear.

  The present invention is suitable for an FM radio receiving apparatus having a function of detecting an FM demodulation failure during reception and suppressing noise generation.

DESCRIPTION OF SYMBOLS 1 Radio receiver 2 Radio demodulation system 10 Antenna 11 Low noise amplifier 12 Front end part 13 Converter 14 Detector 15 Noise generation detection part 16 Stereo demodulation part 21 Band pass filter for low frequency sideband signals 22 Band pass filter for pilot signals 23 Band-pass filter for high-frequency sideband signal 24 Energy comparison unit 31 Noise suppression control type demodulator

Claims (2)

An FM detector for detecting a frequency-modulated (FM) received signal;
A stereo demodulator that stereo-demodulates the detection signal output from the FM detector;
A control unit that controls the stereo demodulation unit based on a detection signal output from the FM detector;
An FM radio demodulation system comprising:
The controller is
A pilot signal strength detector that detects a signal strength of a pilot signal included in the detection signal output from the FM detector;
A sideband signal strength detection unit that detects a signal strength of a sideband signal that is a signal in the vicinity frequency band of the pilot signal,
An FM radio demodulation system, wherein the stereo demodulation unit is controlled based on a ratio value between the detected signal strength of the pilot signal and the detected signal strength of the sideband signal. The sideband signal strength detector is
A high-frequency sideband signal strength detection unit that detects a signal strength of a high-frequency sideband signal that is a signal in the vicinity of the pilot signal and included in a frequency band higher than the pilot signal;
A low-frequency sideband signal strength detection unit that detects a signal strength of a low-frequency sideband signal that is a signal in the vicinity of the pilot signal and included in a frequency band lower than the pilot signal;
The controller is
Control the stereo demodulator based on the ratio of the detected signal strength of the pilot signal and the sum of the detected signal strength of the high frequency sideband signal strength and the signal strength of the low frequency sideband signal strength The FM radio demodulation system according to claim 1.

Images