JPH0771048B2 - Frequency shift amount measuring method, its apparatus, and receiving apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention, when a series of signals having a predetermined frequency band are transmitted by shifting the frequency in the transmission process, the shift amount is changed on the receiving side. The present invention relates to a frequency shift amount measuring method and apparatus for measuring a shift amount received by an original series of signals based on the received signals so that the original series of signals can be correctly reproduced even if unknown.

Especially, SSB (Single Side Ban) whose carrier frequency is unknown
d) The present invention relates to a frequency shift amount measuring device capable of detecting the frequency of an original audio signal from an audio signal whose frequency is shifted by being demodulated from a signal with imperfect tuning.

From the above, it can be applied to a receiving device.

[Conventional technology]

Conventionally, a general receiving device 7 for receiving a radio wave propagated by the SSB method has been configured as shown in FIG. 9 (a). The incoming radio wave is converted into an intermediate frequency signal by the mixer 7b by the local oscillation signal from the local oscillator 7a, and
After only the required frequency component is selected by the IF circuit 7c, the detector 7d
Is detected by and converted into an audio signal. At this time,
Although the incoming radio wave is tuned by the frequency of the local oscillator signal (hereinafter referred to as the tuning frequency), this tuning frequency is manually set in the local oscillator 7a, and the carrier frequency of the transmitting side of the incoming radio wave (hereinafter referred to as the tuning frequency) is set. , The carrier frequency is the one on the transmitting side), so I heard the sound from the speaker 7g (to be referred to as voice and musical instruments, etc.) (to be described later) with my ears. The audio signal that is roughly tuned in this way and output from the detector 7d is frequency-shifted by the amount of tuning deviation, so in order to perform finer tuning, frequency demodulation is performed by the signal from the BFO (Beat Fre-guency Oscilator) 7e. Speaker 7g frequency-shifted by device 7f
Is output as sound. The adjustment of the frequency shift amount by the BFO 7e at this time was also performed while listening to the sound from the speaker 7g. In such a receiving device 7, as described above, the demodulated sound can be heard most naturally or the contents of communication can be heard clearly so as to be manually tuned.

On the other hand, as a method for measuring the carrier frequency of the SSB signal at the time of transmission on the receiving side, there has conventionally been the configuration shown in FIG. 9 (b). This is CCIR (International Advisory Committee on Radio Communications)
This is a device for measuring the carrier frequency by the offset method recommended by.

This offset method can be accurately measured when the incoming radio wave is a DSB (Double Side Band) signal and a residual carrier SSB signal that is propagated while the carrier signal remains. Next, the operation of the offset method when measuring these carrier frequencies will be described, and then the measurement of the carrier frequency of the suppressed carrier SSB signal having no carrier will be described.

First, the frequency from the offset signal generator 10 for frequency measurement is set to zero or fixed, and the detected audio signal (frequency F _o ) of the receiver 7 is the CRT11 of the oscilloscope.
Connected to drive the vertical axis of. Reference oscillator 12
Set the frequency from to the frequency (F _r ) of the audio signal you want to observe, and connect this to drive the horizontal axis of the CRT11. In this state, incoming radio waves (carrier frequency
The receiving device 7 is tuned so that F _X ) can be received. Next, the output frequency of the offset signal generator 10 is adjusted so that the Lissajous figure appearing on the screen of the CRT 11 of the oscilloscope is stationary. This output frequency is measured by the counter 9, and if this value is F _T , then F _o = F _r , F _X = F _T ± F _r , and the carrier frequency of the incoming radio wave F _X
Can be measured accurately.

The carrier frequency F _X of the suppressed carrier SSB signal was measured by stopping the observation of the Lissajous figure of the CRT 11 by the offset method shown in FIG. 9 (b), turning off the BFO 7e, and turning off the sound from the speaker 7g of the receiver 7. The output frequency (F _T ) of the offset signal generator 10 is adjusted so as to be the most natural when heard. The output frequency F _T at this time was used as the carrier frequency.

[Problems to be Solved by the Invention]

In the conventional receiving device as described above, it is difficult to obtain an accurate voice because the tuning work of the receiving device depends on the hearing of the operator. Similarly, it is difficult to measure the carrier frequency with high precision, and there is a problem in that the measured data has individual differences and reliability is poor.

In order to solve the above problems, the present invention first analyzes a transmitted signal, for example, a large number of transmitted voices, and pays attention to the following characteristics.

Although the frequency distribution of the voice depends on the vocalization condition, the voice of a man generally has a fundamental frequency of 100 Hz or higher and the voice of a woman has a harmonic overtone of an integral multiple of 200 Hz or higher. For example, when two types of voices are selected from a series of voices and spectrum analysis is performed in the frequency domain, a harmonic wave including each fundamental wave (hereinafter, referred to as a harmonic wave train) is observed. The two types of harmonic sequences of voice are arranged at different frequency intervals, but their frequency axes are the same and the frequency zero point is also common. The mutual frequency relationship between the harmonic sequences of these two types of voice does not change even if the frequency is shifted during the transmission process. Therefore, if the notch is added at each harmonic interval so as to extend the two types of frequency-shifted harmonic sequences on the frequency axis, there are spectral points where the two types of harmonic sequences are common,
This spectral point is the zero point of the original frequency. If the frequency at this point is measured, the amount of frequency-shifted shift can be directly used.

The present invention pays attention to the above points, and when a series of signals having a predetermined frequency band is transmitted as a transmission signal whose frequency is shifted by the transmission process, the transmission signal is received and the original series of signals is received. It is an object of the present invention to obtain a frequency shift amount measuring method and device capable of accurately measuring the frequency shift amount in the transmission process by detecting the frequency position of.

Furthermore, by using this frequency detection device as an SSB signal reception device, the original frequency of the audio signal that has been SSB-transmitted, frequency-shifted, and demodulated is detected, and the frequency difference before and after the frequency shift, that is, the frequency shift amount, is detected. Measuring is also an objective.

[Means for Solving the Problems]

As a method for achieving the above object in the present invention, a series of signals whose frequencies have been shifted by a transmission process are received, the spectrum is measured, and a plurality of harmonic sequences having different harmonic intervals are obtained from the spectrum. Select and calculate each harmonic interval, and then calculate a virtual spectrum to extend at each harmonic interval on the low frequency side of each of the multiple harmonic strings, and Among them, the frequency position of the common virtual spectrum is detected, and this is set as the frequency shift amount received by the series of signals.

Furthermore, an analysis unit that receives a series of audio signals whose frequencies have been shifted in the transmission process, measures and outputs the spectrum, and selects a harmonic train that includes the fundamental wave of multiple audio signals from the spectrum and The data pre-processing unit that calculates the harmonic intervals of each of the multiple audio signals and selects the representative spectrum that represents each of the multiple audio signals, and the new harmonics at the higher harmonic intervals on the lower frequency side than the typical spectra of each of the multiple audio signals. A virtual spectrum is calculated so as to form a wave train, a frequency position of the virtual spectrum common to a plurality of audio signals is detected, and a series of audio signals is output as an amount of frequency shift received.

Further, by providing the above frequency detecting device in the receiving device for receiving the SSB signal, it is possible to detect the frequency shift amount of the SSB demodulated audio signal and perform accurate tuning.

[Action]

In the frequency shift amount measuring method of the present invention, a series of signals whose frequencies have been shifted by a transmission process are received to measure the spectrum, and a plurality of high-frequency trains having different harmonic intervals are selected from the spectrum, and Calculates each harmonic interval, calculates a virtual spectrum so that it extends to the low frequency side of each of the multiple harmonic strings, and extends the harmonic spectrum at each harmonic interval. The frequency position of the virtual spectrum is detected, and this is taken as the frequency shift amount received by the series of signals.

In the frequency shift amount measuring device, the analysis unit receives a series of signals whose frequencies have been shifted by the transmission process, measures the spectrum thereof, and outputs the spectrum. Based on the spectrum, the data preprocessing unit selects multiple harmonic sequences with different harmonic intervals, calculates the respective harmonic intervals, and selects a representative spectrum that represents each of the multiple harmonic sequences. To do. The calculation unit calculates a virtual spectrum so that a new harmonic sequence is formed at each harmonic interval on the low frequency side of the representative spectrum of each of the multiple harmonic sequences, and the new harmonic sequence is common. The frequency position of the virtual spectrum is detected.

〔Example〕

 A frequency shift amount measuring method according to the present invention will be described.

FIG. 1 is a flow chart showing the main flow of the measurement method, and FIG. 2 is a diagram for explaining the main flow of FIG. 1, showing an example of a voice spectrum.

Here, an audio signal is taken as an example of a series of signals having the same frequency region, and it is assumed that this audio signal is frequency-shifted and transmitted in the transmission process. The amount of shift when the frequency of the voice signal before transmission is frequency-shifted can be roughly divided into the first amount in order to obtain it from the voice signal transmitted.
There are three levels as shown. This will be described below with reference to FIG. The detailed flow is the same as the operation flow (FIG. 4) of the invention of the frequency shift amount measuring device described later.

Step (10): The spectrum of the transmitted voice signal is analyzed, the result is stored each time the analysis is performed, and the result is sequentially output at a predetermined time interval.

Step (20): measuring the frequencies of the stored spectrum within the specified level range and at least two harmonic sequences having different harmonic spacings (ie two frequency-shifted audio signals, For the sake of convenience, this is named voice signal 1 and voice signal 2), and the respective harmonic intervals PITCH ₁ and PITCH ₂ are calculated. It is to be noted that one by one (one harmonic sequence at one of the predetermined time intervals) is selected from the spectra sequentially output from step (10) until the above-mentioned two objective harmonic sequences are obtained.

Here, the relationship between the selected high frequency and the desired shift amount f _SIFT will be described.

The spectra of the audio signals 1 and 2 are shown in FIGS. 2 (a) and 2 (b). Since the audio signal 1 and audio signal 2 undergo the same amount of frequency shift, the high frequency interval PITC obtained earlier
Based on H ₁ and PITCH ₂ , when a virtual spectrum is set up so as to extend the two high-frequency trains of FIGS. 2 (a) and 2 (b) to the low-frequency side, these two high-frequency trains are extended. The frequency position of the common virtual spectrum can be detected.
The frequency positions of this virtual spectrum are shown in FIGS. 2 (c) and 2 (d).

The frequency position of the common virtual spectrum is the point where the frequency of the original audio signal is zero. Therefore, the frequency itself of the common virtual spectrum is the shift amount f _SIFT obtained.

In FIGS. 2 (c) and 2 (d), F _S1 and F _S2 are reference points for extending the harmonic train, and in this case, as a representative spectrum serving as the reference point, FIG. a),
It is the frequency when the one showing the maximum level is selected from the spectra of the audio signals 1 and 2 of (b).

As in the case of the audio signals 1 and 2 shown in FIGS. 2 (a) and 2 (b), it may be a harmonic wave train having no unnecessary wave component (for example, when a known harmonic wave train is transmitted for a test of a transmission system). For example, even if the representative spectrum is not provided, a virtual spectrum may be set up so as to extend the measured harmonic sequence.

When there are unnecessary components other than harmonics in the spectrum, some measures have been taken to calculate the parameters PITCH ₁ , PITCH ₂ , F _S1 , and F _S2 . This will also be described later (explanation of FIG. 4).

Step (30): From the above description, the following equation holds based on each parameter obtained in step (20).

An integer of F _S1 −n × PITCH ₁ = F _S2 −m × PITCH ₂ n, m: n, m satisfying this formula is calculated.

Next, the value of F _S1 −n × PITCH ₁ is obtained, and this is the shift amount f
Output as _SIFT .

FIG. 3 is a diagram showing a configuration of an embodiment of the frequency shift amount measuring device according to the present invention, and FIG. 4 is a flow showing an operation of the device of FIG. 3, that is, a shift amount f _SIFT measuring flow. Further, although many detailed flows for achieving the main flow of FIG. 1 are conceivable, FIG. 4 is also a flowchart showing an example thereof.

In FIG. 3, reference numeral 1 is an analysis unit for analyzing an input signal, for example, a frequency component of an audio signal, which is an A / D converter 1a for sampling the audio signal and converting it into digital waveform data,
A wave memory 1b for storing the waveform data, and an FFT processing unit 1c for converting the waveform data into a frequency spectrum by FFT operation and storing the spectrum data.
Consists of. The FFT processing unit 1c outputs a series of spectral data each time it is stored. Reference numeral 2 is a data pre-processing unit, which is composed of the following 2a to 2c. Reference numeral 2a is a spectrum detector that detects the frequency and level of each spectrum from the spectrum data. Reference numeral 2b denotes a computing unit, which selects a plurality of harmonic sequences including fundamental waves of two audio signals, for example, audio signals 1 and 2 based on the data detected by the spectrum detection unit 2a, and selects a harmonic interval of each. (PITCH ₁ , PITCH ₂
Then, a representative spectrum representing each of them is selected (the frequencies are _defined as F _S1 and F _S2 ). Reference numeral 2c is a memory for storing the calculation result of the calculator 2b. Reference numeral 3 denotes an arithmetic unit, which calculates n, m that satisfies the equation in step (30) in FIG. ₁ to obtain the value of F _S1 −n × PITCH ₁ and shift amount
Output as f _SIFT . The arithmetic unit 2b and the arithmetic unit 3 have means for storing data of respective arithmetic processes. Reference numeral 4 is a monitor for displaying and monitoring the data of each unit after the FFT processing unit 1c by CRT. 5 is a set of measurements,
A panel for resetting and setting analysis conditions, and 6 is a control unit, which stores a program for scheduling and controlling each unit, and controls each unit according to an instruction from the panel 5.

Here, each circuit of the spectrum detector 2a, the calculator 2b, the calculator 3 and the controller 6 is composed of a CPU and a memory in this example.

In FIG. 4, steps (11) and (12) correspond to step (10) in FIG. 1 and show the operation of the analysis unit 1. Steps (21) to (28) correspond to step (20) in FIG. 1 and show the operation of the data preprocessing unit 2. Step (21) ~
(38) corresponds to step (30) in FIG. 1 and shows the operation of the calculation section 3.

Next, measurement of the shift amount f _SIFT according to this embodiment will be described with reference to FIGS. 3 and 4.

Steps (11) and (12): The operation of the analysis unit 1 is as described in the above configuration. However, the analyzable frequency band and frequency resolution of the analysis unit 1 are set to desired conditions from the panel 5. The frequency band of the analysis unit 1 should be set according to the expected frequency band of the frequency-shifted audio signal that is the input signal.

In any case, the frequency measurement limit is determined in the narrower band of the frequency band of the analysis unit 1 and the expected band of the voice signal. The frequency of this measurement limit is f _LIMIT
And

Steps (21) and (22): The spectrum detection unit 2a receives the spectrum data output from the analysis unit 1 according to the instruction from the computing unit 2b, and detects the maximum level L _MAX from the spectrum data. This spectral data contains the fundamental and harmonics of the audio signal.

Step (23): The spectrum detection unit 2a has preset a limit value α of the level measurement determined by the S / N ratio and the like, and the specified level range (α) based on α
And spectrum data are compared, and specified level range (α)
Detect the frequency of the spectrum that falls within. The frequency of this spectrum is stored in the memory 2C as f _PEAK (i).
Here, i is the number of the detected spectrum, and takes values of i = 1, 2, ..., N in order from the lowest frequency. N is the total number of spectra.

Step (24): If the conditions of L _MAX > α and i ≧ 2 are not satisfied, the harmonic interval cannot be detected in the next step, so restart from step (21).

Step (25): Details of the calculation process here will be described later on the basis of the flow of FIG. 5, the spectrum diagram of FIG. 6 and the calculation data of FIG. Here, only the main points will be described with reference to FIG.

First, the frequency of the spectrum is calculated from f _PEAK (i) to calculate the frequency interval between the spectra (hereinafter, referred to as spectrum interval: see P (1) to P (3) in FIG. 6). f _PEAK
Since (i) may include a spectrum that is not high frequency (f _PEAK (2)), the unnecessary spectrum interval (P (1), P (2)) is removed from the spectrum interval, and the error due to the measurement resolution is further removed. The average required harmonic interval PITCH is calculated in consideration of (P (3) + P (3) / 2).

In this calculation, if the harmonic interval PITCH cannot be obtained, that is, if the result in the flow of FIG. 5 results in FAIL, the process restarts from step (21).

Step (26): In this way, the harmonic intervals obtained for at least two harmonic trains are stored as PITCH ₁ and PITCH _2.
Remember in 2c. In this example, the target harmonic train is set to two because it is the minimum number for calculating the shift amount f _SIFT . In order to calculate the shift amount f _SIFT more accurately, it is better to increase the number of target harmonic trains.

Step (27): The representative spectrum f _S of the harmonic series is calculated from the spectrum frequency f _PEAK (1). But,
Since f _PEAK (i) may include a spectrum that does not form harmonics (for example, f _PEAK (2) in FIG. 6),
It is necessary to select a representative spectrum f _S from the spectra forming the harmonic series from the spectrum f _{PEAK (i)} .

FIG. 8 shows the flow of selecting this representative spectrum f _S. FIG. 8 shows f _PEAK (i) in which the spectral interval between both neighbors matches the harmonic interval PITCH obtained in step (25), and i
The smaller one is f _S. From this selection flow, for example, f _PEAK (4) in FIG. 6 becomes f _S.

In this way, the representative spectrum f ₂ obtained for each of the two harmonic sequences is _defined as f _S1 and f _S2 , and the harmonic intervals PITCH ₁ and PITCH ₂ obtained in step (26) are corresponded to the same harmonic sequence. Then, it is stored in the memory 2c.

Step (28): Check the conditions of f _S1 ≠ f _S2 and PITCH ₁ ≠ PITCH ₂ , and if the conditions are satisfied, proceed to the next step (31). If the conditions are not satisfied, restart from step (21).

Step (31): From here on, the operation of the calculation unit 3 is started. First, f _S1 , f _S2 , PITC stored in the above equation and the memory 2c
Set the following equation based on H ₁ and PITCH ₂ , and set n = as the initial condition.
Let m = 0.

F _a = F _S1 −n × PITCH ₁ F _b = F _S2 −m × PITCH ₂ Steps (32) to (37): These steps are representative spectra F _S1 and F _{S2 of} two harmonic sequences, respectively.
The virtual spectrum F _a and F _b are calculated by changing the harmonic orders n and m so as to extend the harmonic series to the lower frequency side, and the virtual spectrum that is common to the two extended harmonic series. Find (F _a = F _b ), and shift the frequency position (F _a = F _S1 −n × PITCH ₁ ) of this common virtual spectrum by the shift amount f
_SIFT .

In this flow, the step (32) to (34), when it is determined that F _a> F _b in step (32), by changing the n until F _a ≦ F _b the (F _b is fixed) F _a calculate. When F _a <F _b , or when F _a <F _b in the above calculation, in steps (32) to (37), change m until F _a ≧ F _b (F _a is fixed. ) Calculate F _b . Like this seesaw
In the process of calculating F _a and F _b , steps (35),
_Fa = _Fb is detected in (38). Steps (34), (37)
In that case, there is no point in calculating F _a and F _b beyond the f _LIMIT explained in steps (11) and (12). In such a case, restart the spectrum data from step (21). I will let you.

Next, details of the process of calculating the harmonic interval PITCH in step (25) will be described based on the flow of FIG. 5 and with reference to the spectrum diagram of FIG. 6 and the calculated data of FIG.

First, the symbols in FIG. 5 will be described.

i is the number of the spectrum detected as described above, and i =
1,2, ... N is also the total number of spectra.

j is a number assigned corresponding to the magnitude of the detected spectrum interval, and j = 1, 2, ... (<N).

P (j) is a spectrum interval corresponding to the number j. For example, the number j has the same value for the same spectrum interval as shown in FIG.

SUM (j) is the sum of the values of P (j) for each number j, that is, for each type of spectrum interval.

COUNT (j) indicates the number of P (j).

The flow will be described below with reference to FIG.

Steps (41), (42): Initial condition setting.

Steps (43) to (51): SUM (j) and COUNT (j) are calculated. Among them, in steps (43) to (47), a spectrum interval of the same kind as the spectrum interval ΔF with respect to the number i obtained in step (43) is searched for, and if there is a SUM ( j) and COUNT (j) are integrated, and if there is not, the spectral interval ΔF in step (49).
To SUM (j) and COUNT (j).

In steps (50) and (51), the above calculation is performed until i = N.

Step (52): From here to step (58) is a flow for obtaining the average harmonic interval PITCH. In this step, the initial setting for that is performed. The number NOMINAL is the value of j of the spectral interval P (j) with the largest number of detections.

Steps (53) to (58): The one showing the maximum value among COUNT (j) obtained in steps (43) to (51) (this is C
OUNT (NOMINAL)) and searched COUNT
The harmonic interval PITCH is calculated in step (58) based on (NOMINAL) and the corresponding SUM (NOMINAL).

Steps (57) to (59): When the value of COUNT (NOMINAL) is 2 or more, that is, when the largest spectrum interval is 2 or more, it is determined as FAIL (see step (25)).

FIG. 7 is an example of calculating the spectrum in FIG. 6 by the flow in FIG.

In the above flow, avoid erroneous decision by giving a margin of γ% when comparing frequencies such as F _a = F _b , and by adopting the operation that is considered true if the following equation is satisfied: You can Here, γ is a value that should be derived from the error range of the measurement system.

│F _a −F _b │ ≦ (γ / 100) F _{b In the} above analysis unit 1, FFT processing is performed for spectral analysis. However, since each voice signal is short in time, high speed is achieved. It is also possible to use analysis functions other than the FET processing as long as the spectrum of the input signal can be analyzed by.

Further, by providing the conventional receiving device 7 for SSB signals shown in FIG. 9 (a) with the device shown in FIG. 3 to provide a new receiving device, accurate reception can be performed. That is, the device of FIG. 3 receives an audio signal from the frequency demodulator 7f of FIG. 9 (a),
It is possible to measure the shift amount in which the audio signal is frequency-shifted due to the tuning shift. Accurate reception can be performed by tuning again with the local oscillator 7a and the BFO 7e based on the measured shift amount.

Furthermore, if the frequencies of the local oscillator 7a and the BFO 7e are accurately detected by a counter or the like, the carrier frequency of the SSB signal can also be accurately measured. This is effective in radio wave monitoring.

In the description of the above embodiment, since the audio signal contains a high frequency, the frequency shift amount is measured based on this audio signal. However, according to the present invention, two or more different signals are included in the signal containing a high frequency. If it is a series of signals that can capture the harmonic train,
The amount of frequency shift can be measured.

In addition, in order to accurately calculate the frequency of the spectrum of the input signal, which changes moment by moment, the method described in the following documents can also be used.

IEICE Transactions A, Vol J70-A No5, pp, 798-80
5, May, 1987 “High-precision frequency determination method using FFT” (Effect of the invention) As described above, the frequency shift amount measuring method according to the present invention has a series of frequencies that have been shifted in the transmission process. Receive the signal, measure its spectrum, select multiple harmonic trains with different harmonic intervals from the spectrum, calculate the harmonic intervals of each, and then add each of the multiple harmonic trains to the low-frequency side. The virtual spectrum is calculated so as to extend at the harmonic intervals, the frequency position of the common virtual spectrum in the extended multiple harmonic sequences is detected, and this is taken as the frequency shift amount received by the series of signals. And a frequency shift amount measuring device, an analyzer for receiving a series of signals whose frequencies have been shifted by a transmission process and measuring and outputting the spectrum thereof,
A data pre-processing unit that selects a plurality of harmonic sequences with different harmonic intervals based on the spectrum to calculate the respective harmonic intervals, and selects a representative spectrum that represents each of the plurality of harmonic sequences, A virtual spectrum is calculated so that a new harmonic sequence is formed on the lower frequency side of the representative spectrum of each of the multiple harmonic sequences at each harmonic interval, and the frequency of the virtual spectrum in which the new harmonic sequence is common. Since the calculation unit for detecting the position is provided, there is an effect that the reception side can measure the frequency shift amount of a series of signals whose frequencies have been shifted in the transmission process with high accuracy.

By applying the present invention to a receiving device for SSB signals, the receiving device can be tuned easily, quickly and with high accuracy, and in particular, it has a remarkable effect in monitoring the carrier frequency of radio waves.

[Brief description of drawings]

FIG. 1 is a flow chart showing an embodiment of the frequency shift amount measuring method according to the present invention, and FIGS. 2 (a) to 2 (d).
FIG. 4 is a diagram for explaining FIG. 1, a diagram showing examples of a voice spectrum and a virtual spectrum, FIG. 3 is a diagram showing a configuration of an embodiment of a frequency shift amount measuring device according to the present invention, and FIG. Is a flow chart showing the operation of the embodiment of FIG.
FIG. 5 is a detailed flowchart for calculating the harmonic interval in FIG. 4, FIG. 6 is a spectrum diagram for explaining FIG. 5, and FIG. 7 is a diagram showing an example of calculated data. FIG. 8 is a flow chart for selecting a representative spectrum, and FIGS. 9 (a) and 9 (b) are diagrams showing an example of a conventional SSB signal receiving apparatus and carrier frequency measuring apparatus. In the figure, 1 is an analysis unit, 1a is an A / D converter, 1b is a wave memory, 1c is an FFT processing unit, 2 is a data preprocessing unit, 2a is a spectrum detection unit, 2b is a computing unit, 2c is a memory, and 3 Is the arithmetic unit, 4
Is a monitor, 5 is a panel, 6 is a control unit, 7 is a conventional receiver, 7a is a local oscillator, 7b and 8 are mixers, 7c is an IF circuit, 7d is a detector, 7e is a BFO, and 7f is a frequency demodulator. , 7g is a speaker,
9 is a counter, 10 is an offset signal generator, 11 is a CRT, 1
2 is a reference oscillator.

Claims (3)

[Claims] 1. A frequency shift amount received by the series of signals when the series of signals having a predetermined frequency band is transmitted as a transmission signal whose frequency is shifted by a transmission process. A frequency shift amount measuring method for measuring, measuring the spectrum of the received transmission signal, selecting a plurality of harmonic sequences having different harmonic intervals from the spectrum, and calculating the respective harmonic intervals, Further, the virtual spectrum is calculated so as to be extended at the respective harmonic intervals on the low frequency side of each of the plurality of harmonic sequences, and the frequency position of the virtual spectrum common to the extended plurality of harmonic sequences is calculated. A frequency shift amount measuring method which detects and uses this as the frequency shift amount received by the series of signals. 2. An analysis unit (1) for receiving a series of audio signals whose frequencies have been shifted in the transmission process, measuring and outputting the spectrum, and a harmonic including fundamental waves of a plurality of audio signals from the spectrum. A data pre-processing unit (2) for selecting a wave train, calculating respective harmonic intervals, and selecting a representative spectrum representative of each of the plurality of audio signals, and the representative spectrum of each of the plurality of audio signals. A virtual spectrum is calculated so as to newly form a plurality of harmonic sequences at each of the harmonic intervals on the low frequency side, and the frequency position of the virtual spectrum common to the new plurality of harmonic sequences is detected. A frequency shift amount measuring device, comprising: an arithmetic unit (3) for outputting the frequency shift amount received by the series of audio signals. 3. A receiver for receiving an SSB signal and converting and varying the frequency of the SSB signal to demodulate the SSB signal to output an audio signal, in order to measure the frequency shift amount of the SSB demodulated audio signal. Claim (2)
A receiving device comprising the frequency shift amount measuring device according to.