12 sets the third target value and the third predetermined number.

Note that it is needless to say that the number of the stages in which the target value G and the predetermined number N are changed is not limited to three.

The storage part 13 stores values which the target value G can take (for example, the above first target value, second target value, and third target value) and values which the predetermined number N can take (for example, the above first predetermined number, second predetermined number, and third predetermined number). The storage part 13 is constituted by, for example, the above-described EEPROM.

FIG. 4 is a flowchart showing a procedure of a process of correcting a reference frequency of a reference signal in the radio device 1. The following will describe a frequency control process of a case where the target value and the predetermined number are changed in three stages as described above.

First, an error of a frequency will be described. The radio device 1 handles a frequency as high as 10 GHz. An error by 1 ppb relative to 10 GHz is 10 Hz. In a frequency modulation (FM) mode of armature radio, an error of approximately 10 Hz in the frequency due to fluctuation of a PPS signal has little impact. In addition, in a single side band (SSB) mode of armature radio, an error of the frequency is directly reflected in pitch of a sound, but an error of approximately 10 Hz has little impact. However, when the error increases to approximately 100 Hz, such an error has great impact. For example, in FT8 communication of armature radio, a shift in frequency as large as 100 Hz results in reception of a signal at a completely different channel. Thus, an error of the frequency needs to fall within a range of at most 2 or 3 ppb, preferably within a range of 1 ppb.

First, the radio device 1 is turned on, so that the control unit 11 is activated. Then, the changing part 12 reads out the first target value and the first predetermined number, from the storage part 13 and sets the first target value and the first predetermined number (step S1), as shown in FIG. 4. Here, as the first predetermined number, for example, 5 (equivalent to a gate time of 5 seconds) is set.

Subsequently, the deviation detection circuit 10 detects a deviation between a specified value of the reference frequency and an average value calculated by the averaging circuit 9 (step S2). Before the detection of the deviation, the counter 8 counts a pulse number of the reference signal from the reference oscillator 7, and the averaging circuit 9 calculates an average value of pulse numbers each for one cycle of a PPS signal by dividing, by the first predetermined number, a total of pulse numbers for a time period whose length is the first predetermined number N times the one cycle.

After the detection of the deviation, the changing part 12 determines whether the deviation falls below the first target value (step S3). Here, the first target value is set to, for example, 60 ppb. The theoretical error Ef (ppb) of the PPS signal is represented by the following equation, as described above.

E
⁢
f

=

150
/
Tg

Therefore, in the case of Tg=5 (sec), Ef=±30 ppb.

In a case where the changing part 12 determines, in the step S3, that the deviation falls below the first target value (YES), the changing part 12 reads out the second target value and the second predetermined number, from the storage part 13 and sets the second target value and the second predetermined number (step S4). Accordingly, the changing part 12 changes the first target value to the second target value and changes the first predetermined number to the second predetermined number. Here, as the second predetermined number, for example, 80 (equivalent to a gate time of 80 seconds) is set.

In a case where the changing part 12 determines, in the step S3, that the deviation does not fall below the first target value (NO), the process returns to the step S2.

Subsequently, the deviation detection circuit 10 detects a deviation between the specified value of the reference frequency and an average value calculated by the averaging circuit 9 (step S5). Before the detection of the deviation, the averaging circuit 9 calculates the average value of pulse numbers each for the one cycle in a time period whose length is the second predetermined number N times the one cycle. In the step S5, the deviation detection circuit 10 calculates a deviation between the average number and the specified value.

After the detection of the deviation, the changing part 12 determines whether the deviation falls below the second target value (step S6). Here, the second target value is, for example, 4 ppb. In the case of Tg=80 (sec), Ef=+2 ppb.

In a case where the changing part 12 determines, in the step S6, that the deviation falls below the second target value (YES), the changing part 12 reads out the third target value and the third predetermined number, from the storage part 13 and sets the third target value and the third predetermined number (step S7). Accordingly, the changing part 12 changes the second target value to the third target value and changes the second predetermined number to the third predetermined number. Here, as the third predetermined number, for example, 150 (equivalent to a gate time of 150 seconds) is set.

In a case where the changing part 12 determines, in the step 6, that the deviation does not fall below the second target value (NO), the process returns to the step S5.

Subsequently, the deviation detection circuit 10 detects a deviation between the specified value of the reference frequency and an average value calculated by the averaging circuit 9 (step S8). Before the detection of the deviation, the averaging circuit 9 calculates an average value of pulse numbers each for the one cycle in a time period whose length is the third predetermined number N times the one period. In the step S8, the deviation detection circuit 10 calculates a deviation between the average number and the specified value.

After the detection of the deviation, the changing part 12 determines whether the deviation falls below the third target value (step S9). Here, the third target value is set to, for example, 2 ppb. In the case of Tg=150 (sec), Ef=+1 ppb.

In a case where the changing part 12 determines, in the step S9, that the deviation falls below the third target value (YES), the process ends. In a case where the changing part 12 determines, in the step S9, that the deviation does not fall below the third target value (NO), the process returns to the step S8.

Through such a process, the reference frequency of the reference signal is controlled so that an error of the reference frequency falls within a predetermined range.

After the process has ended, a series of the frequency control processes carried out by the counter