JPH06289072A - Burst signal analyzer - Google Patents

Abstract

PURPOSE:To analyze a frequency in a short time by temporarily storing signal data exceeding one period of a burst signal in a waveform memory, and extracting only necessary data including data of a signal section from them. CONSTITUTION:A waveform memory 8 stores signal data D A/D-converted from an analog burst signal (a) of one signal period or longer by an A/D converter 5. The starting position of a signal section in one reading section stored in the memory 8 is obtained by comparing the data D stored in the memory 8 with a threshold value by a comparator 13. Data collecting section is set from the starting position of the signal section in the memory 8. The data collecting section is so set that the data D belonging to the signal section falls in all the section or the data D belonging to the section are always included at a predetermined ratio. The data D belonging to the data collecting section is read by a reader 11, and frequency-analyzed by an FFT 14. Accordingly, the data D input to the FFT 14 are always extracted from a predetermined section including the signal section of the signal (a).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a burst signal analyzing apparatus for efficiently and accurately measuring frequency characteristics of a signal section in a burst signal including a signal section and a non-signal section within one signal cycle.

[0002]

2. Description of the Related Art For example, in JDC (Japan Digital Car Phone), ADC (US Digital Car Phone), GSM (European Digital Car Phone), JDCT (Japan Digital Cordless Phone), etc., in order to effectively use the communication line. In addition, a TDMA (time division multiple access) method is adopted as a communication method.

In such a TDMA system, one signal period T ₀ as shown in FIG. 8 is provided between the master station and each slave station.
The burst signal a in which the signal section T ₁ and the non-signal section (T ₀ −T ₁ ) are incorporated is transmitted and received. Generally, the occupied bandwidth B _W of a signal used for communication is required to be within an allowable range so as not to adversely affect adjacent channels. As shown in FIG. 10, the occupied bandwidth B _W is a frequency band in which the total value of the spectrum (power) of each frequency in the frequency analysis of the corresponding signal is 99% of the total value of the spectrum over the entire frequency band. Defined as width.

A general method for measuring the occupied bandwidth B _W in the case where the signal under measurement is a general continuous wave that is not a burst signal is as follows. The frequency characteristic distribution shown in is obtained, and the occupied bandwidth B _W is calculated from this frequency distribution characteristic. When the signal under measurement is the burst signal a as shown in FIG. 8, the signal section detection circuit 1 detects the signal section T ₁ of the input burst signal a as shown in FIG. Then, the signal section detection circuit 1
Is a control signal b which is “ON” when the signal is in the signal section T _{1 and} is “OFF” when there is no signal section (T ₀ −T ₁ ).
Is applied to the spectrum analyzer 2. The burst signal a is input to the spectrum analyzer 2. In this spectrum analyzer 2, the control signal b is "O.
Frequency sweeping is performed during the period “N” to perform frequency analysis on the signal in the signal section T ₁ of the burst signal a.

The analysis result of the spectrum analyzer 2 is transmitted to the next occupied bandwidth calculation unit 3. The occupied bandwidth calculation unit 3 calculates the occupied bandwidth B _W of the measured burst signal a according to the above-described calculation method.

[0006]

However, even in the burst signal analyzing apparatus in which the analog spectrum analyzer 2 as described above is incorporated, there are still the following problems to be improved. When the cycle T ₀ of the burst signal a is, for example, 20 ms, the signal section T ₁ and the non-signal section (T ₀ −T ₁ ) exist within this cycle T ₀ (20 ms). In this case, the frequency component for one wave is measured in the signal section T ₁ while avoiding the non-signal section (T ₀ -T ₁ ).
For example, when the signal section detection circuit 1 does not exist, the rising timing of the signal section of the burst signal is asynchronous with respect to the measuring device. Therefore, when the frequency component for one wave is always measured in the signal section, 20 ms is required for one measurement.

On the other hand, as shown in FIG. 10, the frequency is set on the horizontal axis and the spectrum (power) is set on the vertical axis.
Considering the case where the analysis result is displayed on the display screen of the CRT display device, the frequency axis is, for example, 500 points, that is, 50 points.
Zero wave resolution is required. Therefore, cycle 20
It is necessary to repeat the measurement of one wave per ms 500 times by changing the frequency component.

Therefore, in order to obtain one frequency analysis result shown in FIG. 10, a measurement time of 20 (ms) × 500 (wave) = 10 s (second) is required. Specifically, the frequency sweep speed is slowed to perform the frequency sweep over a plurality of signal sections T ₁ . As a result, there is a problem that the time required to obtain one frequency analysis result increases significantly.

The present invention has been made in view of the above circumstances, and by temporarily storing in a waveform memory signal data that exceeds at least one cycle of a burst signal, and extracting only the necessary data from it. An object of the present invention is to provide a burst signal analysis device that can perform frequency analysis processing even in the case of a burst signal in a short time and can significantly improve the signal analysis processing efficiency.

[0010]

In order to solve the above-mentioned problems, in the burst signal analyzer of the present invention,
An A / D converter that converts an analog burst signal in which a signal section and a non-signal section are incorporated into one signal cycle into a digital signal at a predetermined sampling cycle, and in synchronization with the sampling clock from the A / D converter. By comparing each data stored in the waveform memory with a predetermined threshold value, a waveform memory that captures and stores each data that is sequentially output over a reading period that is wider than at least one signal period Of the signal section start position detecting means for detecting the signal section start position and the signal section start position detected by the signal start position detecting means, and each data belonging to the signal section is included in the whole section or at a constant rate. Data sampling section setting means for setting the sampling section and data reading for reading each data of the data sampling section set by the data sampling section setting means from the waveform memory And the step, in which a FFT processing means for outputting the frequency characteristic in the signal interval of the read is a burst signal by performing FFT processing on each data by the data reading means.

[0011]

In the burst signal analyzing apparatus configured as described above, the waveform memory stores digitized signal data (sampling data) for one signal period or more, which is composed of the signal section and the non-signal section of the burst signal. Remembered.
Then, the signal section start position in the stored one reading section can be easily obtained by comparing each data stored in the waveform memory with the threshold value. When the signal section start position on the waveform memory is obtained, the data sampling section is set from this signal section start position. This data collection section is set so that data belonging to the signal section is entered in all sections, or data belonging to the signal section is always entered at a constant rate.

Then, each data belonging to the data sampling section in the waveform memory is read and frequency-analyzed by FFT (Fast Fourier Transform) means. Therefore, each data input to the FFT means is always extracted from a constant section including the signal section on the burst signal.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of the burst signal analysis device of the embodiment. FIG. 3 input from the input terminal 4
In one signal period T ₀ shown in, the signal section T ₁ and the non-signal section (T
_The analog burst signal a incorporating _0- T ₁ ) is input to the A / D converter 5. The burst signal a input from the input terminal 4 has already been frequency-converted to the intermediate frequency in the preceding stage.

When the measurement start signal b is applied from the operation panel 7, the A / D converter 5 samples the input analog burst signal a in synchronization with the clock of the clock signal c input from the outside. And convert it into a digital signal. Since the frequency (sampling frequency) f _S of the clock signal c needs to be measured at a frequency that is at least twice the occupied bandwidth B _W of the signal in the signal section T ₁ of the burst signal a that is the signal under measurement, Expected occupied bandwidth B _W
However, it is set so that the following condition is satisfied.

F _S > 4 · B _W and FFT for N _M data D on the time axis
The frequency resolution when performing the (fast Fourier transform) process is (f _S / N _M ). Each data D output from the A / D converter 5 in synchronization with the sampling clock of the clock signal c is applied to the input terminal of the next waveform memory 8. The operation of writing / reading data to / from the waveform memory 8 is performed by the write controller 10 and the read controller 1 in the memory controller 9.
It is carried out by 1. The memory control unit 9 is composed of a plurality of tasks (processing units) constructed on a computer program.

Next, referring to FIG. 3, the signal period T ₀ , the signal section T ₁ , the no signal section (T ₀ -T ₁ ), the reading section T _R ,
The mutual relation of the data sampling section T _M will be described. In addition, T represents an actual time (section), N represents the number of sampling data when the burst signal a is sampled in synchronization with the clock signal c, and the reading section T _R is the data included in the section. N _R is set to a value slightly larger than the number of data N ₀ corresponding to the signal period T ₀ . Since the start position of the reading section T _R is determined by the timing when the measurement start signal b is input, it is asynchronous with the start position of the signal period T ₀ of the burst signal a.

On the other hand, the data sampling section T _M is the signal section T
_It is set at a position including ₁ as the center, and includes the number N _M of data obtained by adding N ₂ data at both ends of the data number N ₁ corresponding to the signal section T ₁ . That is, in the data sampling section T _M , the total number of data N ₁ included in the signal section T ₁ and the (2N ₂ ) number of data included in the non-signal section (T ₀ −T ₁ ) are added. Several N _M of data are included.

N _M = N ₁ + 2N ₂ The starting position of the data sampling section T _M always has a fixed relationship with the starting position of the signal section T ₁ in the burst signal a. The writing control unit 10 includes an operation panel 7
From a state where the data number N _R of the reading section T _R is set, the measurement start instruction b is inputted in synchronization with the clock signal c, and data D applied to the input terminal of waveform memory 8 Write address WA in order from the start address R _AO
Specify and write in order. Then, the reading section T
_{When the} data D is written by the number N of _R data, the write operation is stopped. When the write operation is completed, the write controller 10 sends a write end signal d to the read start position determiner 12.

The read control unit 11 includes a read start position determination unit 1
When the read address R _A is designated from 2, the waveform memory 8
The data D of the corresponding address R _{A in} the
Send to. Further, when the start address R _AS is input from the read start position determination unit 12, the read control unit 11 reads each data D stored at each address after the start address R _AS in synchronization with the clock signal c, FFT unit 14
Send to.

When the start address R _AS is input from the read start position determining unit 12, the counting unit 15 outputs N _M pieces of data D included in the data sampling section T _M set on the operation panel 7. When the counting is completed, a data collection end command is sent to the FFT unit 14. When the start address R _AS is input from the read start position determining unit 12, the FFT unit 14 fetches each data D output from the waveform memory 8 until a data collection end command is input. As a result, the data D is acquired by the number N _M corresponding to the data collection section T _M.
Then, the FFT unit 14 performs FFT (fast Fourier calculation) processing on the acquired N _M pieces of data D to obtain the frequency characteristic as shown in FIG. 10 described above, and calculates the next occupied bandwidth. It is sent to the section 16. The occupied bandwidth calculation unit 16 calculates the occupied bandwidth B _W of the signal included in the signal section T ₁ in the burst signal a, which is the signal under measurement, by the method described above.

Further, the comparison unit 13 of the memory control unit 9 compares each data D read from the waveform memory 8 with the threshold value D _SH stored in the threshold value memory 17 to determine the comparison result as the read start position. It is sent to the unit 12. Next, the operations of the read start position determination unit 12, the comparison unit 13, and the read control unit 11 in the memory control unit 9 will be described with reference to the flowchart shown in FIG.

When the write end signal d is input from the write control unit 11 to the read start position determining unit 12, this flow chart starts and the read address R _{A is set} to the start address R of the waveform memory 8.
_It is set to _AO and one piece of data D stored in the corresponding address _RA is read (P2). The read data D value is compared with the threshold value D _{SH of the} threshold value memory 17 (P3). If the data D value is less than the threshold value D _SH , the corresponding position is a no signal section (T ₀ -T ₁ ). Then, the read address R _A is updated (R _A = R _A +1) (P4) and the process returns to P2.

In P4, if the data D value is equal to or greater than the threshold value D _SH , the corresponding position is in the signal section T ₁ , so it is confirmed that the current read address R _A is not the start address R _AO (P5 ), adds the data N _M corresponding to data sampling interval T _M to the read address R _a (P6). Then, the data D of the read address _RA after addition is read (P7) and compared with the threshold value D _SH (P8). Data D value is a consequential value D _SH
If it is less than this, it is determined that the data sampling section T _M includes all the data D in the signal section T ₁ as shown in FIG.

Even in the signal section T ₁ , the sampling data value may be less than or equal to the threshold value D _SH depending on the signal waveform. Therefore, in P8, for example, a plurality of data are consecutively less than or equal to the threshold value. Is a condition. Next, the start address R _AS of this data collection section T _M is _calculated by the following equation (P9). _{R AS = R A -N M -N} 2 i.e., signal the start position of the data sampling interval T _M interval T
Set N ₂ pieces from the start position of _{1 to} the front position. Then, the determined start address R _AS is sent to the read control unit 11 and the counting unit 15 (P10).

If the current read address R _A is equal to the start address R _AO at P5, it is determined that the start position of the read section T _R is in the signal section T ₁ , and the read address R _A is updated. ( _RA = _RA + 1) (P11), and the data D at the updated read address _RA is read. And
The result is compared with the threshold value D _SH (P13). If the data D value is not less than the threshold value D _SH , the process returns to P11 again and the read address R _A is updated.

When the data D falls below the threshold value D _SH at P13, it is determined that the corresponding position has entered the no-signal section,
Return to P2. According to the burst signal analysis device configured as described above, the asynchronous burst signal a as shown in FIG.
Is input, N _R pieces of data D in the reading section T _R set at an arbitrary position determined by the application timing of the measurement start signal b in the burst signal a are stored in the waveform memory 8. Then, by going compared to batchwise D _SH value of the data D in the order of N _R stored in the waveform memory 8, and it determines the data sampling interval T _M. And FF
The T section 14 section obtains a frequency characteristic using the N _M pieces of data D in the data sampling section T _M , and the occupied bandwidth computing section 1
The occupied bandwidth B _W is calculated in 6.

In this way, the waveform data for one signal period of the burst signal a is temporarily stored in the waveform memory 8,
The FFT processing is performed on the data in the data sampling section including the signal section from the stored data. Therefore, even if the signal is not a continuous wave signal but an asynchronous burst signal, the frequency characteristic is always obtained from the same portion of the signal waveform, and the accuracy of the measurement result is improved.

Since only the signal for one signal period T ₀ is used to obtain the frequency characteristic shown in FIG. 10, the speed of measuring the frequency characteristic is higher than that of the conventional apparatus using the analog spectrum analyzer 2. Can be greatly improved. Further, by setting the data sampling section T _M to include the entire signal section T ₁ as in the embodiment apparatus, in addition to the evaluation of the influence on the occupied bandwidth B _W and the adjacent channel,
Since it is not necessary to perform amplitude suppression at both ends of the signal using the window function, the signal section T ₁ in the burst signal a
A more complete spectrum distribution of is obtained. In addition, accurate power can be obtained for all signals included in the signal section T ₁ .

FIG. 6 is a block diagram showing the schematic arrangement of a burst signal analyzing apparatus according to another embodiment of the present invention. The same parts as those in the embodiment apparatus of FIG. 1 are designated by the same reference numerals.
Therefore, detailed description of the overlapping portions is omitted. First, the data sampling section T _M1 set in the apparatus of this embodiment will be described with reference to the waveform diagram of FIG. As shown in the figure, the data sampling section T _M1 is set so as to extract a part of the signal in the signal section T ₁ . That is, the N _M1 pieces of all data D included in the data sampling section T _M1 are composed of only the data D of the signal section T ₁ . Specifically, it is composed of N _M1 pieces of data excluding predetermined N ₃ pieces of data D at both ends of the N ₁ pieces of data in the signal section T ₁ .

In the apparatus of this embodiment, as shown in FIG.
As shown in, a window function multiplication unit 18 is arranged between the waveform memory 8 and the FFT unit 14. This window function multiplication unit 1
The window function W (t) set to 8 has a time function waveform having the characteristic shown in FIG. 5, for example. That is, with respect to the data sampling section T _M1 set by the memory control unit 9, the Hanning time window function has a small value near the start position and the end position of the section and a large value at the central position.

For example, as shown in FIG. 4, when frequency analysis is performed by cutting out a part of a continuous signal using, for example, a time gate or the like, the end face of the cut out section is vertical, and therefore this end portion is vertical. Contains high frequency components.
Therefore, in order to reduce the influence of the high frequency component due to this clipping, each data is multiplied by the window function W (t) having the time characteristic shown in FIG. It is possible to remove frequency components that are not originally included from the frequency characteristics after the FFT processing at 14.

Then, the window function W (t) shown in FIG. 5 is multiplied to each of the N _M1 data D in the data sampling section T _M1 output from the waveform memory 8. That is, the digital data having a mountain-shaped waveform in which the first and last data D are largely attenuated and the middle is hardly attenuated is input to the FFT unit 14. FIG. 7 is a flow chart showing the order of processing performed by the memory control unit 9 for setting the data sampling section T _{M1 in} which each data belonging to the signal section T ₁ shown in FIG. 4 is included in all the sections. The same parts as those in the flowchart of FIG. 2 are designated by the same reference numerals. Therefore, detailed description of the overlapping portions will be omitted, and only different portions will be described.

That is, when the write end signal d is input from the write control section 11 to the read start position determining section 12, this flow chart starts. Then, in P5, when it is confirmed that the current read address R _A is not the start address R _AO ,
It is determined that the corresponding position has reached the beginning of the signal section, and the read address _RA is the number of data N corresponding to the data sampling section T _M1.
M1 and adds a predetermined number N ₃ (P6a).

R _A = N _M1 + N ₃ Then, the data D stored at the read address R _A after addition is read (P7) and compared with the threshold value D _SH (P
8). If the data D is greater than or equal to the threshold value D, it is determined that the data sampling section T _M1 includes only the data D of the signal section T ₁ as shown in FIG. 4, and the start address R _AS of this data sampling section T _M1 is determined. It is calculated by the following formula (P9).

R _AS = _RA −N _M1 That is, the start position of the data sampling section T _M1 is the signal section T
_It is set to a position within N ₃ signal sections T ₁ from the start position of ₁ . Then, the determined start address R _AS is sent to the read control unit 11 and the counting unit 15 (P10). In the burst signal analyzing apparatus configured as described above, the waveform memory 8 temporarily stores the waveform data for one signal period of the burst signal a, and the data included in the signal section T ₁ from the stored data. The FFT process is performed on the data in the data sampling section configured by only D.

Therefore, similarly to the apparatus of the embodiment shown in FIG. 1, even if the signal is not a continuous wave signal but an asynchronous burst signal, the frequency characteristic is always obtained from the same portion of the signal waveform, and the measurement result is obtained. The accuracy of is improved. Also,
Since the data collection section is set using the window function and the amplitude of the collected data is corrected, even if the signal section T
_{When 1} is longer than necessary and it is not necessary to collect all the data included in the signal section T ₁ , by using this window function, an arbitrary number of data can be obtained without affecting the frequency characteristic. Frequency analysis can be performed.

[0037]

As described above, according to the burst signal analyzing apparatus of the present invention, it is necessary to temporarily store in the waveform memory the signal data of at least one cycle of the burst signal, and to include the data of the signal section from the waveform data. FFT processing is carried out by extracting only such data. Therefore, even with a burst signal, frequency analysis processing is performed in a short time, and the signal analysis processing efficiency can be greatly improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a burst signal analysis device according to an embodiment of the present invention.

FIG. 2 is a flowchart showing the operation of the apparatus of the embodiment.

FIG. 3 is a diagram showing a relationship between a burst signal waveform and a data sampling section in the apparatus of the embodiment.

FIG. 4 is a diagram showing a relationship between a burst signal waveform and a data sampling section in a burst signal analysis device according to another embodiment of the present invention.

FIG. 5 is a characteristic diagram showing a window function in the apparatus of the embodiment.

FIG. 6 is a block diagram showing a schematic configuration of the apparatus of the embodiment.

FIG. 7 is a flowchart showing the operation of the apparatus of the embodiment.

FIG. 8 is a general burst signal waveform diagram.

FIG. 9 is a schematic diagram showing a signal analysis device using a conventional analog spectrum analyzer.

FIG. 10 is a diagram showing the relationship between frequency characteristics and occupied bandwidth.

[Explanation of symbols]

5 ... A / D converter, 7 ... Operation panel, 9 ... Memory control unit, 10 ... Write control unit, 11 ... Read control unit, 12 ... Read start position determination unit, 13 ... Comparison unit, 14 ... FFT unit, 15
... Counting unit, 16 ... Occupied bandwidth calculating unit, 17 ... Threshold memory, 18 ... Window function multiplying unit.

Claims (1)

[Claims] 1. An A / D converter (5) for converting an analog burst signal in which a signal section and a non-signal section are incorporated into one signal cycle into a digital signal at a predetermined sampling cycle.
And a waveform memory for fetching and storing each data sequentially output from the A / D converter in synchronization with the sampling clock over at least a reading section wider than the one signal period.
(8), and a signal start position detection means (13, 17) for detecting the signal section start position in the reading section by comparing each data stored in the waveform memory with a predetermined threshold value, Using the signal section start position detected by this signal start position detection means, data collection section setting means (12) for setting a data collection section in which each data belonging to the signal section is included in all sections or at a constant rate And a data reading means (11) for reading each data of the data sampling section set by the data sampling section setting means from the waveform memory, and F for each data read by the data reading section.
A burst signal analysis device comprising: an FFT processing means (14) for executing an FT process to obtain a frequency characteristic of the burst signal in the signal section.