JPH021073A - Method for reading multichannel waveform - Google Patents

Abstract

PURPOSE:To easily and promptly convert a multichannel waveform on a recording paper into multichannel data by reading the above-mentioned wave form by an optical sensor, storing the read waveform for each channel, and after that separately obtaining an analog signal by a prescribed operation. CONSTITUTION:The scanning light of an image sensor 3 crosses on a recording paper 1 in an (l) direction shown in a figure. At this time, a 1-bit output signal is obtained at a channel 1, and 1-bit through 1... 1-bit signals are obtained at a channel 3. The respective channel signals obtained in this manner are partitioned in units of 8 bits and parallel-converted in an S/P converting circuit 5 and successively written to RAMs71-7n by a DMA circuit 6. CPUs81-8n read data corresponding to respective self channels from the RAMs by the circuit 6. Data for zero calibration are previously recorded on the recording paper 1, and the respective CPUs detect zero positions corresponding to themselves and execute operations for obtaining the differences between the respective zero value data and respective actually measured data. Obtained results are analogized by D/A converting circuits 91-9n and respectively outputted from respective terminals 101-10n as the data.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a reading method for reading a multi-channel waveform, such as an electroencephalogram σ, recorded on recording paper.

[Conventional technology]

Detected drum shapes obtained from multiple points simultaneously, such as brain waves and vibrations, are recorded on recording paper by a multichannel pen recorder. In order to analyze the waveform χ recorded on this recording paper, it is necessary to convert it into an electrical signal.

Of course, there is also a method of directly recording the detected waveform on a magnetic tape using a multi-channel data recorder, but it is difficult to determine whether or not it is correctly recorded on the magnetic tape. Furthermore, since data recorders are relatively expensive, it is impossible to record on the data recorder each time. On the other hand, it is not uncommon for an experienced person to clarify the phenomenon of a waveform recorded on a recording paper by looking directly at the waveform f'L.

Therefore, when detecting and recording non-reproducible phenomena, pen recorders are often used in consideration of reliable recording and storage. Pen recorders are also used because of their cost advantages.

When a recorded waveform is analyzed using a waveform analyzer, a computer, etc., it has been done manually by reading the waveform on recording paper with a digitizer and converting it into an electrical signal. Furthermore, in an era when data recorders were not widely used, data was recorded using a pen recorder. Therefore, when analyzing waveforms of this old data, it was necessary to convert it into electrical signals.

[Problem to be solved by the invention]

As mentioned above, since the conversion to electrical signals is done manually, there is a problem in that the conversion cannot be done quickly. Also, its accuracy was not high.

The present invention has been made in view of the above points, and an object of the present invention is to provide a multi-channel waveform reading method that can quickly convert data waveforms on recording paper into electrical signals.

[Means for solving problems]

The above purpose is groove fi! i. The present invention is a method for independently obtaining a plurality of electrical signals corresponding to waveforms of a plurality of channels recorded in parallel on a recording sheet, the method comprising one-dimensional, two-dimensional or An individual optical sensor is arranged to face the recording sheet, and when the sensor is a one-dimensional or two-dimensional sensor, it is possible to read all of the waveforms of the plurality of channels in a direction intersecting the time axis of the waveform. arranging the sensor and electrically eroding the sensor in a direction crossing the time axis; if the sensor is an individual sensor, moving the individual sensor in a direction crossing the time axis; and recording the sensor. Either or both of the sheet and the sensor are moved continuously or intermittently in the direction of the time axis of the waveform T! obtaining read outputs of the waveforms of the plurality of channels by performing a waveform shaping on the readout outputs obtained from the sensor to form a digital signal; The method includes writing data into the memories of the plurality of channels, reading digital signals of a predetermined area to which the own channel belongs from the memory of the plurality of channels, and outputting data corresponding to the waveforms of the plurality of channels to Dokdo. The present invention relates to nine methods for reading multichannel waveforms characterized by the following.

[For production]

In the above invention, the sensor does not continuously output read outputs of the same channel, but sequentially outputs read outputs of multiple channels. Since the waveform signals of a plurality of channels are output in a predetermined order, it is possible to divide them into one section. The digital signals corresponding to the read outputs are preferably serial-parallel converted, and more preferably, addresses are assigned and written to memories for multiple channels, respectively! It will be done. All waveform data is written to the memory of each channel! fi, it is possible to extract the data of the own channel.

〔Example〕

FIG. 1 is an explanatory diagram showing one embodiment of the present invention.

In the figure, reference numeral 1 denotes a recording paper on which waveforms at a plurality of locations are simultaneously detected and recorded by a pen recorder, and is fed by a feeding device 2 in the direction of the arrow A at a constant speed. 3
The one-dimensional image sensor scans the recording paper 1 from end to end and optically (A/D conversion circuit) converts the output signal of the image sensor 3 into a digital signal.

5 is a serial/parallel conversion circuit (S/P conversion circuit), and 6 is a direct memory access circuit (DMA circuit). 71
．． 72.73...φ...7n is a random access memory (RAM) provided corresponding to the channel, 81
．． 82.83 ***-am 80 is a microcomputer (CPU) provided corresponding to the channel. 91.92.93Φ...φφφ9n is digital-to-analog conversion circuit (D/A conversion circuit), 101, +0
2, lO3...lOn are output terminals, respectively.

And RAM81.82.83...8n,
CPU91, 92, 93 e-***a9
A distributed processing arithmetic circuit 11 is formed by n and the like.

Here, the S/P conversion circuit 5 includes CPU81.82.83.
...8n is 8 bit word length, so 8 bits
This is a unit conversion. The DMA circuit 6 controls the timing of writing and reading digital signals into the RAM 71.72.73 mm-*-70 so that writing and reading do not overlap.

Figure 2 shows a recording paper l on which waveforms are recorded on each channel, channels chi, ch2, ch3, φ, .
Different waveforms in hn are data D1, O2, O3...
・Recorded as Dn. A reference signal S1 representing time is recorded in chm.

Next, the operation in the above configuration will be explained.

FIG. 3 shows the output signal of the image sensor 3, and S2 is a synchronization signal for scanning. S3 is a clock signal and consists of 4096 bits per scan. The scanning light from the image sensor 3 traverses the recording paper 1 as shown by the imaginary line in FIG. The presence or absence of a data line is determined by the scanning light and is obtained as an electrical signal at the output of the image sensor 3.

This electrical signal is synchronized with the clock signal S3 and has a waveform! ! 1
As shown in FIG. 3(C), the digital signal S4 is
obtained as. As shown in FIG. 3(D), when the scanning direction is perpendicular to the data line, as shown by arrow B, as in channels ch1 and ch2, "1. A 1” bit output signal is obtained. However, as shown in channel ch3, when the data line and the scanning direction intersect diagonally, there are relatively long and continuous "1" lines.
An output signal of "1.1...1" bit is obtained.

In this way, as shown in FIG. 3(C), channels chi, ch2, ch3... Chn
The position of the line is obtained as a digital signal. This signal S4 is divided into 8-bit units in the S/P conversion circuit 5 as shown in FIG. 3(E). Then, the first 8 bits are converted in parallel as "11000000", and the DMA circuit 6 stores them in the RAM 7s, 72.73
...Written to 7n. Then the second 8b
It does the same and RAM71.72.73...
7n. Thereafter, data for one scan is written in order. In this way, data for -scanning is written into the RAMs 71, 72, 73, . . . , 7n of each channel. And CPU81.82.83...8n
The DMA circuit 6 transfers data corresponding to its own channel to the respective RAMs 71, 72, 73...7n.
Read from. The timing of this readout is controlled by the DMA circuit 6 so as not to overlap with the timing of writing to the RAMs 71, 72, 73, . . . , 7n.

Next, the operation of the CPU 81.82.83*8n reading data corresponding to its own channel will be described.

As shown in FIG. 4, zero calibration data is recorded on the recording paper 1 by a pen recorder before data is recorded. This indicates the reference position of the waveform and is represented by a straight line parallel to the running direction of the recording paper 1. C
Each of the PU51, 82, 83...8n detects the position based on its own zero calibration waveform, and performs a predetermined processing area based on the detected zero position. Programmed to read data.

This procedure is as shown in Figure 5, after starting the reading operation,
In step S1, each channel chi, ch2, ch3,
...The zero calibration waveform of chn is read, and in step S2, this read value is set as the zero value for each channel.
PU1, CPU2, CPU3... Stored in CPUn. Thereafter, in step S3, each of the measured channels shown in FIG.
- Read the chn waveform and calculate the difference between that data and zero value data. This operation in step S3 is continuously repeated to convert the waveform on the recording paper l into an electrical signal.

For convenience of explanation, FIG. 6 is a block diagram showing a configuration for zero calibration. This is the operation section 2
When starting reading from 0, by performing a calibration operation, the CPU81.82.83...
8n to the calibration mode. This mode conversion is performed by switch circuit 211.212.21
3... It is done by 2In. This switch circuit 21! , 212.213...2In is CP
It is inserted between U81.82.83 e-so-*8n and D/A conversion circuit 91.92.93...9n, and CPU81.82.83... D/A conversion circuit 91.92.93...9n
and latch/drive circuits 221, 222, 223...
・The supply is switched between 22n and 22n.

231.232.233...23n is a switching transistor, 241.242.243...
...24n is a light emitting diode, and when a "1" signal is supplied to the latch/drive circuits 221, 222, 223...22n, this is latched and the light emitting diode continues to turn on. It is now possible to do so. 251
．． 252.253* * * -* -25n is CPU
81.82.83 m-**-* Memory circuit for storing zero value data attached to 8n.

Therefore, by the zero calibration operation, the CPU 81
．． 82', 83 ・am-**8n, switch circuit 21
1.212.213...Contest 21n is set to zero calibration state. In this state, each RAM71.
Zero value data of each channel is written in 72, 73...7n. In this state, the CPU 81 corresponding to channel ch1 reads the first data, and
U82' is the second, CPU83 is the third...
- The CPU 8n reads the nth data and stores this value in the memory circuits 251, 252, 253...25r1.
memorize each. This stored data means in which bit of 4096b i there is a line. Here, since the zero calibration waveform of the CPU 81.82.83...8t is within an approximately fixed range, the RAM 71.72.73...
...The processing range for reading data from 7n is a predetermined fixed range. This is because the pen recorder is able to make fine adjustments because it is necessary to change the pen position slightly depending on the data content, etc., so each time, the recording paper l, which has a different zero position for each channel, is in a stationary state. , if the fourth
As shown in the figure, for example, if the line at the position P for which zero calibration is to be performed for two channels is dropped or blurred,
I can't read the waveform. At that time, the CPU
Since the data "1" is not obtained from the output of the light emitting diode 232, it is found that the light emitting diode 232 does not emit light and zero calibration cannot be obtained. In this case, zero calibration is made possible by manually shifting the recording paper 1 slightly to allow it to escape from the drop-off location.

As shown in FIG. 7, data in a preset processing area is read from the RAM based on the set zero value.

Therefore, the CPU can read data of its own corresponding channel. This data is stored in memory circuits 251, 252, 253...25n as shown in FIG.
and the comparison calculation circuit 261.262.263...
. . . 26n performs a comparison operation and obtains digital data corresponding to the analog value.

This digital value is the D/A conversion circuit 91.92.93.
. . 9n are sequentially converted into analog signals and obtained as a continuous waveform as shown in FIG. 7, which is the same as the waveform recorded on the recording paper 1.

After all the conversion operations are completed, by operating a reset button (not disclosed) on the operation unit 20, all circuits are reset to a stop state.

In this way, each channel chi%ch2, ah3
. . . chn is independently set to its own zero value, and the waveform on the recording paper 1 is converted into an electrical signal using this as a reference. Therefore, a highly accurate electrical signal having the same waveform and peak value on the °C1 recording paper 1 can be obtained.

Further, the electric signal obtained from the image sensor 1 is converted into a digital signal and sent to the CPU 81, 82, 83, .
Since each signal is processed independently and simultaneously and converted into an analog signal by n, the processing speed is high even if the CPU and its program are relatively simple. Therefore, high-speed processing can be performed with a low-cost circuit configuration.

Incidentally, in the case where the data S4 is 1 consecutively, such as the channel ch3 shown in FIG. 3(C), the CPU 83 calculates the average thereof.

Furthermore, in one embodiment of the present invention, as shown in FIG. 3(C), 11'' is used as a negative line, but the CPU processes 1 on the left as the center of the line.

Even such processing does not affect accuracy at all.

Furthermore, even if a portion of the data waveform on the recording paper 1 is missing, a substantially continuous electrical signal can be obtained by providing a means for holding a previous value in the CPU.

Furthermore, in this embodiment, even if the waveform data D+=Dn is recorded in such a way that it exceeds the area assigned to itself, it can be read. FIG. 9 shows this principle in principle. In this embodiment, waveform data D1 to D, 6 from the 1st channel to the 16th channel
are recorded in parallel.

Since the image sensor 6 has 4096 bits (C512 bytes) and reads II@36 cm, simply 512
If Baitoke is divided into 16 channels, each channel will have 62 bytes. ffl, the number of bytes allocated to one channel is 62 bytes.

However, @ (,: PU 8+~8 extracts 44 bytes, which is more than 32 bytes, from the data of the − scan period.

Fill and extract data from each 6 bytes that protrude from the earthwork. When the S/P converter circuit 5 generates an 8-bit parallel digital signal, an address signal is attached to each byte of the -scanning period, so the CPUs 8□ to 8n use the memory 71 to 7 in response to this address signal. It is possible to easily extract the data of a predetermined area (44 bytes, 131 area) to which the own channel belongs. Since the scanning direction of the image sensor 6 is the direction that intersects the time axis of the waveform data, -
During the scanning period, a number of read outputs corresponding to the number of drums are obtained. CPU 8. ~8n may need to read adjacent waveform data as well, but it should be possible to determine whether the data is from the own channel or the adjacent channel by comparing it with the previous data.

Further, in this embodiment, data l is stored in a plurality of memories 71-7.
) DM A (Direct % iemory A
Since the waveform data read V signal by the image sensor 6 can be easily and quickly processed.

[Modifications] The present invention is not limited to the above-described embodiments, but can be modified as follows, for example.

(In the 11th embodiment, the analog signals obtained from the output terminals 10□ to 10n711) are recorded on a magnetic tape by a data recorder, but the digital data obtained from the CPUs 8□ to 8 are input to a computer or a waveform analysis device. You may let them. In addition, a waveform analysis function may be provided to the CPUs 8+ to 8n.

(In Embodiment 21, one recording sheet is continuously fed at a constant speed, but it may be fed intermittently. Also, the image sensor 6 may be fed continuously or intermittently in the direction in which the time axis of the waveform data D1 to Dn extends. You may move.

In short, it is sufficient to move the recording paper 1 and/or the sensor 6χ to obtain a relative scanning motion.

(3) In the embodiment, the sensor 3 is a -dimensional sensor consisting of a CCD, but a two-dimensional image sensor may be used instead.Also, the sensor 3 is an individual sensor,
This individual sensor may be moved in a direction opposite to the time axis of the waveform data. Further, the sensor 6 is not limited to a solid-state image sensor, and may be an image pickup tube or a non-visible light image sensor.

〔Effect of the invention〕

As described above, according to the present invention, a recording sheet (recording paper)
The above multi-channel waveforms can be easily and quickly converted into independent multi-channel data.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram showing a waveform reading device according to an embodiment of the present invention, FIG. 2 is an explanatory diagram showing waveforms on recording paper for explaining the present invention, and FIG. 3 is an explanatory diagram showing an embodiment of the present invention. A waveform diagram showing the operation in FIG.
Fig. 5 is a time chart for explaining the operation of zero calibration, Fig. 6 is a block diagram showing the configuration of zero calibration in an embodiment of the present invention, and Fig. 7 is an electric diagram in an embodiment of the present invention. FIG. 8 is a block diagram of the arithmetic processing circuit for obtaining the data in FIG. 7, and FIG. 9 is a diagram for explaining the data image area read by the CPU. be. 1... Recording paper, 3... Image sensor, 4... Digital signal forming circuit, 5... Series/parallel conversion circuit,
6... Direct access times Wr, 71-7n...
Random access memory, 8+~8n...CPU,
9□~9n...Digital-to-analog conversion circuit, ch
l, ch2, ch3~chn...channel. Agent: Nori Takano Figure 2 Figure 3 '8 Reyu? -Mu Figure 7 Figure 8 Figure 2! . (8yo Yur3 12!~)

Claims (1)

[Scope of Claims] [1] A method for independently obtaining a plurality of electrical signals corresponding to waveforms of a plurality of channels recorded in parallel on a recording sheet, comprising one-dimensional, two-dimensional or An individual optical sensor is arranged to face the recording sheet, and when the sensor is a one-dimensional or two-dimensional sensor, it is possible to read all of the waveforms of the plurality of channels in a direction intersecting the time axis of the waveform. arranging the sensor and electrically scanning the sensor in a direction intersecting the time axis; if the sensor is an individual sensor, moving the individual sensor in a direction intersecting the time axis; obtaining reading outputs of the plurality of channels of waveforms by moving one or both of the sheet and the sensor continuously or intermittently in the direction of the time axis of the waveform; forming a digital signal by shaping the read output to form a digital signal; writing the digital signal to each of the plurality of channels of memory corresponding to the waveform of the plurality of channels; out of the plurality of channels of memory to which the own channel belongs; A multi-channel waveform reading method comprising reading digital signals in a predetermined area and independently outputting data corresponding to the waveforms of the plurality of channels.