JP3430484B2 - Transmitter, receiver, transceiver, and data transmission system - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a transmitter, a receiver, a transceiver for transmitting non-image data such as numerical data and character data, and a data transmission system.

[0002]

2. Description of the Related Art In data communication using a fixed network or a mobile communication network, data to be transmitted is usually compressed on the transmitting side and expanded on the receiving side in order to improve transmission efficiency. In recent years, with the progress of multimedia in communication, compression / decompression technology suitable mainly for image data (still image data, moving image data) has been developed one after another.
In transmission of image data, transmission at a higher compression rate is realized as compared with transmission of non-image data such as numerical data and character data.

In image data compression / decompression technology,
Compared to reversible technology, JPEG (ISO / IEC 1
The progress of non-reciprocal technology represented by non-reciprocal method of 0918 standard) is remarkable. The reason for this is that the compression ratio in the reversible technology is close to the theoretical upper limit, it is difficult to expect a significant improvement in the compression ratio, and it is difficult to achieve a compression ratio that satisfies various conditions such as transmission line capacity. In addition, if the human visual characteristics are sufficiently taken into consideration, it is possible to suppress the deterioration of the image quality even with the irreversible type. For example, in the JPEG lossy method, discrete cosine transform (DCT) is used to suppress deterioration of image quality and to achieve a high compression rate (for example, 1/10).
Can be realized.

[0004]

As described above, with respect to image data, more efficient compression / decompression technology is being developed and standardized, but non-image data such as numerical data and character data is old. General compression from
Since the decompression technology has to be applied, as a result, the compression rate is at best about 1/2, which is significantly lower than the compression rate when the image compression / decompression technology such as JPEG is applied to the image data. Therefore, the effect of reducing the communication time and the communication cost due to the application of the compression / decompression technique is suppressed.

Further, as described above, lossy compression /
The decompression technology uses human characteristics to suppress image quality deterioration, and is difficult to adapt to non-image data that requires strict reproduction. For example, even if the color of one dot in a forest image changes from green to yellow-green, it is difficult for humans to visually recognize it.
If the leading bit in the binary representation "11111111" of the numerical value of 5 changes, the value becomes 127, and if a calculation using this value is performed, a completely incorrect result will be obtained. Similarly, the character code 1
It is possible that the character "A" may be changed to a completely different character "B" simply by changing the bit. Therefore, no matter how high the compression rate is, application of lossy compression / decompression technology to non-image data is not considered at all.
There was also no technology for applying decompression technology to non-image data.

The present invention has been made in view of the above-mentioned circumstances, and is a transmitter, a receiver, a transceiver, and data transmission capable of improving the transmission efficiency of non-image data such as numerical data and character data. The purpose is to provide a system.

[0007]

In order to solve the above-mentioned problems, the present invention reduces the capacity of transmission data by converting non-image data into image data and applying an image compression / decompression technique. is doing. Furthermore, the present invention reduces retransmission overhead by applying majority logic. The specific means will be described below.

The transmitter according to the present invention selects a two-dimensional pattern corresponding to the value of non-image data input from the outside from a preset two-dimensional pattern group , and selects the selected two-dimensional pattern.
A pattern conversion unit that repeats a dimensional pattern a plurality of times to generate repeated data, and image data representing an image including the repeated data generated by the pattern conversion unit is compressed by a predetermined image compression method. And a transmitting unit that transmits the image data compressed by the image compressing unit .

[0009]

A receiver according to the present invention receives a compressed image data, an image decompressing unit for decompressing the received data of the receiving unit to restore the image data, and an image decompressing unit for decompressing the image data. From the image represented by the image data,
Repeated data generated by repeating a two-dimensional pattern included in a preset two-dimensional pattern group a plurality of times is extracted.
When it is output, the repetitive data is added to the two-dimensional pattern.
And a pattern matching unit that outputs a plurality of non-image data by dividing the pattern matching unit.
From the multiple non-image data divided by
It comprises a majority logic unit for determining one of the non-image data I that has been characterized.

[0011]

The transceiver according to the present invention selects a two-dimensional pattern corresponding to the value of non-image data input from the outside from a preset two-dimensional pattern group , and selects the selected two-dimensional pattern .
Repeat two-dimensional pattern multiple times to generate repeated data
A pattern conversion section for forming, raw by the pattern conversion section
Receiving an image compression unit, a transmission unit for transmitting the image data compressed by the image compressing unit, the compressed image data to compress the image data representing an image including the formed by repeating data in a predetermined image compression method a receiving section that, an image decompression unit for restoring the image data by decompressing the received data of the receiving unit, from the image represented by the restored image data by the image decompression unit, included in the two-dimensional pattern group Repeat generated by repeating a two-dimensional pattern multiple times
When data is extracted, the repeated data is
A pattern matching unit for outputting a plurality of non-image data <br/> divided into each repetition of the original pattern, the pattern matcher
Many from multiple non-image data divided by the ching part
Majority theory that determines one non-image data by decision logic
It is characterized by comprising a processing section.

[0013]

[0014]

The data transmission system according to the present invention, the above-described transmitter, receiver, and that you are characterized by combining a transceiver suitable for transmitting non-image data.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. <1. Configuration of First Embodiment><1-1. Configuration of Transmission Side> FIG. 1 is a block diagram showing the configuration of the data transmission system according to the first embodiment of the present invention. In this figure, reference numeral 1 is a transmitter for transmitting non-image data such as numerical data and character data supplied from the outside via a public network, and arithmetic processing means (CPU etc.) and storage means (ROM, RAM etc.). , I / O interface, communication means (modem, etc.), and the like. In the transmitter 1, 11 is a pattern conversion unit that converts the non-image data into a two-dimensional pattern and generates image data including at least one two-dimensional pattern, and performs processing described below.

FIG. 2 is a flow chart showing the flow of processing by the pattern conversion unit 11, and FIG. 3 is a diagram showing an example of an image represented by the image data obtained as a result of the processing shown in FIG. As shown in these figures, when the non-image data is supplied, the pattern conversion section 11 corresponds to a predetermined size area (for example, 640 × 480 dots, 1-bit basic image for each dot) secured in the storage means. Area)
A start code for writing image data representing a two-dimensional pattern generated in a step described later is written (step SA1). This start code is image data, and represents a two-dimensional pattern (for example, “ >> ”, which will be referred to as a start pattern hereinafter) of a predetermined size (for example, 24 × 24 dots) indicating the start of the data area. In the present embodiment, the writing position of the image data is controlled so that the image represented by the image data is continuous in the row direction (right direction in the drawing) on the basic image as shown in FIG. When the edge of the image is reached, it is supposed to shift in the column direction (downward in the figure).

Next, the pattern conversion section 11 determines whether or not the supplied non-image data is non-character string data (step SA2). Numerical data are examples of non-character string data, and character data and symbol data are examples of constituent elements of character string data. That is, the character string data is data for which the value of each constituent element should be recognized as a character code, and the other non-image data is non-character string data.
In the present embodiment, it is assumed that the character string data also includes data having one component (character data).

If the non-image data is non-character string data, the result of the determination in step SA2 is "YES", so the pattern conversion section 11 performs a conversion process from the non-character string data to the character string data. (Step SA3). As a specific example, when the non-character data is numeric data, the value of each digit is calculated based on the radix used when displaying the numeric value, and the character representing the value of each digit is changed from the upper digit to the lower digit. Numerical data is represented by sorting in order. For example, numerical data that is 255 in decimal notation is represented by character string data indicating a character string “255”. Needless to say, the alignment order of the characters is not limited to the above-described order (from upper to lower).

Step SA4 is a process subsequent to step SA3, and also a process when the judgment result in step SA2 is "NO" (when character string data is supplied). Here, the pattern conversion is performed. Part 11
Converts each character data in the character string data into a two-dimensional pattern in order. Since the actual value of the character data is the character code, the pattern conversion unit 11
Refers to the comparison table of the character code and the two-dimensional pattern stored in the storage means (not shown), and the corresponding 2
By selecting the dimension pattern, 2
Can be converted into a dimensional pattern. The format of the two-dimensional pattern is a bitmap format with a predetermined size (for example, 24 × 24 dots). Further, in step SA4, the pattern conversion unit 11 writes the image data representing the selected two-dimensional pattern in the area on the storage unit corresponding to the basic image, so that the two-dimensional pattern corresponding to the non-image data is predetermined. Arrange them on the basic image in order.

Next, the pattern conversion section 11 judges whether or not the arrangement of the two-dimensional pattern has been completed for all the non-image data to be transmitted (step SA5). If not completed, the process proceeds to step SA1. Return to. As a result, as shown in FIG. 3, each two-dimensional pattern string corresponding to each non-image data is divided on the starting pattern such as ">>" and arranged on the basic image. That is, the start pattern also represents a data delimiter.

When it is determined in step SA5 that the placement of the two-dimensional pattern has been completed for all the non-image data to be transmitted, the pattern conversion section 11 determines an end pattern (for example, "<< )) Is written such that the end pattern is located at the end of all two-dimensional patterns on the basic image (step SA6).

In the transmitter 1 of FIG. 1 again, 12 is an image compression unit, which is well known as image data (image data of an area corresponding to a basic image) in which data representing non-image data is written by the pattern conversion unit 11. The compressed image data is generated by compressing with the image compression method of. Any image compression method can be adopted here, but in the present embodiment, JP
The non-reciprocal method of EG is adopted. In the transmitter 1, 13 is a transmission unit that transmits the compressed image data generated by the image compression unit 12 via the public network.

<1-2. Configuration of Receiving Side> On the other hand, 2 is a receiver that extracts and outputs non-image data from received data received via a public network, and like the transmitter 1, arithmetic processing means (CPU etc.), storage means (ROM, RAM, etc.),
It is realized by an I / O interface, communication means (modem, etc.) and the like. In the receiver 2, 21 is a receiving unit that receives data via the public network, 22 is an image decompressing unit, and the received data of the receiving unit 21 is regarded as compressed image data, and this is the image compression method of the image compression unit 12. It is expanded by the method corresponding to. Note that it is easy to configure the image decompression unit 22 to be able to decompress compressed image data in various formats, but in order to avoid complication of the description, in the present embodiment, only decompression by the JPEG lossy method is possible. It has a simple structure.

Further, in the receiver 2, 23 is a pattern matching unit for scanning the image data expanded by the image expansion unit 22 to extract non-image data, and the scanning direction is two-dimensional by the pattern conversion unit 11. It is set to be the same as the arrangement direction of the pattern. FIG. 4 is a flowchart showing the flow of processing by the pattern matching unit 23. As shown in FIG.
3 continues extracting a two-dimensional pattern of a predetermined size (the same size as the two-dimensional pattern used in step SA4 of FIG. 2) until it finds a two-dimensional pattern that matches the start pattern previously registered in the storage means ( Step SB
1, SB2).

However, "matching" here does not require strict matching, and even if a part of the two-dimensional pattern is different, it is sufficient if it can be specified that it is the start pattern registered in the storage means in advance. is there. Since such a pattern matching method is well known, the description of the specific procedure will be omitted.

When the start pattern is found, the pattern matching unit 23 extracts the end pattern or the two-dimensional pattern that matches the start pattern registered in the storage means in advance, that is, until the end pattern or the start pattern is extracted. The extraction of the dimensional pattern and the acquisition of the corresponding character code are repeated (step SB
3 to SB6). In addition, the storage unit (not shown) includes a transmitter 1
Contains the same contrast table that
The pattern matching unit 23 refers to this comparison table to acquire the character code. Specifically, the pattern matching unit 23 refers to the comparison table to extract a two-dimensional pattern. If the two-dimensional pattern is neither a start pattern nor an end pattern, the pattern matching unit 23 refers to the above-mentioned comparison table to extract the two-dimensional pattern. The character code corresponding to is acquired, and the character code is additionally stored in a buffer (not shown).

In step SB5, when the start pattern is detected, the pattern matching unit 23 checks whether or not the character sequence represented by the series of character codes obtained immediately before satisfies the numerical expression. If not satisfied, the series of character codes is determined to be character string data and the process returns to step SB3. If satisfied, it is determined to be non-character string data (numerical data) (step SB7).

If it is determined that the data is non-character string data, the pattern matching unit 23 determines the non-character string data (based on the value indicated by the character represented by each character code and the acquisition order of each code). After converting to numerical data, the process returns to step SB3 (step SB8). For example, when the character string represented by a series of character codes is "225", the values indicated by the characters "2", "2", and "5" are 2, 2, 5 respectively.
Therefore, assuming that the acquisition order is in the described order, 2 × 102 = 200 2 × 101 = 205 × 100 = 5, and thus the data is converted into numerical data having a value of 225.

When the end pattern is detected in step SB4, the pattern matching section 23 carries out the processes of steps SB9 and SB10. Step SB
Steps 9 and SB10 are substantially the same as steps SB7 and SB8 described above, except that the process does not return to step SB3. The acquired character string data and non-character data are sequentially output to the outside as non-image data.

<2. Operation of First Embodiment><2-1. Operation on Transmission Side> According to the configuration described above, the non-image data is converted into a two-dimensional pattern string by the pattern conversion unit 11 of the transmitter 1, and is arranged on the basic image following the start pattern (for example, “ >> ”). To be done. After this process is repeated for the number of non-image data to be transmitted, the end pattern (for example, "<<") is arranged, and the basic image as shown in FIG. 3 is obtained. The image data representing the basic image thus created is lossy-compressed by the image compression unit 12, and the transmission unit 1
3 via the public network.

<2-2. Operation on receiving side> The compressed image data received by the receiving unit 21 via the public network is the image decompressing unit 22.
The image data is decompressed and restored by. The image data restored in this way usually has a lower image quality than the original image data. Therefore, if pattern matching that requires strict matching is applied, it may not be possible to recognize a two-dimensional pattern corresponding to non-image data, but the pattern matching unit 23 performs pattern matching that does not require strict matching. Therefore, the two-dimensional pattern can be extracted.

The two-dimensional pattern string sandwiched between the start patterns or between the start pattern and the end pattern is regarded as representing non-image data, and the two-dimensional pattern is converted into a character code for each two-dimensional pattern string. To be done. When the original non-image data is character string data (for example, "SUNDAY"), the character code string thus obtained can be directly used as the non-image data. When the original non-image data is non-character string data (for example, numerical data), the character code string is converted to non-character string data by the conversion process described above, and then output as non-image data. It

<3. Summary of First Embodiment> According to the above-described first embodiment, the image compression / decompression technique can be applied to non-image data such as numerical data and character string data, so that the compression rate is improved and the transmission efficiency is improved. Can be raised. Moreover, since a non-reversible image compression / decompression method can also be applied, a significant improvement in compression rate can be expected.

Further, since the pattern matching which does not require strict matching at the receiving side is adopted, even if the received data has an error, the non-image data is correctly processed as long as the characteristics of the two-dimensional pattern are not changed. Can be restored. This means that the quality required for the transmission path can be suppressed low and the frequency of retransmission requests can be reduced. Furthermore, an image data output section is provided on the receiving side, and this image data output section allows
By outputting the decompressed image data to an external display, etc., the non-image data can be visually displayed on the image represented by the image data. The data may be visible to the naked eye.

Incidentally, regardless of the reversible / irreversible type, JP
Image compression / decompression technology other than EG's lossy method can be adopted. Further, it is possible to apply not only to data transmission via the public network but also to data transmission using a dedicated line.
Of course, it can be applied to data transmission via ISDN (Integrated Digital Communication Network) or mobile communication network, and can also be applied to inter-process communication in a computer.

Although an example in which a general character is used as the two-dimensional pattern has been shown, if the two-dimensional pattern to be used is preset between the transmitting side and the receiving side, an arbitrary pattern can be obtained. Can be used. Furthermore, when the non-image data to be transmitted is limited to one type (for example, numerical data), the determination process of step SA2 of FIG. 2 and steps SB7 and SB9 of FIG. 4 can be omitted. In addition, the number of non-image data to be included in the image data representing one basic image, the size of the basic image,
How to divide a two-dimensional image corresponding to non-image data is an appropriate design matter.

Furthermore, in the first embodiment, it is premised that the numerical data is expressed in a decimal number when the two-dimensional pattern is formed, but the numerical data is not limited to the decimal number, and may be expressed in any radix. Further, if the information indicating the type of data (character string data / non-character string data) is written in the image data of the basic image, the determination process in steps SB7 and SB9 in FIG. 4 can be omitted.

Further, instead of arranging the two-dimensional patterns in a fixed direction, the two-dimensional patterns are randomly arranged as shown in FIG. 5, and information representing a random number sequence at that time is included in the image data of the basic image. Even if a third party refers to the image represented by the received image data, it becomes difficult to read the non-image data. That is, the confidentiality of the data can be easily ensured.

Needless to say, if transceivers having both the function of the transmitter 1 and the function of the receiver 2 are arranged on the transmitting side and the receiving side, bidirectional communication becomes possible. Further, if the first direction of communication, the transmitter and the transceiver, it is possible with a combination such as receiver and transceiver.

By the way, in general data communication, data that has failed to be transmitted is retransmitted. However, in an environment where it is difficult to maintain the quality of a communication channel such as mobile communication, the overhead required for retransmission increases, which causes a problem. This causes a decrease in transmission efficiency. In the first place, since there is a potential factor in reducing the transmission efficiency in general retransmission processing, it is desirable to avoid the retransmission processing itself. The retransmitting process and its problems in general data communication will be described below.

In general data communication, the transmission side transmits error-detected / correction-encoded data, and the reception side performs error detection / correction processing on received data.
If the normal reception fails (reception error), the sender is requested to retransmit. For example, commonly used BS
In the C (binary synchronous communication) protocol, on the transmitting side,
In the case of normal reception, an error detection code such as BCC (block check character) is added to the transmission data, and the reception side performs error detection using the error detection code to determine normal reception / reception error. A signal (ACK) indicating that fact is returned to the above, and a signal (NACK) indicating that is returned in the case of a reception error. In addition, NA
The transmission side that receives the CK retransmits the data corresponding to the NACK.

In such a BSC protocol, the overhead of retransmission processing at the time of a reception error is roughly classified into the transmission of NACK and the retransmission of data. The latter depends on the size (length) of data to be transmitted, but the former is constant. That is, if a reception error occurs during the transmission of small data, an excessive overhead is generated compared to the size of the data to be transmitted, resulting in a decrease in data transmission efficiency and a longer communication time. I will end up.

In wireless data transmission represented by mobile communication, the quality of the transmission path is often deteriorated during data transmission due to factors such as changes in electric field strength and fading, and reception errors often increase. . Further, due to the nature of the wireless system, it is extremely difficult to predict the deterioration of the transmission path quality, and the longer the communication time, the higher the probability that the deterioration will occur within the communication period. Therefore, in wireless data transmission, it is desired to reduce or reduce the above-mentioned overhead not only from the viewpoint of improving the data transmission efficiency but also from ensuring communication reliability. Here, a second embodiment of the present invention that can satisfy such a demand will be described.

<4. Configuration of Second Embodiment><4-1. Structure of Transmission Side> FIG. 6 is a block diagram showing the structure of a data transmission system according to the second embodiment of the present invention. In this figure, the same parts as those of FIG. The description is omitted. In FIG.
Reference numeral 10 designates a transmitter for transmitting non-image data such as numerical data and character data supplied from the outside through a mobile communication network, and its realization means is that it has a mobile device as a communication means. Different from the transmitter 1. This transmitter 10
Is different in function from the transmitter 1 of FIG. 1 in that the pattern conversion unit 11 is replaced with a pattern conversion unit 11a and error detection
The point is that the correction encoding unit 14 is provided.

The error detection / correction coding unit 14 applies a predetermined error detection / correction code (for example, CR to the data to be transmitted (hereinafter referred to as transmission data) supplied from the outside).
C) is added and output, and the pattern conversion unit 11a converts the output data of the error detection / correction encoding unit 14 into a two-dimensional pattern and generates a two-dimensional pattern string including at least one two-dimensional pattern. Below, the pattern conversion unit 11
The processing performed by a will be described with reference to FIG. 7.

FIG. 7 is a flowchart showing the flow of processing by the pattern conversion unit 11a, and FIG. 8 is a diagram showing an example of an image represented by the image data obtained as a result of the processing shown in FIG. In FIG. 7, the same parts as those in FIG. 2 are designated by the same reference numerals, and the description thereof will be omitted.

As shown in FIG. 7, in step SA7 following step SA2 and step SA3, the pattern conversion unit 11a sequentially forms the two-dimensional pattern of the character string data obtained up to the previous step. Then, in step SA8, it is determined whether or not the length of the two-dimensional pattern string obtained in step SA7 is less than or equal to a predetermined value, and if it is less than the predetermined value, the two-dimensional pattern string is sequentially repeated an odd number of times. To generate a repeating pattern (step S
A9). However, a two-dimensional pattern (for example, "●", hereinafter referred to as a delimiter pattern) indicating a delimiter of each two-dimensional pattern string is inserted in the repeating pattern.
Note that the number of repetitions of the two-dimensional pattern sequence can be set arbitrarily, and in the present embodiment, it is set to three.

Step SA10 is a process subsequent to step SA9, and also a process when the determination result in step SA8 is "NO". The pattern conversion section 11a causes the repetitive pattern obtained in step SA9, Alternatively, the image data representing the two-dimensional pattern string determined to be longer than the predetermined value in step SA8 is written in the writing position described above.

The pattern conversion section 11a uses the step SA1.
~ SA3, SA7 to SA10 are repeated until the area of the predetermined size reserved in the storage means is full or the non-image data to be transmitted is exhausted (step SA5). As a result, as shown in FIG. 8, each two-dimensional pattern sequence corresponding to each non-image data is divided by the start pattern and arranged on the basic image.

When any of the above-mentioned conditions is satisfied, the pattern conversion section 11a outputs image data representing an end pattern (for example, "<<") indicating the end of non-image data, the end pattern on the basic image. Writing is performed in the area corresponding to the basic image so as to be located at the end of all the two-dimensional patterns (step SA6).

<4-2. Configuration of Receiving Side> Meanwhile, 2 in FIG.
Reference numeral 0 denotes a receiver that extracts and outputs non-image data from received data received via a mobile communication network.
It is realized by the same means as. The functional difference of the receiver 20 from the receiver 2 of FIG. 1 is that it has a pattern matching unit 23 a, a majority logic unit 24, and an error detection / correction unit 25 instead of the pattern matching unit 23.

FIG. 9 is a flow chart showing the flow of processing by the pattern matching section 23a. In this figure, the same parts as in FIG. 4 are assigned the same reference numerals and explanations thereof are omitted. In FIG. 9, step SB9
-SB11, SB14, SB15 and SB9a-SB11a, SB14a, SB15a is the same processing content.

As shown in FIG. 9, when the start pattern is found and the judgment result in step SB2 is "YES",
The pattern matching unit 23a extracts the two-dimensional pattern and responds until the two-dimensional pattern that matches the start pattern or the end pattern registered in the storage unit is extracted, that is, until the start pattern or the end pattern is extracted. The acquisition of the character code is repeated (steps SB3, SB12, SB13, SB6).

If the start pattern is extracted during the repeated acquisition of the character code, the pattern matching unit 23a determines whether or not the two-dimensional pattern represented by the character code string on the buffer is a repeated pattern. (Step SB14). This determination is realized by checking the presence or absence of the character code corresponding to the delimiter pattern in the character code string. If the determination result in step SB14 is "YES", the pattern matching unit 23a divides the character code string in the buffer with the character code of the delimiter pattern as a boundary (step SB15). It should be noted that each character code string after division does not include a character code representing a delimiter pattern.

Step SB9 is a process subsequent to step SB15, and is also a process when the determination result in step SB14 is "NO". In this step, the pattern matching unit 23a causes the character code on the buffer. It is determined whether the string is non-character string data. Note that, in the present embodiment, since only non-character string data exists as non-character string data, the above determination is actually performed by checking whether or not the character code sequence on the buffer does not satisfy the numerical expression. To be done. When there are a plurality of character code strings in the buffer (when the character code strings are divided in step SB15), the above determination is performed only for the leading character code string.

The determination result in step SB9 is "YES".
When the character code string on the buffer is determined to be non-character data, the pattern matching unit 23a determines the value represented by the character represented by each character code and the acquisition order of each code. To convert the character code string on the buffer into non-character string data (numerical value data) (step SB1).
0).

Step SB11 is a process subsequent to step SB10, and also a process when the determination result in step SB9 is "NO". In this step, the pattern matching unit 23a obtains by the above-described process. The output data is output to the subsequent stage, the buffer is cleared, and the process returns to step SB3. At this time, the pattern matching unit 23a outputs the data to the majority logic unit 24 to be described later when the data to be output is plural, and the data to be described later when the data to be output is singular. Output to the error detection / correction unit 25.

Steps SB3, SB12, SB13, S
If the end pattern is extracted during the processing of B6 (step S
B13), the pattern matching unit 23a, step S
B14a, SB15a, SB9a, SB10a, SB1
The processing of 1a is performed. This process is the same as step SB described above.
14, SB15, SB9, SB10, SB11 are almost the same processes, and the only difference is that the process does not return to step SB3 and ends.

Again in FIG. 6 , the majority logic unit 24.
Determines one data with the lowest probability of including an error from the plurality of data output from the pattern matching unit 23a. FIG. 10 is a flow chart showing the flow of the above-mentioned decision processing by the majority logic unit 24. As shown in this figure, the majority logic unit 24 first checks whether each data matches or does not match (step SC1). A match / mismatch determination method is arbitrary, but in the present embodiment, a match is obtained when all the exclusive ORs of corresponding bits are 0 (false),
If even one is 1 (true), it is considered as a mismatch. Further, in the present embodiment, the matching data is represented by one of the data, and thereafter, the representative data is compared with other data to reduce the number of comparisons.

Then, the majority logic unit 24 determines whether or not all the data do not match (step SC2),
If this determination result is “YES”, information indicating a reception error is output to the error detection / correction unit 25 (step S
C3). On the contrary, if the determination result in step SC2 is "NO", the data having the maximum number of matches is output (step SC4). Since there are a plurality of pieces of data with the maximum number of matches, it is arbitrary which of these pieces of data is selectively output.

Referring again to FIG. 6 , the error detection / correction unit 25
Performs error detection processing and error correction processing corresponding to the error detection / correction coding unit 14 of the transmitter 10 on the data output from the majority logic unit 24 and the data output from the pattern matching unit 23a. However, when the information indicating the reception error is supplied from the majority logic unit 24, the above process is not performed and this information is directly output.

The error detection / correction unit 25 outputs the result of the error detection / correction processing to the outside. The results here are the non-image data obtained by removing the information and the error detection / correction code when no error is detected, and the error was detected but could be completely restored by error correction. In this case, the non-image data includes information to that effect and the non-image data after restoration, and if the information cannot be completely restored by error correction, the non-image data includes information to that effect and the error after restoration.

<5. Operation of Second Embodiment><5-1. Operation on Transmission Side> According to the above-described configuration, the non-image data is stored in the error detection / correction encoding unit 14 of the transmitter 10.
Error detection / correction coding in the pattern conversion unit 1
In 1a, it is converted into a two-dimensional pattern sequence. In this pattern conversion process, when the length of the two-dimensional pattern string is equal to or less than the predetermined value, the two-dimensional pattern is repeated, a repeating pattern including a delimiter pattern is generated, and the repeated pattern is arranged on the basic image. On the contrary, when the number exceeds the predetermined value, the obtained two-dimensional pattern sequence is arranged as it is on the basic image. In the basic image, the start pattern is arranged at the beginning of the two-dimensional pattern sequence (or the repeating pattern) corresponding to each non-image data, and the end pattern is arranged at the end of the whole (see FIG. 8). The image data representing the basic image thus created is lossy-compressed by the image compression unit 12 and transmitted by the transmission unit 13 via the mobile communication network.

<5-2. Operation on receiving side> The compressed image data received by the receiving unit 21 via the mobile communication network is decompressed by the image decompressing unit 22, and the image data is restored. The image data restored in this way usually has a lower image quality than the original image data. Therefore, if pattern matching that requires strict matching is applied, it may not be possible to recognize a two-dimensional pattern corresponding to non-image data, but the pattern matching unit 23a performs pattern matching that does not require strict matching. Therefore, the two-dimensional pattern can be extracted.

Each two-dimensional pattern string sandwiched between the start patterns or between the start pattern and the end pattern is considered to represent non-image data, and is converted into image data for each two-dimensional pattern string, resulting in an error. It is supplied to the detection / correction unit 25. However, since a two-dimensional pattern string including a delimiter pattern is recognized as a repeating pattern in the pattern matching 23a, a plurality of image data represented by a plurality of two-dimensional pattern strings obtained by dividing the repeating pattern are majority-determined. It is supplied to the logic unit 24.

In the majority logic unit 24, when a plurality of image data are supplied from the pattern matching unit 23a, the image data are compared with each other, and if there is even one set of matching image data, the matching image data The image data having the maximum number is selected and supplied to the error detection / correction unit 25.
On the contrary, when there is no matching image data, information indicating a reception error is supplied to the error detection / correction unit 25.

In the error detection / correction unit 25, when the information indicating the reception error is supplied from the majority logic unit 24, this is through-outputted, and in other cases, the pattern matching unit 23 and the majority logic unit. The error detection / correction process is performed on the image data supplied from any one of 24. As a result of this error detection / correction processing, the error detection / correction status (whether or not there is an error, whether or not it has been completely corrected) and the data (non-image data) after the error detection / correction processing are output.

<6. Summary of Second Embodiment> The advantages of the second embodiment that cannot be obtained by the first embodiment will be described below. In the second embodiment, when the data to be transmitted is short, the transmitting side generates a repeating pattern, and the receiving side selects correct data from a plurality of data obtained by dividing the repeating pattern by majority logic. As a result, even in an environment where the transmission path quality is likely to change, such as mobile communication, without retransmitting,
Data can be transmitted correctly.

Of course, if a transmitting / receiving unit is provided on each of the transmitting side and the receiving side and a reception error occurs (when all data is different in the processing by the majority logic unit 24 or error correction fails), the receiving side May request the sender to resend, and in such cases,
There is an advantage that the frequency of retransmission can be reduced by using the majority logic. In particular, when embedding a plurality of non-image data in the image data representing one image as in the second embodiment, the amount of data to be retransmitted becomes large, so that the frequency of retransmission occurrence is reduced, Transmission efficiency is greatly improved.

The majority logic unit 24 may request retransmission processing when the number of matching data is less than the majority. Furthermore, instead of comparing the entire data, it may be compared for each bit or character forming the data. In this case, each bit or each character forming the finally determined data is a value (0 or 1) having the largest number of identical values or a character.

Further, in the second embodiment, since the repeating pattern is generated by repeating the two-dimensional pattern string, the transmitting side is compared with the case where the data in which the non-image data is repeated is two-dimensionally patterned. The burden can be reduced. Further, since the repetitive pattern is used only for the data having a large overhead of the retransmission processing (the data having a short two-dimensional pattern string), the amount of data to be transmitted is not increased so much and the transmission efficiency is greatly reduced. It is possible to reduce the occurrence of reception errors during data transmission.

In the second embodiment, an example in which communication is performed via a mobile communication network has been described. However, the present invention is not limited to this, and the above-described case can be applied to any line or line network in which the transmission line quality is likely to change. Similar effects are obtained. Furthermore, a repeating pattern may be generated and used for all non-image data. Further, in the aspect of making a retransmission request, the frequency of retransmission requests may be monitored, and when the frequency of occurrence exceeds a predetermined threshold, the number of repetitions may be changed to improve the reliability of communication.

Further, in the second embodiment, the delimiter pattern is sandwiched between each two-dimensional pattern row in the repetitive pattern, but a pattern indicating the repetitive pattern is placed at the head, and then each You may make it arrange | position a two-dimensional pattern showing the length of a two-dimensional pattern sequence, the number of repetitions, etc., and arrange | position each two-dimensional pattern sequence after that. Further, the error detection / correction encoding / decoding may not be performed.

<7. Application example><7-1. Structure of Application Example> FIG. 11 is a block diagram showing a structure of a rockfall detection system obtained by modifying the data transmission system according to the first embodiment. In this figure, 3 is an observation point side device installed at a location where rockfall is a concern, such as a rock wall surface or a wall surface of an underground cavity. Data indicating a sign of rockfall is used for mobile communication network and fixed network (hereinafter communication network). Say) to send via. Reference numeral 4 denotes a center side device that collects, announces, and analyzes data transmitted from the observation point side device 3 via the communication network.

<7-1-1. Configuration of observation point side device 3>
In the observation point side device 3, 31 is a data logger to which a plurality of (for example, 32 × 2 = 64) cable sensors for detecting a precursor of a rock fall are connected, and after amplifying the signal from each cable sensor, A / D conversion is performed at 100 KHz, and each is output as 12-bit sensor data. Although the cable sensor itself is well known, a detailed description thereof will be omitted. However, the cable sensor is a cable-shaped sensor having an outer diameter of about 5 mm and outputs a voltage signal according to an impact received from the outside. In this rockfall detection system, a plurality of cable sensors are laid in a grid pattern at predetermined intervals (for example, 80 cm intervals) so as to cover the wall surface of the observation target, and the voltage value of the output signal of each cable sensor indicates the sign of a rockfall. The voltage value corresponds to the distribution of vibrations including the destructive sound and small-scale collapse sound.

Reference numeral 33 is a computer incorporating a CPU, ROM, RAM, hard disk, various I / O interfaces, and the like, and 34 is a mobile device for mobile communication.
34 is connected to the computer 32. The computer 32 inputs / analyzes each output data of the data logger 31 to determine whether there is an abnormality. When it is determined that there is an abnormality, a simple analysis process is performed, and the analysis result is treated as emergency data via the mobile unit 33 to the center. It is transmitted to the side device 4. Further, the computer 32 performs processing (described later) such as control of each element, data compression, and transmission of normal data according to an instruction supplied via the mobile device 33.

<7-1-2. Configuration of Center Side Device 4> On the other hand, in the center side device 4, 41 is a modem connected to a communication network, 42 is a computer incorporating a CPU, ROM, RAM, hard disk, magneto-optical disk drive, various I / O interfaces, etc. 43 is an input device such as a keyboard, 44 is an output device such as a display or plotter, and each of the elements 41, 43 and 44 is a computer 4
Connected to 2.

The computer 42 transmits an instruction to the observation point side device 3 via the modem 41 at a predetermined time interval, and when normal data is input from the modem 41 in accordance with the instruction, the normal data is expanded. Processing such as extraction of sensor data and storage of each data (described later) is performed.
Also, when emergency data is input via the modem 41,
The computer 42 performs output processing and storage processing of the emergency data. In addition, when the monitor operates the input device 43, the computer 42 reads the data stored in the external storage device 43, performs an analysis process on the read data, and notifies the monitor of the analysis result by the output device 44. .

<7-2. Operation of Application Example> The computer 32 of the observation point side device 3 performs the normal operation when all the output data of the data logger 31 is less than or equal to a predetermined threshold,
If even one of the output data exceeds a predetermined threshold value, an emergency operation is performed. The threshold value is set in advance according to the level of noise to be mixed, and may be 0.

<7-2-1. Normal Operation> In normal operation, the computer 32 of the observation point side device 3 sequentially stores the sensor data supplied from the data logger 31.
In the hard disk of the computer 32, an area for storing all sensor data generated in 30 seconds is secured as a storage area for sensor data. That is,
When handling one sensor data (12 bits) as 2-byte data, 64 [lines] x 2 [bytes] x 1
00000 [Hz] x 30 [seconds] = 384000000
An area larger than [byte] is secured.

Then, when a predetermined instruction is supplied from the center side device 4 through the mobile unit 33, the observation point side device 3
Reads the stored sensor data for 30 seconds and performs the same processing as the processing shown in FIG. 2 to generate binary basic image data. However, since the data to be transmitted is only the sensor data (numerical data), the process of step SA2 in FIG. 2 is unnecessary, and the process of step SA3 is performed on all non-image data (sensor data). . The basic image data thus obtained is compressed in the computer 32 by a reversible method such as JPEG, and is transmitted to the center side device 4 via the mobile device 33. The reversible method is used here because high-precision data is required.

When the center side device 4 receives the compressed data via the modem 41, the computer 42
Disconnects the line connection with the observation point side device 3, decompresses the received data to acquire basic image data, and performs the same processing as in FIG. 4 to extract non-image data. However, since the non-image data included in the basic image data is only the sensor data (numerical data), step SB in FIG.
The processing of 7 and SB9 is unnecessary, and the processing of step SA8 or SA10 is performed on all non-image data. Non-image data (sensor data) obtained in this way
Are stored in a hard disk or a magneto-optical disk in the computer 32.

Here, when the monitor operates the input device 43 and gives a predetermined instruction, the computer 32 stores the stored information.
Predetermined analysis processing is performed based on the sensor data for 0 seconds, and the result is output from the modem 41 to notify the supervisor.

<7-2-2. Emergency operation> In the emergency operation, the computer 32 displays 0.1 from the start of the emergency operation.
A simple analysis process is performed based on the 10 × 64 = 640 sensor data obtained by the second, and the result is transmitted to the center side device 4 via the mobile device 33. When the center side device 4 receives the compressed data via the modem 41, the computer 42 stores the received data (result of simple analysis) in the hard disk or the magneto-optical disk in the computer 32, and outputs the output device 41.
Notify the watchman by.

Further, the observation point side device 3 has 300 × 640 = 1 obtained 30 seconds after the start of the emergency operation.
92,000 pieces of sensor data are transmitted to the center side device 4 in the same procedure as in the normal operation, and the center side device 4 also
The sensor data is extracted from the received data and stored in the same procedure as the normal operation. Thereafter, performing a predetermined analysis process based on the instruction of the supervisor is the same as in the normal operation.
During an emergency operation, the line connection is disconnected from the observation point side device 3 side.

<7-3. Summary> As explained above,
In the normal operation, the non-image data is compressed / decompressed by using the image compression / decompression technique, so that a high compression rate is realized, and the sensor data necessary for the analysis processing in the center side device 4 is used as a monitoring system. Can be transmitted in a time that does not impair the function of.

It should be noted that the observation point side device 3 is provided with a CCD camera or the like (image data input section) for picking up an image of the wall surface, and in a normal operation, an unused area of the image data of the wall surface by the CCD camera (for example, the observation contrast is circular). In this case, the image data obtained by overwriting the two-dimensional pattern representing the sensor data on the four corners becomes the unused area) may be used as the basic image data. According to this, since the non-image data can be transmitted by utilizing the empty area of the image data, the overall transmission efficiency can be remarkably improved.

[0089]

As described above, according to the present invention,
Transmission using high compression rate image compression / decompression technology
Data volume can be reduced and repeated data
And the majority logic is used to reduce the probability of receiving errors.
Because it can be reduced, such as numerical data and character data
It is possible to improve the transmission efficiency of non-image data .

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a data transmission system according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing a processing flow by a pattern conversion unit 11 in the transmitter 1 in the same system.

FIG. 3 is a diagram showing an example of an image represented by image data obtained as a result of processing by the pattern conversion unit 11.

FIG. 4 is a flowchart showing a flow of processing by a pattern matching unit 23 in the receiver 2 in the system.

FIG. 5 is a diagram showing another example of an image represented by image data obtained as a result of the processing by the pattern conversion unit 11.

FIG. 6 is a block diagram showing a configuration of a data transmission system according to a second embodiment of the present invention.

FIG. 7 is a flowchart showing a processing flow by a pattern conversion unit 11a in the transmitter 10 in the same system.

FIG. 8 is a diagram showing an example of an image represented by image data obtained as a result of processing by the pattern conversion unit 11a.

FIG. 9 is a flowchart showing a processing flow by a pattern matching unit 23a in the receiver 20 in the same system.

FIG. 10 is a flowchart showing a processing flow by the majority logic unit 24 in the receiver 2 in the system.

FIG. 11 is a block diagram showing a configuration of a rockfall detection system obtained by modifying and applying the data transmission system according to the first embodiment of the present invention.

[Explanation of symbols]

1, 10 ... Transmitter, 2, 20 ... Receiver, 11, 11a ...
Pattern conversion unit, 12 ... image compression unit, 13 ... transmission unit, 1
4 ... Error detection / correction coding unit, 21 ... Receiving unit, 22 ... Image decompressing unit, 23, 23a ... Pattern matching unit, 24
... majority decision logic section, 25 ... error detection / correction section.

─────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-8-204967 (JP, A) JP-A 1-144182 (JP, A) JP-A 5-37700 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) H03M 7/40 H04N 1/41 H04N 7/24

Claims (15)

(57) [Claims] 1. A two-dimensional pattern corresponding to a value of non-image data input from the outside is selected from a preset two-dimensional pattern group , and the selected two-dimensional pattern is repeated a plurality of times.
A pattern conversion unit that generates a repetitive repeat data, an image compression section that compresses the image data representing an image comprising a repeating data generated by the pattern conversion unit at a predetermined image compression method, is compressed by the image compression unit And a transmitter for transmitting the image data. 2. The transmitter according to claim 1, wherein the image compression method is a lossy compression method. 3. An image data input unit for inputting image data, wherein the image compression unit is arranged in the pattern conversion unit in an unused area on an image represented by the image data input by the image data input unit. The transmitter according to claim 1, wherein image data representing an image obtained by arranging the repetitive data generated by the method is compressed by a predetermined image compression method. 4. An error detection / correction coding unit that performs error detection / correction coding on non-image data input from the outside and supplies the error detection / correction coding to the pattern conversion unit, wherein the pattern conversion unit is Two-dimensional pattern corresponding to the value of non-image data supplied from the error detection / correction coding unit
Pattern from a preset two-dimensional pattern group,
Repeat the selected two-dimensional pattern multiple times to
The transmitter according to claim 1 , wherein the transmitter generates a data. 5. A receiving unit that receives compressed image data, an image expanding unit that expands received data of the receiving unit to restore image data, and an image data that is restored by the image expanding unit. From the image that is generated by repeating the two-dimensional pattern included in the preset two-dimensional pattern group a plurality of times.
When data is extracted, the repeated data is
A pattern matching unit for outputting a plurality of non-image data <br/> divided into each repetition of the original pattern, a plurality of non-image divided by the pattern matching unit
One non-image data from image data by majority logic
A majority decision logic unit for making a decision . 6. The receiver according to claim 5 , further comprising an image data output unit that outputs the image data expanded by the image expansion unit. 7. Error detection / correction processing for performing error detection / correction processing on the non-image data determined by the majority logic unit.
The receiver according to claim 5 , further comprising a correction unit. 8. A two-dimensional pattern corresponding to a value of non-image data input from the outside is selected from a preset two-dimensional pattern group , and the selected two-dimensional pattern is repeated a plurality of times.
A pattern conversion unit that generates a repetitive repeat data, an image compression section that compresses the image data representing an image comprising a repeating data generated by the pattern conversion unit at a predetermined image compression method, is compressed by the image compression unit A transmitting unit for transmitting the image data, a receiving unit for receiving the compressed image data, an image decompressing unit for decompressing the received data of the receiving unit to restore the image data, and an image decompressing unit for decompressing the image data. from an image represented by image data, extract the repeated data generated with two-dimensional pattern is repeated a plurality of times included in the two-dimensional pattern group
When it is output, the repetitive data is added to the two-dimensional pattern.
Repeatedly dividing each pattern matching unit for outputting a plurality of non-image data and a plurality of non-image divided by the pattern matching unit of
One non-image data from image data by majority logic
And a majority logic unit for making a decision . 9. The transceiver according to claim 8 , wherein the image compression method is a lossy compression method. 10. An image data input unit for inputting image data, and an image data output unit for outputting the image data expanded by the image expansion unit, wherein the image compression unit is input by the image data input unit. Characterized in that image data representing an image obtained by arranging the repeated data generated by the pattern conversion unit in an unused area on the image represented by the generated image data is compressed by a predetermined image compression method. The transceiver according to claim 8 . 11. An error detection / correction encoding unit which performs error detection / correction encoding on non-image data input from the outside and supplies the error detection / correction encoding to the pattern conversion unit, and a non-determination unit determined by the majority logic unit. An error detection / correction unit for performing error detection / correction processing on image data, wherein the pattern conversion unit has a two-dimensional shape corresponding to the value of the non-image data supplied from the error detection / correction encoding unit. putter
Pattern from a preset two-dimensional pattern group,
Repeat the selected two-dimensional pattern multiple times to
The transceiver according to claim 8 , wherein the transceiver generates a data. 12. A data transmission system comprising the transmitter according to claim 1 and the receiver according to claim 5 . 13. A data transmission system comprising the transmitter according to claim 1 and the transceiver according to claim 8 . 14. A data transmission system comprising the receiver according to claim 5 and the transceiver according to claim 8 . 15. A data transmission system comprising at least two transceivers according to claim 8 .