JPH0955708A - Data compression and transfer system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data compressing means which displays an effect even on data wherein characters are not successive as to the data compression and transfer system of a data transfer system. SOLUTION: Of the data transfer system which transfers input data to a transmission side device 100 to a reception side device 200 through a transfer path 300, the transmission side device is provided with a repeated part deleting means 101 which retrieves and deletes partial data appearing plural times in the input data from the input data, a reproduction information generating means 102 which generates reproduction information indicating the use range of the data after deletion when the input data are reproduced, and a reproduction information adding means 103 which adds the reproduction information to the data after deletion and transfers them to the reception side device 200 through the transfer line 300. The reception side device 200 is provided with a reproduction information separating means 201 which separates the reproduction information from the received data after deletion and a reproducing means 202 which generates repeated partial data from the data after deletion according to the reproduction information to reproduce the input data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data compression transfer method in a data transfer system, and more particularly to a data compression system in a data transfer system for transferring input data input to a transmission side device to a reception side device via a transfer path. Regarding transfer method.

[0002]

2. Description of the Related Art FIG. 11 is a diagram illustrating a data transfer system to which the present invention is applied, FIG. 12 is a diagram illustrating a conventional processing device, and FIG. 13 is a data compression situation in FIG. It is a figure.

FIG. 11 illustrates a remote maintenance system for an exchange as a data transfer system to which the present invention is applied. In FIG. 11, the plurality of exchanges (1) are respectively provided with a speech path device (2), a processing device (3) and a file memory (4), and are not shown in the figure and accommodated in the speech path device (2). Switching connection is established between the subscribers under the control of the processing unit (3).

The file memory (4) stores system files and station data necessary for the processor (3) to execute the exchange process, and operation data for recording the operation result of the exchange (1). ing.

Each exchange (1) is connected to a remote maintenance device (6) via a communication line (5). The remote maintenance device (6) comprises a processing device (7) and a file memory (8), and in order to monitor the operation status of each exchange (1), the file memory (4) of each exchange (1). Operation data stored in the communication line (5)
The updated system file stored in the file memory (8) is communicated in order to collect it via the network, store it in the file memory (8), and update the system file used by each exchange (1). Download to the file memory (4) of each exchange (1) via the line (5).

When the processing device (3) of each exchange (1) receives the operation data transfer command from the remote maintenance device (6), it reads the operation data from the file memory (4), and the communication path device (2) and When it is transferred to the remote maintenance device (6) via the communication line (5) and a system file download command is received from the remote maintenance device (6), the remote maintenance device (6) passes through the communication line (5). It is necessary to receive the system file that is transferred in the following manner and store it in the file memory (4).

The processing device (3) is equipped with a data transfer system as shown in FIG. 12 in order to reduce the load required to transfer the operational data or the system file as much as possible and prevent the replacement processing capacity from being pressed. Was there.

In FIG. 12, a file reading section (31)
Reads out data to be transferred (for example, pre-compression data shown in FIG. 13A) such as designated operation data from the file memory (4) and transmits it to the zero compression unit (32).

In the pre-compression data shown in FIG. 13A, the character (A) is arranged at the 0th position, the character (B) is arranged at the 1st position, and then the second to 51st characters are arranged. 50 characters (0) are arranged consecutively, the character (C) is arranged at the 52nd position, the character (0) is arranged at the 53rd position, and the character (D) is arranged at the 54th position. There is.

The zero compression section (32) arranges the input pre-compression data other than the character (0) in the same compressed data, but only the character (0) becomes one character (0). , And the number of characters to be arranged to create compressed data as shown in FIG. 13 (b).

As a result, the 0th character (A) of the uncompressed data
The first character (B) and the first character (B) are arranged as the 0th character and the 1st character of the same compressed data, but the 50th character (0 ) Is converted into one character (0) and one character (50) and arranged in the second and third characters of the compressed data, respectively.

The 52nd character (C) of the uncompressed data is
The fourth character of the compressed data is arranged, but the 53rd character (0) of the uncompressed data is converted into a single character (0) and a character (1), and the fifth character of the compressed data is obtained. Arrange to the character and the 6th character.

Finally, the 54th character (D) of the uncompressed data
Is arranged in the seventh character of the compressed data. As described above, the zero compression unit (32) compresses the uncompressed data composed of 55 characters into the compressed data composed of 7 characters, and then transmits the compressed data to the transmission unit (33).

The transmission unit (33) sends the compressed data transmitted from the zero compression unit (32) to the communication line (5) via the communication path device (2) and sends it to the remote maintenance device (6). Forward.

On the other hand, the processing device (7) in the remote maintenance device (6) reads out data to be transferred such as a system file [for example, uncompressed data shown in FIG. 13 (a)] from the file memory (8). 1, the same zero compression processing as the zero compression section (32) in the processing device (3) is performed.
It is assumed that the compressed data shown in 3 (b) is created and transferred to the exchange (1) via the communication line (5).

In the exchange (1), the processing device (3)
When the receiving unit (34) therein receives the compressed data arriving from the communication line (5) via the communication path device (2),
It is transmitted to the zero reproducing unit (35).

The zero reproducing section (35) arranges the compressed data transmitted from the receiving section (34) except the character (0) in the normal reproducing data, but only the character (0) is converted into the character ( Characters (0) are consecutively arranged by the number indicated by the character arranged following (0) to reproduce the same reproduction data as the pre-compression data as shown in FIG. 13 (a).

As a result, the 0th character (A) of the compressed data
And the first character (B) are arranged as the 0th character and the 1st character of the normal reproduction data, but the 2nd character of the compressed data is arranged.
The character (0) is continuously arranged by the number indicated by the following third character (50) from the second to the 51st of the reproduction data.

The fourth character (C) of the compressed data is arranged in the 52nd character of the normal reproduction data, and the fifth character (0) of the compressed data is indicated by the following sixth character (1). Arrange as many as the 53rd character of the reproduction data.

Finally, the seventh character (D) of the compressed data is
It is arranged in the 54th character of the normal reproduction data. As described above, the zero reproducing unit (35) reproduces the compressed data composed of 7 characters into the uncompressed data composed of 55 characters which is the same as the uncompressed data, and then the file writing unit (36). Communicate to.

The file writing unit (36) stores the reproduction data transmitted from the zero reproduction unit (35) in the file memory (4).

[0022]

As is clear from the above description, in the conventional data transfer system, the character (0) in which the zero compression section (32) is continuous is converted into a character (0) and a character (0). Since it was converted to the number of consecutive characters of 0),
For example, when the number of consecutive characters (0) is large, such as 50, the compression effect is large, but when the number of consecutive characters is small, such as 1, the number of characters is rather increased and there is a problem that the opposite effect occurs.
If the character (0) does not appear in the data,
There was a problem that the compression effect could not be exhibited at all.

It is an object of the present invention to realize a data compression means which can exert an effect even on data in which characters (0) are not continuous.

[0024]

FIG. 1 is a diagram illustrating the principle of the present invention. In FIG. 1, reference numeral 100 is a transmission side apparatus, 200 is a reception side apparatus, and 300 is a transfer path, which constitutes a data transfer system.

Reference numeral 101 denotes a transmitting side device (10
It is a duplicated portion deleting means provided in 0). 102 is
The reproduction information creating means is provided in the transmission side device (100) according to the present invention.

Reference numeral 103 denotes a transmitting side device (10
0) is a reproduction information adding means. 201 is
The reproduction information separating means is provided in the receiving side device (200) according to the present invention.

Reference numeral 202 denotes a transmitting side device (10
It is a reproducing means provided in 0). The overlapping part deleting means (101) searches for partial data that appears multiple times in the input data, deletes the detected overlapping part data from the input data, and creates deleted data.

The reproduction information creating means (102) creates reproduction information indicating the use range of the deleted data in order to reproduce the input data from the deleted data created by the overlapping portion deleting means (101).

The reproduction information adding means (103) adds the reproduction information created by the reproduction information creating means (102) to the deleted data after the overlapping part deleting means (101) deletes the overlapping part, and transfers the transfer path. Transfer to the receiving device (200) via (300).

The reproduction information separating means (201) separates the reproduction information from the deleted data transferred from the transmission side device (100) via the transfer path (300). Reproduction means (2
02) analyzes the reproduction information separated by the reproduction information separating means (201), thereby deleting the overlapping portion deleting means (10).
The overlapping portion data deleted in 1) is created from the deleted data, and the input data is reproduced.

The reproduction information creating means (102) arranges the reproduction information, in order to reproduce the input data, the start position of the usage range and the data length of the usage range in the deleted data in the order of reproduction of the input data. And then generated.

Further, the overlapping part deleting means (101) sequentially extracts the unit data constituting the input data from the unit data having the smallest sequence number, and the unit having the sequence number smaller than the sequence number of the extracted unit data. It is considered to search for the presence or absence of duplication of the extracted unit data by sequentially comparing the data as unit data to be compared for duplication in descending order of sequence numbers.

Further, it is considered that the overlapping part deleting means (101) limits the number of unit data to be compared with the extracted unit data within the range of a predetermined threshold value. Further, when the overlapping portion deleting means (101) detects the unit data to be compared that overlaps with the extracted unit data, the unit data having a larger order number than the extracted unit data and the unit data to be overlapped are compared. It may be considered to compare unit data having a larger sequence number with each sequence number in ascending order to search for the presence or absence of duplication of a plurality of consecutive unit data.

The overlapping portion deleting means (101) determines the number of continuous unit data from a predetermined threshold value when detecting continuous unit data to be compared, which overlaps with a plurality of continuously extracted unit data. Only when it is large, it is considered to delete the unit data in which duplicates are continuously detected.

The overlapping portion deleting means (101) detects a plurality of sets of continuous unit data to be compared, which overlaps with a plurality of continuously extracted unit data, and detects the number of continuous unit data of each detected set. In comparison, it is considered that the set having the maximum number of consecutive unit data is the target of duplicate detection.

Further, the transmission side device (100) has a transfer path (3
00) for monitoring the congestion state and controlling the operation / stop of the duplicated portion deleting means (101) according to the degree of congestion of the transfer path (300), or the input data to the duplicated portion deleting means (101). It is considered that the number of unit data is changed or the duplicate search range of the duplicate part deleting means (101) is changed.

Further, the transmission side device (100) monitors the congestion state of the duplicated portion deleting means (101), and operates the duplicated portion deleting means (101) according to the congestion degree of the duplicated portion deleting means (101). It is considered to control the stop or change the unit data number of the input data to the duplicated portion deleting means (101) or change the duplicate search range of the duplicated portion deleting means (101).

Therefore, even for input data which does not include the continuous character "0", it is possible to compress the transfer data by searching for and deleting the overlapping unit data, and the degree of congestion of the data transfer system. The degree of compression can be changed according to the above, and the data transfer efficiency of the data transfer system can be greatly improved.

[0039]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below with reference to the drawings. 2 is a diagram illustrating a processing device according to an embodiment of the present invention, FIG. 3 is a diagram illustrating a compression processing process in FIG. 2, and FIG. 4 is a diagram illustrating an index data editing processing process in FIG. 5, FIG. 5 is a diagram illustrating the reproduction processing process in FIG. 2, FIG. 6 is a diagram illustrating the response time monitoring processing process in FIG. 2, and FIG. 7 is an example of the compression processing time monitoring processing process in FIG. 8 and FIG.
2 is a diagram showing an example of the compressed data structure in FIG. 2, FIG. 9 is a diagram showing an example of the data compression and reproduction state (part 1) according to FIG. 2, and FIG. 10 is a diagram showing the data compression and reproduction state according to FIG. 2 (part 2). FIG. The same reference numerals indicate the same objects throughout the drawings. The target data transfer system is as shown in FIG.

In FIG. 11, a processing device (3) is provided as the transmission side device (100) and the reception side device (200) in FIG. 1, and the communication line (5) is provided as the transfer path (300) in FIG. ) Is provided.

FIG. 2 shows a data transfer system provided in the processing device (3) shown in FIG. 11, which includes a file reading unit (31), a transmitting unit (33) and a receiving unit (34). The file writing unit (36) is shown in FIG.
1, the duplicated portion search unit (371) and the text data memory (372) provided in the compression unit (37) correspond to the duplicated portion deleting unit (101) in FIG. , Again, index data editor (373)
The index data memory (374) corresponds to the reproduction information creating means (102) in FIG. 1, and the transfer data assembling section (375) further includes the reproduction information adding means (10) in FIG.
It corresponds to 3).

The transfer data decomposing section (381) provided in the reproducing section (38) corresponds to the reproduction information separating means (201) in FIG. 1, and the reproducing section (382), working memory (383) and The reproduction data memory (384) corresponds to the reproduction means (202) in FIG.

In FIG. 2, of the processing device (3) shown in FIG. 11, only the portion related to the compression and reproduction processing is shown, and the other portions are omitted. First, a compression processing process according to an embodiment of the present invention (claims 1 to 6) will be described with reference to FIGS. 2, 3, and 8 to 10.

2, FIG. 3, and FIG. 8 to FIG. 10, the file reading unit (31) reads out the specified data to be transferred (hereinafter referred to as an input data block) from the file memory (4) and compresses it. To the section (37).

The input data block (X ₁ ) is shown in FIG.
, M characters (A _i ) [where i is the character number in the input data block (X ₁ ) and is 1 to M-
1 represents an arbitrary number].

For example, the input data blocks illustrated in FIGS. 9 (a) and 10 (a) are both 36 [that is, i = 3].
6] is composed of English letters. The compression unit (37)
In the process shown in FIG. 3, the compression process of the input data block (X ₁ ) is executed.

In FIG. 3, steps S32 to S3B
In the [loop 1] including, the work data (i) is stepped from (0) to (M-1), and each character (A _i ) forming the input data block (X ₁ ) is addressed to one. In the [loop 2] which is to be extracted and subjected to compression processing, and which includes steps S33 to S3A, the work data (n) is stepped from (1) to (i), and is extracted so far. The extracted characters (A _(in) ) are sequentially extracted, and the extracted characters (A _(in) ) that match the characters (A _i ) extracted in loop 1 are searched.

The work data (n) in the [loop 2] is provided with an upper limit [hereinafter referred to as the search range (S)] to prevent the search time, and thus the compression processing time, from being extended indefinitely. are doing.

The [loop 3] including steps S37 to S39 is the work data (m) when the extracted character (A _(in) ) matching the character (A _i ) is detected by the [loop 2]. ) By sequentially stepping from (1),
Consecutive characters in the character _{(A i) (A i +} m) is, searches whether matches the consecutive characters to extract already letters _{(A (in)) (A} (in + m)) The work data (m) is
This corresponds to the number of characters (A _i ) to (A _{(i + m)} ) for which a match is continuously detected.

Even when a continuous matching character string of the number of characters (m) is detected, additional information [index data (X ₂ ) described later, FIG. ), The compression effect cannot be obtained unless the number of characters is sufficiently larger. Therefore, the numerical value for determining the presence or absence of the compression effect [hereinafter referred to as the compression effect determination number (P)] is set in advance and the number of consecutive characters is determined. After confirming whether (m) is equal to or more than the compression effect determination number (P) [step S3C in FIG. 3], it is determined whether or not the compression processing is executed.

When it is determined that the compression processing is not executed,
Is the input data block (X₁) Character extracted from (A
_i) Is transferred to the receiving side (hereinafter referred to as text data).
(X _Three), Refer to FIG. 8 (b)]
(B_j) [However, j is the text data (X_Three) Character number
Stored in the text data memory (372)
[FIG. 3 Step S3F].

If it is determined that the compression process is to be executed, the character (A) extracted from the input data block (X ₁ )
_i ) is not stored in the body data memory (372) as characters (B _j ) forming the body data (X ₃ ), and index data (X ₂ ) for notifying the receiving side of the compression range (FIG. 8 (c)) Index data memory (37) as a character number (j) [hereinafter referred to as a start position (C _kS )] and a continuous character number (m) [hereinafter referred to as a column length (C _kL )] of the first character constituting the
4) (step S3D in FIG. 3).

Similar index data (X ₂ ) is created for the body data (X ₃ ) and the index data memory (3
74) (FIG. 3, step S3F). When the compression processing from the input data block (X ₁ ) to the index data (X ₂ ) and the body data (X ₃ ) by the duplicated portion search unit (371) is completed, the transfer data assembly unit (375)
The body data (X ₃ ) stored in the body data memory (372) and the index data (X ₂ ) stored in the index data memory (374) are extracted, and transfer data (X ₃ ) shown in FIG. Assemble X ₄ ).

In FIG. 8 (d), the 0th character (D ₀ ) of the transfer data (X ₄ ) is a compression flag ( _FCMP ) indicating the presence / absence of compression processing [ _FCMP = logical "1" when compression is executed). , C _CMP = logical “0”] when compression is not executed, the first character (D ₁ ) indicates the start position (H _X2 ) of the index data (X ₂ ), and the second character (D ₂ ) indicates Indicates the column length (L _X2 ) of the index data (X ₂ ), the third character (D ₃ ) indicates the start position (S _X3 ) of the body data (X ₃ ), and the fourth character (D ₄ ).
Indicates the column length (L _X3 ) of the body data (X ₃ ), and the index data (X ₂ ) and the body data (X ₃ ) are arranged after the fifth character (D ₅ ).

The above compression process is shown in FIG. 9 and FIG.
The input data block shown in FIG.
First, the input data block (X ₁ ) shown in FIG.
Is input to the compression unit (37).

The duplication part search unit (371) sequentially extracts the 0th character (A ₀ = A) to the 10th character (A ₁₀ = K) of the input data block (X ₁ ) and repeats the extracted character. As a result of searching for, none of them match the extracted characters,
In any case, the body data memory (37) is used as the 0th character (B ₀ = A) to the 10th character (B ₁₀ = K) of the body data (X ₃ ).
2) and stores the index data (X ₂ ) at the 0th start position (C _0S = (0)) and the 0th column length (C _0L = (1
1)) is stored in the index data memory (374).

Next, the eleventh input data block (X ₁ )
When a character (A ₁₁ = K) is extracted and a duplicate extracted character is searched for, the 10th character (A
_When it is detected that ₁₀ = K), the 11th character (A
_11th character (A ₁₂ = K) and subsequent 11th character (A ₁₀ = K) and 11th character (A ₁₀ = K)
₁₁ = K) and the subsequent characters are sequentially compared, and the 25th character (A ₂₅ = K)
Up to the 24th character (A ₂₄ = K), the 26th character (A ₂₄
It is detected that ₂₆ = L) does not match the 25th character (A ₂₅ = K) [m = 15], and the number of characters [m = 15] is larger than the number of compression effect judgments (P) and there is a compression effect. , The 11th character (A ₁₁ = K) to the 25th character (A ₂₅
= K) is not stored in the body data (X ₃ ), but the first start position (C _1S = (10)) and the 0th position of the index data (X ₂ ) are stored.
Only the column length (C _1L = 15) is stored in the index data memory (37
4).

Subsequently, the duplicated portion search unit (371) sequentially extracts the 26th character (A ₂₆ = L) to the 35th character (A ₃₅ = U) of the input data block (X ₁ ) and all have been extracted. Since they do not match the characters, they are stored in the body data memory (372) as the 11th character (B ₁₁ = L) to the 20th character (B ₂₀ = U) of the body data (X ₃ ) and the index data ( The second start position (C _1S = (26)) and the first column length (C _1L = 10) of X ₂ ) are stored in the index data memory (37).
4).

Next, the transfer data assembling unit (375)
From the data memory (372), the text data (X_Three) Extract
The index data from the index data memory (374).
(X₂) Is extracted and the compression flag (F_CMP= Logic
"1"), index data (X₂) Start position (H_X2= 5)
And column length (L_X2= 12), body data (X_Three) Beginning
Position (H _X3= 17) and the column length (L_X3= 21) is added
Then, transfer data (X_Four) Assembled
And transmits it to the transmitter (33).

As described above, when the same character (K) continues by searching for a match with the extracted character (A ₁₁ ) in descending order of the character number (i) from the character (A ₁₀ ) that has been extracted immediately before. In addition, the search process is greatly simplified.

Next, it is assumed that the input data block (X ₁ ) shown in FIG. 10 is input to the compression section (37). The duplication part search unit (371) uses the input data block (X ₁ )
The 0th character (A ₀ = A) to the 10th character (A ₁₀ = K) are sequentially extracted, and as a result of searching for duplicate extracted characters, none of them match with the extracted characters. The 0th character (B ₀ = A) to the 10th character (B) of the data (X ₃ ).
₁₀ = K) is stored in the body data memory (372) and the 0th start position (C _0S =) of the index data (X ₂ ) is stored.
0) and the 0th column length (C _0L = 11) are stored in the index data memory (374).

Next, the input data block (X₁) Of 11
Letter (A₁₁= A) and search for duplicate extracted characters
As a result, the 0th character (A₀
= A), the 11th character (A₁₁
= A) and the 12th character (A ₁₂= B) or later and comparison target
0th character (A₀= A) followed by the first character (A₁=
B) and subsequent ones are sequentially compared, and the 14th character (A₁₄Up to = D)
Third character (A_Three= D), the 15th character (A_Fifteen=
H is the fourth character (A_Four= E)
[M = 4] and the number of characters [m = 4] is the compression effect determination number
When it is confirmed that the compression effect is larger than (P),
11th character (A₁₁= A) to the 14th character (A₁₄= D)
Body data (X_Three), The index data (X₂)of
First start position (C_1S= 0) and the first column length (C_1L= 4)
Only stored in the index data memory (374).

Subsequently, the duplicated portion search unit (371) sequentially extracts the 15th character (A ₁₅ = H) to the 18th character (A ₁₈ = K) of the input data block (X ₁ ) and performs the same process as described above. Through the process, it is detected that the extracted 7th character (A ₇ = H) to the 10th character (A ₁₀ = K) are respectively matched, and the compression effect is confirmed. Then, the 15th character (A ₁₅ = H) is detected. Through first
Eight characters (A ₁₅ = K) are not stored in the text data (X ₃ ),
Only the second start position (C _2S = 7) and the second column length (C _2L = 4) of the index data (X ₂ ) are stored in the index data memory (37
4).

Subsequently, the duplicated portion search unit (371) sequentially extracts the 19th character (A ₁₉ = L) to the 22nd character (A ₂₂ = O) of the input data block (X ₁ ) and performs the same processing as described above. According to the process, none of the characters match the extracted characters, so that none of the 11th character (B ₁₁ = L) to 14th character (B ₁₄ = O) of the body data (X ₃ ) is stored in the body data memory (37).
2), the third start position (C _3S = 19) and the third column length (C _3L = 4) of the index data (X ₂ ) are stored in the index data memory (374).

Subsequently, the duplication part search unit (371) sequentially extracts the 23rd character (A ₂₃ = H) to the 26th character (A ₂₆ = K) of the input data block (X ₁ ) and performs the same processing as described above. By the process, it is detected that the extracted 15th character (A ₁₅ = H) to the 18th character (A ₁₈ = K) respectively match, and the compression effect is confirmed, the 23rd character (A ₂₃ = H) The 26th character (A ₂₆ = K) is not stored in the body data (X ₃ ) but the fourth start position (C _4S = 15) of the index data (X ₂ ).
And only the fourth column length (C _4L = 4) is stored in the index data memory (374).

Subsequently, the duplication part search unit (371) sequentially extracts the 27th character (A ₂₇ = A) to the 30th character (A ₃₀ = D) of the input data block (X ₁ ) and performs the same processing as described above. By the process, it is detected that the extracted 11th character (A ₁₁ = A) to 14th character (A ₁₄ = D) match, respectively, and the compression effect is confirmed, the 27th character (A ₂₇ = A) Thru the 30th character (A ₃₀ = D) is not stored in the body data (X ₃ ) but the fifth start position (C _5S = 11) of the index data (X ₂ ).
And only the fifth column length (C _5L = 4) is stored in the index data memory (374).

Further, the duplication part search part (371) uses the 31st character (A ₃₁ = E) through the 3rd character of the input data block (X ₁ ).
Five characters (A ₃₅ = I) are sequentially extracted, and the extracted fourth character (A ₄ = E) through eighth character (A ₄ ) are extracted by the same process as described above.
₈ = I) and the compression effect is confirmed, the 31st character (A ₃₁ = E) to the 35th character (A ₃₅ = I) are not stored in the text data (X ₃ ). , The sixth start position (C _6S = 4) of the index data (X ₂ ) and the sixth position
Only the column length (C _6L = 5) is stored in the index data memory (374)
To be stored.

As a result of the above, the index data memory (374)
The index data (X ₂ ) stored in the
_0S ) to the sixth start position ( _C6S ) and the 0th column length ( _C0L ) to the sixth column length ( _C6L ), which is a total of 14 characters.

Next, an index data editing process according to the embodiment of the present invention (claim 7) will be described with reference to FIGS. 2, 3 and 4. In the embodiment of the present invention (claim 7), after the step S3D in FIG. 3, the index data editing process for executing the process shown in FIG. 4 [step S3 in FIG.
3E] is inserted instead of step S3D _A.

In FIG. 4, steps S43 to S44.
[Loop 4] including "," the start position (C ₂ ) of the index data (X ₂ ) when the duplicate character string is detected by Loop 3 in FIG.
_kS ) and column length (C _kL ) are searched for another character string starting with the same character among the character strings already stored, and the steps S47 and S48 are included in [loop 5]. , [Loop 4] is to confirm the continuous match between the other extracted character string detected by [loop 4] and the target character string, and steps S49 to S4B are performed when comparing with other extracted character strings. When the number of consecutive matching characters (m ') is larger than the number of consecutive matching characters (m) that have been compared, the corresponding start position (C _kS ) and column length (C _kL ) of the index data (X ₂ ) are updated. is there.

The above-described index data editing process is shown in FIG.
Of the index data (X ₂ ) shown in, the 27th character (A ₂₇
= A) to the 30th character (A ₃₀ = D). First, the duplicated part search part (371) detects the 27th character (A
₂₇ = A) to the 30th character (A ₃₀ = D) are sequentially extracted, and the extracted 11th character (A ₁₁ =
A) to the 14th character (A ₁₄ = D) are detected, and then the fifth start position (C _5S = 11) and the fifth column length (C _5L = 4) of the index data (X ₂ ) are detected. Is stored in the index data memory (374), and then the index data editing unit (373) is activated.

Index data editing unit (373) which is activated
Executes the index data editing process shown in FIG. 4 to first compare the 27th character (A ₂₇ = A) with the extracted 0th character (A ₀ = A) and confirm that they match. When confirmed [loop 4 in FIG. 4], the succeeding characters (A ₂₈ = B) of the 27th character (A ₂₇ = A) and the succeeding characters of the 0th character (A ₀ = A) (A ₁ = B) By sequentially comparing with the following,
It is detected that the characters up to the 35th character (A ₃₅ = I) and the 8th character (A ₈ = I) respectively match [loop 5 (m ′ in FIG. 4].
= 9)].

As a result, the number of consecutively matched characters of the 27th character (A ₂₇ = A) to the 30th character (A ₃₀ = D) and the 11th character (A ₁₁ = A) to the 14th character (A ₁₄ = D) From [m = 4], the 27th character (A ₂₇ = A) to the 35th character (A ₃₅ =
I) and the 0th character (A ₀ = A) to the 8th character (A ₈ = I)
Since the number of consecutively matched characters [m = 9] is large, the index data (X ₂ ) already stored in the index data memory (374)
The fifth start position (C _5S = 11) and the sixth start position (C _6S = 4), the fifth column length (C _5L = 4) and the sixth column length (C _6L = 5) 5 Update to start position (C _5S = 0) and fifth column length (C _5L = 9).

As a result, the index data (X ₂ ) stored in the index data memory (374) has the 0th start position (C _0S ) to the 5th start position (C _5S ) and the 0th column length (C _0L). ) To the fifth column length (C _5L ), a total of 12 characters, which is reduced by 2 characters as compared with the case where the index data editing unit (373) is not activated.

As described above, the transfer data assembling section (375)
The transfer data (X ₄ ) assembled by is as shown in FIG. Next, a reproducing process according to an embodiment of the present invention (claim 1) will be described with reference to FIGS. 2, 5, and 8 to 10.

In the reproducing process, contrary to the compressing process, the input data block (X ₁ ) shown in FIG. 8 (a) is reproduced from the transfer data (X ₄ ) shown in FIG. 8 (d). to those therefore referred to as a reproduced data block input data block _{(X 1) (X 1)} .

In FIG. 2, FIG. 5, and FIG. 8 to FIG. 10, the receiving unit (34) is connected to the communication line device (2) from the communication line (5).
Transfer data (X ₄ ) arriving at the processor (3) via
Is received, the compression flag (F
_CMP ) is analyzed, and when the compression flag (F _CMP ) is set to compression non-execution [= logical “0”], the transfer data (X ₄ ) is not transmitted to the reproducing unit (38), The data is directly transmitted to the file writing unit (36) and requested to be stored in the file memory (4), but if the compression flag ( _FCMP ) is set to compression execution [= logical "1"], reproduction is performed. To the section (38).

First, the transfer data decomposing unit (381) determines the head position (H _X2 ) added to the first character (D ₁ ) to the fourth character (D ₄ ) of the transferred transfer data (X ₄ ). , The column length (L _X2 ), the start position (H _X3 ) and the column length (L _X3 ), the transfer data (X ₄ ) is decomposed into the index data (X ₂ ) and the body data (X ₃ ), and the working memory is obtained. It is stored in (383).

Subsequently, the reproducing unit (382) executes the reproducing process in the process shown in FIG. 5 by using the index data (X ₂ ) and the body data (X ₃ ) stored in the working memory (383). To do.

In FIG. 5, steps S52 to S59.
[Loop 1] including the work data (k) from 0 to Q
[Where Q is the number of characters in the index data (X ₂ )], each starting position (C _kS ) and column length (C _kL ) are extracted from the index data (X ₂ ) as a set, and the reproduction process is performed. In [loop 2] including steps S53 to S58, work data (i) and (j) are stepped, and work data (n) is changed from 1 to the column length (C). _By stepping up to _kL ), and comparing the reproduced character position (i) with the start position (C _kS ), the character (A _i ) in the reproduced data block (X ₁ ) is converted into the body data (X ₃ ). From [Step S5
6, S57] or from the reproduction data block (X ₁ ) already stored in the reproduction data memory (384) [step S5
4, S55] to be extracted.

The above reproduction processing steps are shown in FIG. 9 and FIG.
The description will be made with reference to the index data and body data shown in. First, it is assumed that the index data (X ₂ ) and the body data (X ₃ ) shown in FIG. 9 are stored in the working memory (383).

The reproducing section (382) uses the index data (X ₂ )
0th start position (C _0S = 0) and 0th column length (C _0L = 1)
1), based on the reproduction data blocks (X ₁₎ in the reproduction character number (i = 0), the 0 character (B ₀ = A) to 10 characters (B ₁₀ = K the body data (X ₃₎ ) Are sequentially extracted, and the 0th character (A ₀ =) of the reproduction data block (X ₁ ) is extracted.
A) to the tenth character (A ₁₀ = K) are stored in the reproduction data memory (384).

Then, the reproducing unit (382) displays the index data.
(X₂) First start position (C_1S= 10) and the first column length
(C_0L= 15) and the reproduction data block (X₁) Re in
Playback data block based on the raw character number (i = 11)
Ku (X₁) First start position (C _1S= 10) [that is, the text
Data (X_Three) 10th character (B_Ten= K)], the first column length
(C_1L= 15) repeated for the number of characters (= 15)
Return and extract and play data block (X₁) 11th character
(A₁₁= K) through the 25th character (A_{twenty five}= K), play
The data is stored in the data memory (384).

Then, the reproducing unit (382) receives the second start position (C _2S = 26) and the second column length (C _2L = 10) of the index data (X ₂ ) and the reproduction data block (X ₁ ). Based on the reproduction character number (i = 26) of the body text data (X ₃ ) from the 11th character (B ₁₁ = L) to the 20th character (B).
₂₀ = U) are sequentially extracted and stored in the reproduction data memory (384) as the 26th character (A ₂₆ = L) to the 35th character (A ₃₅ = U) of the reproduction data block (X ₁ ).

As described above, in the reproduction data memory (384), as shown in FIG. 9 (a), the input data block (X
_The same reproduction data block (X ₁ ) as in ₁ ) is reproduced,
It has been stored.

The file writing unit (36) stores the reproduction data block (X ₁ ) stored in the reproduction data memory (384) in the file memory (4). next,
It is assumed that the index data (X ₂ ) and the body data (X ₃ ) shown in FIG. 10 are stored in the working memory (383).

The reproducing section (382) uses the index data (X ₂ )
0th start position (C _0S = 0) and 0th column length (C _0L = 1)
1) and the reproduction character number (i = 0) of the reproduction data block (X ₁ ) based on the 0th character (B ₀ = A) to the 10th character (B ₁₀ = K) of the body data (X ₃ ). the extracted sequentially, the 0th character of the reproduced data blocks (X ₁₎ (a ₀ =
A) to the tenth character (A ₁₀ = K) are stored in the reproduction data memory (384).

Subsequently, the reproducing section (382) stores the first start position (C _1S = 0) and the first column length (C _0L = 4) of the index data (X ₂ ) and the reproduction data block (X ₁ ). Based on the reproduction character number (i = 11), from the first start position (C _1S = 0) of the reproduction data block (X ₁ ) to the first column length (C 1
_1L = 4) consecutive characters, that is, the 0th character (B ₀ = A) to the 3rd (character (B ₃ = D)) in the body data (X ₃ ) are continuously extracted and reproduced data block (X ₁ ) The 11th character (A ₁₁ = A) to the 14th character (A ₁₄ = D) are stored in the reproduction data memory (384).

Then, the reproducing unit (382) reproduces the second start position (C _2S = 7) and the second column length (C _2L = 4) of the index data (X ₂ ) and the reproduced data block (X ₁ ). Based on the reproduction character number (i = 15), from the second start position (C _2S = 7) of the reproduction data block (X ₁ ) to the second column length (C
_2L = 4) worth of continuous character, i.e. body data (X ₃₎ seventh character in (B ₇ = H) to 10 (characters (B ₁₀ = K) is continuously extracted, reproduced data blocks (X _It is stored in the reproduction data memory (384) as the 15th character (A ₁₅ = H) to the 18th character (A ₁₈ = K) of 1).

Subsequently, the reproducing section (382) stores the third start position (C _3S = 19) and the third column length (C _3L = 4) of the index data (X ₂ ) and the reproduced data block (X ₁ ). Based on the reproduction character number (i = 19), the 11th character (B ₁₁ = L) to the 14th character (B ₁₄ = 0) of the body data (X ₃ ) are sequentially extracted, and the reproduction data block (X ₁ ) is extracted. 19th character (A ₁₉ = L) through 22nd character (A ₂₂ = O) of the above are stored in the reproduction data memory (384).

Then, the reproducing unit (382) stores the fourth start position (C _4S = 15) and the fourth column length (C _2L = 4) of the index data (X ₂ ) and the reproduced data block (X ₁ ). Based on the reproduction character number (i = 23), continuous characters of the fourth column length (C _4L = 4) from the fourth start position (C _4S = 15) of the reproduction data block (X ₁ ), that is, the body data ( 7th character (B ₇ = H) to 10th character (character (B ₁₀ = K)) in X ₃ ).
Of the reproduced data block (X ₁ )
It is stored in the reproduction data memory (384) as 3 characters (A ₂₃ = H) to 26th character (A ₂₆ = K).

Subsequently, the reproducing unit (382) stores the fifth start position (C _5S = 0) and the fifth column length (C _5L = 9) of the index data (X ₂ ) and the reproduction data block (X ₁ ). Based on the reproduction character number (i = 27), from the fifth start position (C _5S = 0) of the reproduction data block (X ₁ ) to the fifth column length (C
_5L = 8) consecutive characters, that is, the 0th character (B ₀ = A) to the 8th (character (B ₈ = I)) in the body data (X ₃ ) are continuously extracted, and the reproduction data block (X ₁ ) The 27th character (A ₂₇ = A) to the 35th character (A ₃₅ = I) are stored in the reproduction data memory (384).

As described above, the input data block (X ₁ ) shown in FIG. 10 (a) is stored in the reproduction data memory (384).
That is, the same reproduction data block (X ₁ ) as that of ₁ is reproduced and stored.

The file writing unit (36) stores the reproduction data block (X ₁ ) stored in the reproduction data memory (384) in the file memory (4). next,
An embodiment of the present invention (claims 8 and 9) will be described with reference to FIGS. 2, 6 and 7.

The compression control data memory (3
In B), the data string length for determining whether compression is necessary [for example, 40
96 bytes and the like] and a search range (S) [for example, 128 bytes and the like] are stored.

In FIGS. 2, 6 and 7, the compression control data memory (3B) is used from the time when the transmitting unit (33) transfers the transfer data (X ₄ ) to the time when a response is returned from the receiving side. The response time (SC) is monitored [step S6 in FIG. 6].
1], the response time (SC) has a predetermined threshold value (B
When it is detected that L ₁ ) or more is exceeded [step S6
2], it is determined that the transmission process is congested and the compression effect is improved, so that the compression flag ( _FCMP ) is set to the compression execution [= logical "1"] in order to eliminate the congestion. The data length (L _CMP ) of the input data block (X ₁ ) read from the file memory (4) by the file reading unit (31) is expanded by a predetermined amount (for example, about 64 bytes), and the search range ( S) is expanded by a predetermined amount [for example, about 32 bytes] [step S63].

The compression control data memory (3B) is
The reduction unit (37) controls the input data block (X₁) Received
From the transfer data (X_Four) Output compression time
(CC) is monitored [step S71 in FIG. 7] and during compression processing
Interval (CC) is a predetermined threshold value (BL₂) With above
If it is detected that [step S72], the compression process
In order to determine that it is congested and reduce the compression processing load,
Compression flag (F_CMP) Is not executed [= logic "0"]
The file is read by the file reading unit (31) while setting
Input data block (X₁)
Data length (L _CMP) To a predetermined degree [eg 6
4 bytes, etc.], reduced, and search range (S) in advance
Reduce by a specified amount (for example, about 32 bytes)
[Step S73].

As is apparent from the above description, according to the present embodiment, each character (A in the input data block (X ₁ ) is composed.
_i ) is compared with a character (A _i ) having a smaller character number (i) and only the non-overlapping characters (B _j ) are included in the body data (X ₃ ).
And the index data (X ₂ ) is created by the start position (C _kS ) and column length (C _kL ) of each part of the body data (X ₃ ) required for reproducing the input data block (X ₁ ) and the index is created. Since the transfer data (X ₄ ) is assembled from the data (X ₂ ) and the body data (X ₃ ), it is possible to perform compression processing even on the input data block (X ₁ ) that does not include consecutive characters (0). .

Since the characters (A _i ) to be compared are sequentially searched in the descending order of the character numbers, the compression processing time can be shortened when duplicate characters are consecutive. Further, by limiting the search range (S) of the characters (A _i ) to be compared, the compression processing time can be shortened.

Further, the transmission section (33) or the compression section (37)
It is also possible to improve the data transfer efficiency of the data transfer system by controlling the block length and the compression processing range of the input data block (X ₁ ) according to the congestion status of the data transfer system.

2 to 11 are merely examples of the present invention until now. For example, the target input data is not limited to the data block (X ₁ ) in units of characters, and other However, the effect of the present invention does not change in any case. Further, the configurations of the compression section (37) and the reproduction section (38) are not limited to those shown in the figure, and various modifications may be considered, but in any case, the effect of the present invention is different. Absent. Further, it goes without saying that the data transfer system to which the present invention is applied is not limited to the exemplified switching network.

[0102]

As described above, according to the present invention, in the data transfer system, even for input data which does not include the continuous character "0", the transfer data can be transferred by searching for and deleting duplicate unit data. It becomes possible to compress,
Further, the compression degree can be changed according to the congestion degree of the data transfer system, and the data transfer efficiency of the data transfer system can be significantly improved.

[Brief description of drawings]

FIG. 1 is a principle diagram of the present invention.

FIG. 2 is a processing apparatus according to an embodiment of the present invention.

FIG. 3 is a compression process in FIG.

FIG. 4 is a process of editing index data in FIG.

FIG. 5 is a reproduction processing process in FIG.

FIG. 6 is a response time monitoring process in FIG.

FIG. 7 is a compression processing time monitoring processing process in FIG.

FIG. 8 is a compressed data structure in FIG.

FIG. 9: Data compression and reproduction status according to FIG. 2 (No. 1)

FIG. 10: Data compression and reproduction status according to FIG. 2 (Part 2)

FIG. 11 is a data transfer system to which the present invention is applied.

FIG. 12 Conventional processing device

FIG. 13 is a data compression state in FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 switchboard 2 speech path device 3, 7 processing device 4, 8 file memory 5 communication line 6 remote maintenance device 31 file reading unit 32 zero compression unit 33 transmission unit 34 reception unit 35 zero reproduction unit 36 file writing unit 37 compression unit 38 Reproduction unit 3A Monitoring unit 3B Compression control data memory 100 Transmission side device 101 Overlapping part deletion means 102 Reproduction information creation means 103 Reproduction information addition means 200 Receiving side device 201 Reproduction information separation means 202 Reproduction means 300 Transfer path 371 Duplication part search section 372 Text data memory 373 Index data editing unit 374 Index data memory 375 Transfer data assembling unit 381 Transfer data disassembling unit 382 Reproducing unit 383 Working memory 384 Reproducing data memory

Claims (9)

[Claims] 1. Input data input to a transmission side device,
In a data transfer system for transferring to a receiving side device via a transfer path, the transmitting side device searches for partial data that appears multiple times in the input data, and detects the duplicated partial data as the input data. And reproducing information indicating the use range of the deleted data in order to reproduce the input data from the deleted data created by the overlapping part deleting means. Reproduction information creating means for adding the reproduction information created by the reproduction information creating means to the deleted data after the overlapping part deleting means deletes the overlapping part, and receiving the data via the transfer path. Reproduction information adding means for transferring the reproduction information from the deleted data transferred from the transmission side device to the reception side device via the transfer path. And a reproduction information separation unit that separates the reproduction information separated by the reproduction information separation unit to create the overlapped portion data deleted by the overlapped portion deletion unit from the deleted data, and the input A data compression and transfer method, characterized in that a reproducing means for reproducing data is provided. 2. The reproduction information creating means, in order to reproduce the reproduction information from the input data, sets the start position of the usage range and the data length of the usage range in the deleted data to the input data. 2. The data compression transfer method according to claim 1, wherein the data compression transfer method is arranged and generated in a reproduction order. 3. The overlapping part deleting means sequentially extracts the unit data constituting the input data from the unit data having the smallest sequence number, and has a sequence number smaller than the sequence number of the extracted unit data. 2. The data compression transfer method according to claim 1, wherein the unit data to be compared for duplication is sequentially compared in descending order of the sequence numbers to search for the duplication of the extracted unit data. 4. The data according to claim 1, wherein the overlapping portion deleting unit limits the number of unit data to be compared with the extracted unit data within a range of a predetermined threshold value. Compressed transfer method. 5. The overlapping portion deleting means, when detecting the unit data to be compared that overlaps with the extracted unit data, compares the unit data having a larger sequence number than the extracted unit data with the overlapping comparison. 2. The data compression transfer according to claim 1, wherein the unit data having a larger sequence number than the target unit data is compared in ascending order of the respective sequence numbers, and the presence or absence of duplication of a plurality of consecutive unit data is searched. method. 6. The overlapped portion deleting means, when detecting unit data as a continuous comparison target that overlaps with a plurality of continuously extracted unit data, determines the number of continuous unit data from a predetermined threshold value. The data compression transfer method according to claim 1, wherein the unit data in which the duplicates are continuously detected is deleted only when the data is large. 7. The overlapping portion deleting means detects a plurality of sets of continuous comparison target unit data that overlaps a plurality of continuously extracted unit data, and detects the number of continuous unit data of each detected set. The pair which has the maximum number of consecutive unit data by comparison is targeted for duplicate detection.
Data compression transfer method described. 8. The transmission side device monitors the congestion state of the transfer path, controls the operation / stop of the overlapping part deleting means in accordance with the congestion degree of the transfer path, or deletes the overlapping part. 2. The data compression transfer system according to claim 1, wherein the number of unit data of input data to the means is changed, or the duplication search range of the duplication part deleting means is changed. 9. The transmission side device monitors the congestion state of the duplicated portion deleting means, controls the operation / stop of the duplicated portion deleting means in response to the degree of congestion of the duplicated portion deleting means, or 2. The data compression transfer method according to claim 1, wherein the number of unit data of the input data to the overlapping portion deleting means is changed, or the overlapping search range of the overlapping portion deleting means is changed.