KR100417825B1 - A method of abstracting gsmp message for atm exchange - Google Patents

Abstract

The present invention relates to a GSMP message for controlling an asynchronous transfer mode switch, and more particularly, to prevent transmission efficiency degradation and protocol error due to traffic congestion, and to a message compression transmission method for effectively transmitting a message. Compressing a null data area that is inserted unnecessarily to match a frame structure unit of data to be used, compressing bits that are not used and using only valid bits among the header message areas transmitted, and If you do not use all of them and mainly use small area, the data area is divided into many blocks of different sizes, and only the valid data area is transmitted, and the remaining area is compressed. The same contents are repeated among the transmitted message data. Store in cache memory If it is present, only the address and the cache hit area are transmitted. If the repeated message is not stored in the cache memory, the cache hit area is filled with null data and transmitted in addition to the repeated message. The region displaying the relevant serial number among the message data indicates a difference value of the serial number, and the remaining bits are configured to be compressed and transmitted.

Description

A METHOD OF ABSTRACTING GSMP MESSAGE FOR ATM EXCHANGE}

The present invention relates to a general switch management protocol (GSMP) message for controlling an asynchronous transfer mode (ATM) exchange, and in particular, when a large amount of data messages are transmitted, to reduce transmission efficiency and protocol error due to excessive traffic. The present invention relates to a message compression transmission method for effectively transmitting a message.

Asynchronous Transfer Mode (ATM) exchange is a fast (Asynchronous) mode of processing the data to be transmitted, so the data transfer rate is fast, and switching the high-speed data transfer path (Switch) as described above, High-speed data switching and transmission services such as real-time service and non-real-time service can be classified into two broad categories. Real-time services transmit data online such as voice data communication. Non-real-time service is a service that does not occur at the same time to transmit and receive data, such as E-mail transmission, FAX, Graphic transmission.

As described above, the asynchronous transfer mode exchanger controls and monitors the entire exchange from the internal main processing unit, while the operator controls and monitors the internal functions of the ATM exchange from the outside and updates the program of the internal main processing unit. In the event of a change in operating parameters or a failure, an external operating terminal is connected to prompt alarming and immediate confirmation, and the operation terminal and the internal main processor communicate with each other using the GSMP protocol.

Hereinafter, a method for transmitting a GSMP message according to an embodiment of the prior art will be described with reference to the accompanying drawings.

1 is a connection diagram of a general ATM switch and an operation terminal, and FIG. 2 is a diagram illustrating message states transmitted between a master and a slave of a GSMP.

Referring to the drawings, the connection of the general asynchronous transfer mode switch and the operating terminal, the operating terminal (10) consisting of a workstation (Work Station) computer for controlling and monitoring the ATM switch,

It consists of an asynchronous transfer mode (ATM) exchanger 20 for switching the transfer path of high-speed data processed in an asynchronous manner.

The operation terminal 10 is a GSMP master for message data communication with the ATM switch 20 under the control of the upper protocol 12 and the upper protocol 12 capable of controlling and operating the ATM switch. (14) is included,

The ATM switch 20 receives a message from the GSMP master 14, analyzes the message, and transmits the GSMP slave 24 and the entire ATM switch 20 that transmit the result to the GSMP master 14. A main processing device 22 for controlling and monitoring is included.

Hereinafter, a GSMP message transmission method will be described as an embodiment according to the related art having the above configuration.

The General Switch Management Protocol (GSMP) protocol, which has been developed and used for general purpose control of an asynchronous transfer mode (ATM) exchange, uses a format structure for simple message communication between the operation terminal 10 and the ATM exchanger 20. By transmitting and receiving message data, the GSMP master 14 of the operation terminal 10 receives a control signal received from the upper protocol 12 as an example, a message in a unit of 1,500 byte format. Size), and transmits the same to the GSMP slave 24 of the ATM switch 20.

The GSMP slave 24 reconfigures or converts a GSMP protocol message in units of 1,500 bytes received from the GSMP master 14 into a form that can be processed by the main processing unit 22, thereby converting the main processing unit ( 22).

When the main processing unit 22 analyzes and performs the authorized message, the GSMP slave 24 notifies the GSMP slave 24. The slave 24 converts the message into a message of the GSMP protocol format and then masters the message. By transmitting as a response or report message to (14), the master (14) is applied to an upper protocol, so that the operator can confirm the processing result of the control command through a display device not shown in the figure.

In the GSMP protocol message as described above, a call setup message for connecting a path through which high-speed data is transmitted, a configuration control message for checking and controlling the configuration of an ATM switch, and checking whether the connection of the call setup path is maintained. There are connection maintenance message, statistical information message by call setup in ATM switch, and status report message to check and monitor the operation status of ATM switch.It is formatted in 1,500 byte unit according to RFC 1987 GSMP Version 1.1 Spec. It is formatted and sent.

A first step of periodically transmitting and receiving a call setup or connection message and a connection maintenance message between the GSMP master 14 and the slave 24 by a GSMP protocol message as shown in FIG. In the second step, the slave 24 responds as a result message after processing, and if there is a change in the components of the ATM switch 20, i.e., the shape / state, the slave 24 may periodically The third step is to send a message to the master (14).

In the GSMP message transmission method of the prior art as described above, if a message to be transmitted or received between the master 14 and the slave 24 is temporarily congested, there is no separate function for controlling the flow of the message. A situation in which data transfer is temporarily congested occurs, and a path for transmitting a data message between the master 14 and the slave 24 has a data transfer rate or bandwidth. Since it is limited, data messages exceeding the bandwidth cannot be transmitted and thus discarded or lost.

In addition, the GSMP message transmission method according to the prior art is a simple format that does not divide and transmit or compress and transmit a message. When a large amount of data messages is generated, data may not be transmitted properly due to inefficiency. There is a problem that is lost, and there is a limit to the handling (Handling) of the operating system for normal message transmission, there was a serious problem that the error occurs in the operation of the protocol by processing the lost data message.

In the low bandwidth control channel or message transmission path or dedicated line that transmits and receives messages between the master 14 and the slave 24 as described above, when the traffic of data transmission is congested, the GSMP message transmission rate is not only lowered. In addition, since it affects the transmission speeds of other processors and other data messages that are configured in the ATM switch 20 and handle different functions, an important cause of a bottle neck phenomenon that lowers the efficiency of the entire ATM switch 20. It became.

The present invention controls the flow of GSMP messages transmitted and received through a control channel connected between the GSMP master and the slave, adjusts the size of the message, and compresses the message data to prevent congestion of the transmitted traffic, It is an object of the present invention to provide a GSMP message compression method that improves the performance of an ATM switch.

In order to achieve the above object, the present invention provides a first process of eliminating and compressing a null data area inserted unnecessarily in order to match a frame structure unit of data to be transmitted, and the actual transmission of the header message field. The second process of recompressing and compressing a field allocated to a data area wider than the data to the actual required size, and in the case of using a small area mainly without using all of the frame unit data area to be transmitted, the data is predetermined from the least significant bit. If the value of the last bit of the partitioned block and the value of the remaining data except the partitioned block are not the same, the partition encoding of the block encoding repeats partitioning the remaining data into smaller blocks until the same. In step 3 and in the message data transmitted, If the repeated and repeated messages are stored in the cache memory, only a cache hit area of a predetermined size indicating the corresponding address and cache transfer of the cache memory is transmitted; if the repeated messages are not stored in the cache memory, the repetition A fourth process of filling and transmitting the cache hit area with null data in addition to the corresponding message, and an area indicating an associated serial number among the transmitted message data indicates a difference value of the serial number and compresses the remaining bits. It is characterized by the GSMP message compression method consisting of a fifth process.

1 is a connection diagram of a general ATM switch and an operation terminal,

2 is a diagram illustrating message states transmitted between a master and a slave of a GSMP;

3 is a diagram illustrating a null data area of a GSMP call setup message according to an embodiment;

4 is a diagram illustrating a compressible region of a GSMP header message according to an embodiment.

5 is a configuration diagram of message data divided into block encodings according to an embodiment;

6 is a detailed configuration diagram of message data divided into block encodings according to an embodiment;

7 is a flow chart dividing message data into block encoding.

** Explanation of symbols on the main parts of the drawing **

10: operation terminal 12: upper protocol

14: GSMP Master 20: ATM Switch

22: main processing unit 24: GSMP slave

31,312: null data 34: code field

Hereinafter, a method of compressing a GSMP message according to an embodiment of the present disclosure will be described with reference to the accompanying drawings.

FIG. 3 is a diagram illustrating a null data region of a GSMP call setup message according to an embodiment, FIG. 4 is a diagram illustrating a compressible region of a GSMP header message according to an embodiment, and FIG. 5 is an embodiment. FIG. 6 is a detailed diagram illustrating message data divided into block encoding, and FIG. 7 is a flowchart illustrating dividing message data into block encoding.

Referring to the GPS message compression method according to the technique of the present invention with reference to all the accompanying drawings, a frame of GSMP (General Switch Management Protocol) message data transmitted between the GSMP master 14 and the slave 24 ( Frame) unit configuration, in one embodiment, a first process of deleting or compressing a null padding data region inserted unnecessarily without meaning to fit each frame unit into 16 bits or 32 bits;

Of the fields of the GSMP protocol message data header transmitted between the GSMP master 14 and the slave 24, only valid bits indicating actual data values are used, and unnecessary bits that are not used are deleted. Or a second process of compressing,

When a small portion of the area is mainly used without using all of the frame unit areas of the GSMP protocol message data transmitted between the GSMP master 14 and the slave 24, the data area of the frame unit is increased. In another embodiment, a third process of Block Encoding is performed by dividing a block into a plurality of blocks that differ by 1 / N ** 2 times and transmitting only a valid data area and deleting or compressing the remaining areas;

If the same content is repeated among the GSMP protocol message data transmitted between the GSMP master 14 and the slave 24 and the repeated message is stored in the cache memory, the cache memory is a predetermined bit. Transmits only a cache hit area of a predetermined size indicating a corresponding address or an allocation number of and indicating a cache transfer as one bit, and repeating the repeated message if the repeated message is not stored in the cache memory. A fourth step of filling and transmitting a cache hit area including the predetermined bits with null padding data in addition to an actual corresponding message;

Among the GSMP protocol message data transmitted between the GSMP master 14 and the slave 24, a serial number display area consisting of a plurality of bits indicating serial numbers associated with each other is sequentially arranged. The difference value of the serial number by the equation is represented, and the remaining bits (Bit) are composed of a fifth process of transmitting by deleting or compressing.

Particularly, block encoding according to the third process is performed by dividing data in units of frames into blocks of a predetermined size of N bits in reverse order from least significant bits (LSB). 1 step (S10),

A second step of determining whether all the bit signals of the last signal and the remaining signals except the signal divided into the block size are the same digital signal among the bit signals divided into the predetermined size block in the first step S10 (S20). )Wow,

In the case where the digital signal of the last bit Bit and all the remaining bits Bit that are divided in the second step S20 is not the same, the indication that the digital bit is added to the block and is not identical to the display bit consisting of one bit In one embodiment, a third step S30 of displaying '1',

A fourth step (S40) of dividing the remaining undivided digital signal into a block composed of one-half the size of the block from the least significant bit LSB and then feeding back to the second step (S40) )Wow,

In the second step S20, when the last bit of the bit divided into blocks and the digital signals of the remaining non-divided bits are all the same, the indication is equal to the display bit added in units of one bit. '5 and then ends (S50).

Hereinafter, the GPS message compression method of the present invention having the above configuration will be described in detail with reference to the accompanying drawings.

The message transmitted between the GSMP master 14 of the operating terminal 10 and the GSMP slave 24 of the ATM switch 20 is a call setup for connecting a data transmission path between the caller and the called party. Message, a port shape control message for controlling the switching port connection of ATM switch 20 and the functional units by each shape, and initial path connection establishment and path connection for communication between the GSMP master 14 and slave 24. Connection maintenance messages for exchanging messages for continuous connection maintenance, statistical information messages for transmitting various statistical data of the ATM switch 20 to the operation terminal 10, various operational states of the ATM switch 20, and It consists of a status report message that reports the status of a failure.

As described above, according to the RFC 1987-GSMP Version 1.1, the data communication protocol transmitted between the GSMP master 14 and the slave 24 includes data of 1,500 byte units forming one frame. It is a simple communication protocol that is not defined for the flow control or data partitioning and compression, and it is impossible to control the flow in which traffic is excessively increased. This may result in message loss in excess of.

In order to solve the above problems, the present invention relates to a method of compressing and transmitting a message transmitted in the GSMP protocol format and extracting at the receiving side.

In another embodiment, the compression method of the transmitted data is a general digital data compression method using the Huffman coding method, which is highly efficient, exhibits a constant abstraction rate for various kinds of data, and is very easy to apply. Although there is an advantage, the time required for compression and decompression is very long, such that the ATM network 20 and the operation terminal 10 are connected to online lines in real time, such that the data must be transmitted and received. ) Was not appropriate.

Of the data transmitted and received between the GSMP master 14 and the slave 24 as described above, in the call setup message as shown in FIG. 3, inserted without meaning as '0' to configure a frame unit. The null padding data areas 31 and 32 are deleted or compressed by the first process, and a wide data area is allocated as shown in the code field 34 of FIG. Is small, in one embodiment, 8 bits are allocated to transmit a value from 0 to 255, but in actual use, only 2 bits are used when only values of 0 to 3 are used. Since the code area 34 is reconfigured or encrypted, the 6-bit bit is deleted, compressed or reduced, and then transmitted (second process), and the receiving side has previously recognized and promised both sides of the transmission and reception. Compression or shaft Restore the exhausted bits.

The third process referring to the attached FIGS. 5 to 7 is data transmitted in units of frames, and in one embodiment, transmissions constituting one frame unit in 16 bits. Among the data, six bits (Bit) are designated and divided into blocks in order from the least significant bit (LSB) (S10).

As an example, referring to Example 1 (EX 1) of FIG. 6, the 16 bits of the frame unit are '0000000001001111', and 6 bits (Bit) are sequentially changed from the least significant bit (LSB) to the block. The specified and divided data is '111100'.

The 10 bits remaining after being divided into blocks are '0000000001', and the last bit of bits divided from the least significant bit (LSB) into blocks is '0'.

It is determined whether the last divided bit '0' and the remaining 10 bits of the block are signals of the same digital bit (S20).

As a result of the determination in the second step S20, the last bit divided into blocks is '0', and since the 10 bits remaining after the split have '1', it is determined that they are not the same.

As an example, also in Example 2 (EX 2) of FIG. 6, as the result, the remaining 10 bits are divided into '0000000011' and the last bit divided into blocks is determined to be not the same as '0'.

Therefore, in the third step, the display bit of one bit allocated to the next stage of the last bit divided into the six-bit block is not equal, and in one embodiment, '1' is displayed. (S30).

The remaining 10 bits are divided into small blocks corresponding to one-half of the block, that is, the least significant bit (LSB) into blocks of 3 bits (Bit) (S40).

For example, in the case of Example 1 (EX 1), the second divided signal is '100', and in the case of Example 2 (EX 2), it is '110'.

The last bits of Example 1 (EX 1) and Example 2 (EX 2) are each '0', are divided, and the remaining 7 bits (Bit) are each '0000000',

After performing the fourth step S40, since the feedback is fed back to the second step S20, the second divided last bit '0' and the remaining seven bits Bit are the same digital bit. (S20), and the same as '0', the display bit Bit allocated to one bit in the next stage of the second block is displayed as '0', which is the same as an embodiment ( S50) End.

One such embodiment is shown in detail in FIG. 6, and if the last bit divided and the remaining bits are not the same, subsequent partitioning blocks are divided into quarter blocks, 1/8 blocks, as shown in FIG. do.

By transmitting the data signal obtained by block-blocking only valid data by each step of the third process, it is possible to reduce the data transmission speed or bandwidth to be transmitted and received. In other words, instead of having to transmit 16 bits in a frame unit, only 11 bits need to be transmitted. Therefore, the compression effect of the message data is shown, and the data transmission rate is low, so that other data can be transmitted.

On the side of receiving such a signal, 16-bit message data in frame units is restored by recovering data in the reverse order and filling 7 untransmitted bits with '0'.

In the fourth process, the same content, which is frequently repeated in the transmitted frame unit, is extracted, stored in a cache memory of a transmitting side and a receiving side, and a predetermined bit corresponding to an address value of the cache memory. In addition, by allocating and transmitting one bit of using or not using cache memory, it is possible to increase the compression effect and data transmission efficiency, that is, bandwidth of the transmitted data.

As an example, if there are M messages that are often repeated identically, and the messages are 16 bits in size, there should be a cache memory area of M * 16 bits in both sides of the transmission and reception.

In one embodiment, when data of a particular frame is the same as that written in the cache memory, a bit indicating an address of the cache memory, in one embodiment, 4 bits and a 1-bit signal indicating that data in the cache memory is used. If only 5 bits are transmitted, the receiver receives the 5 bits, retrieves the corresponding address from the cache memory, and extracts and displays the data of the corresponding frame.

The formula of the bit representing the message contained in the cache memory as described above is as follows.

In the above formula, 'log ₂ N' is an expression for indicating the address of the cache memory. As one embodiment, if 10 frame data is recorded, it indicates that 4 bits are required, and the following '1' bits indicate one embodiment. For example, when cache memory is used, '1' representing a cache hit is displayed, and when not using cache memory, '0' represents a cache fail.

In the case of transmitting data not stored in the cache memory as described above, since the 'cache indication bit + data to be transmitted' must be transmitted, the traffic actually seems to increase by the 'cache indication bit', but the message data is repeated in general. Since there are many, the effects such as the compression (Abstract) effect of the transmission data and the improvement of the transmission rate or the expansion of the bandwidth.

When the cache memory is not used, all of the 'cache indication bits' are allocated with '0' and transmitted, and the data compression effect is the greatest in the technique of the present invention.

In the fifth process, the message data transmitted from the network is transmitted by adding a sequence number or a serial number and performing a transaction process.

The serial number or serial number as described above is a Locality Pattern. Since the first to fourth processes are difficult to extract, the difference value of the serial number or serial number is obtained by using an appropriate equation. Can only transmit.

As an embodiment, when 16 bits of 2 bytes are allocated to the serial number data field and the serial number of previous message data is '12974', the next message data is '12975'. Must be stored in the serial number data field.

However, only the difference value between the two serial numbers is expressed using the following simple equation.

f (x, y) = | x-y |

In the above equation, 'x' is a serial number value of previous message data, and 'y' is a serial number value of message data to be transmitted at this time.

When using a simple equation as in the above equation, the difference value between the two consecutive numbers is '1', and the difference value '1' needs only 1 bit (digit) as a digit, Compared with Bit, the result is that 15 bits are extracted.

Therefore, since the data transmitted using the GSMP protocol is efficiently compressed and transmitted between the ATM switcher 20 and the operation terminal 10, the data transmission rate or traffic is not increased or congested, and the bandwidth of the data transmission is increased. Get

The technique of the present invention as described above, by effectively compressing and transmitting the control data of the network transmitted in real time, thereby increasing the transmission rate of the real-time control data, increase the bandwidth of the data transmission, and improve the processing speed of the control data In addition, there is an industrial use effect to improve the smooth operation and reliability of the ATM exchange.

Claims (3)

A first process of eliminating and compressing a null data area that is unnecessarily inserted in order to fit a frame structure unit of transmitted data; A second process of reconfiguring and compressing a field allocated with a data area wider than data actually transmitted among header message fields to be actually required; In the case of mainly using a small area without using all of the frame unit data area to be transmitted, the data is divided into the blocks having the predetermined size from the least significant bit, and the last bit value of the divided block and the bits of the remaining data except the divided block. If the value is not the same, the third step of block encoding repeating the above operation of indicating this in the additional bits and dividing the remaining data into smaller blocks until the same, If the same content is repeated among the transmitted message data and the repeated message is stored in the cache memory, only the cache hit area of a predetermined size indicating the corresponding address and the cache transmission of the cache memory is transmitted, and the repeated message is the cache memory. A fourth process of filling the cache hit area with null data in addition to the repeated corresponding message if not stored in the second message; SMP message compression method characterized in that the area indicating the serial number of the message data to be transmitted indicates a difference value of the serial number and the remaining bits are compressed and transmitted. According to claim 1, The block encoding of the third process may include a first step of dividing data in a frame unit into blocks having a predetermined size from the least significant bit; A second step of determining whether the last bit signal divided into a block having a predetermined size in the first step and all the bit signals of the remaining data signals except the block size are the same digital signal; A third step of indicating that the display bits added to the block are not the same if they are not the same in the second step; A fourth step of dividing the remaining undivided signal into half the size of the block from the least significant bit and then feeding back to the second step; And if the same in the second step, the fifth step of indicating that the indication bit is the same. According to claim 1, In the asynchronous transmission mode switch, a method for compressing a message in the asynchronous transmission mode switch, characterized in that the data is transmitted after being compressed at the transmitting side is restored to the original state at the receiving side by the appointment previously applied to both sides.