JP2000339184A - Check slum computer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten a delay time due to check sum calculation on a transmission path at the time of using this check sum computer for error detection or the like at the time of data communication. SOLUTION: The total sum value of fixed partial fields are preliminarily stored in a reference data part 3, and the total sum value of the fixed parts are directly applied to a second adder 5 based on the judged result of a protocol judging part 1. Also, only variable partial field data are extracted by a field dividing part 2 based on the judged result of the protocol judging part 1, and supplied to a first adder 4, and the accumulated and added result of the first adder 4 is supplied to the second adder 5, and the total sum value of the fixed parts are added to the total sum value of the variable parts by the second adder 5. Thus, a time required for the check sum calculation can be shortened.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a checksum calculator, and more particularly to a checksum calculator that reduces the time required for a checksum calculation process.

[0002]

2. Description of the Related Art In recent years, data communication over a plurality of networks as seen on the Internet has become popular. In a communication protocol used for such data communication, data verification is performed using a checksum to check whether a bit error has occurred in communication data. For example, in the Internet Protocol (IP), a header portion is added every 16 bits, a one's complement is added as a checksum, and the receiving side adds and adds 16 bits including a checksum field. If the one's complement of the sum is taken, the value is 0 if there is no bit error such as a transmission error. Alternatively, if the result of the total addition value of the fields calculated including the header checksum field is 0XFFFF (FFFF in hexadecimal), it is determined that the checksum is established and that there is no bit error. It should be noted that some fields in the IP header are rewritten by the passing host, for example, the lifetime, so that the checksum calculation is re-executed at each passing host (gateway), and the header checksum is also rewritten.

FIG. 3 is a diagram showing a configuration example of an IP header in the Internet protocol. The IP header shown in this example has a bit configuration of an integral multiple of 4 bytes (32 bits), and the options and below are variable-length fields that are not required. In the conventional checksum calculation, the transmission source first sets the checksum field before calculation to 0 and sets the IP header to 2 bytes (16 bits).
It is sequentially added every time. At this time, if a carry is found in the addition result, 1 is added to the result. By repeating this, the value obtained by adding the one's complement of the result value added to all the header fields is written to the header checksum field and transmitted. The destination receiving the packet having the IP header performs the addition calculation including the header checksum field, and if the result is 0XFFFF, proceeds with the processing assuming that there is no bit error. Assuming that there is an error, processing such as discarding the received data and requesting retransmission is performed.

[0004] In this way, the checksum is sequentially calculated in the source terminal, the destination terminal, and each host passing between them.

[0005]

However, the above-mentioned conventional checksum calculator has the following problems. In other words, every time a packet arrives, the processing time becomes longer because the addition processing is performed every time for all the fields that are the checksum target of the arrived packet. There is a problem that the total transmission delay time is increased.

That is, in the communication protocol, a part of the parameters used for the header part is generally selected from a plurality of predetermined options in accordance with the actual operating environment in order to make the communication means and the communication system versatile. There is a field for selecting one, and such a field does not change during the operation of one communication, that is, during the period of communication started between the transmission source and the reception destination determined between them. Therefore, it is inefficient to perform the cumulative addition process each time for a fixed field portion (fixed portion) without change, and it is sufficient to perform the cumulative addition process only for the changing field portion (variable portion).

The present invention has been made to solve such a problem, and an object of the present invention is to provide a checksum calculator capable of shortening a checksum calculation processing time.

[0008]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, a checksum calculator according to the present invention is a calculator for a checksum used in data communication, and performs a protocol determination of the input data packet. A protocol determining unit, a field dividing unit that divides a field into a fixed part and a variable part based on the result from the protocol determining unit and extracts only the variable part, and a result of adding each data value of the fixed part in advance A reference data section storing a value, a first adder for sequentially adding the variable part data divided by the field division section, a sum of the addition result from the adder and a fixed partial sum from the reference data section A second adder for calculating a sum with the value.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail based on illustrated embodiments. FIG. 1 is a functional block diagram showing an embodiment of a checksum calculator according to the present invention.

The checksum calculator shown in this example connects input data to a protocol decision unit 1 and a field division unit 2 and outputs an output from the protocol decision unit 1 to the field division unit 2 and the reference data unit 3. Give to. The output of the field division unit 2 is connected to a first adder 4, and the second adder 5 receives the output from the reference data unit 3 and the output from the first adder 4 and outputs checksum determination data. Output.

The checksum calculator shown in FIG. 1 performs a calculation for checking whether or not a bit error has occurred during data transmission by using a checksum, and functions as follows. In addition, here, similarly to the description of the conventional example, the description will be made using the IP header as an example. That is, when a packet-type signal is input as input data, the protocol determining unit 1 identifies a protocol for each layer of the packet. The procedure at this time may be determined, for example, from a lower layer protocol to an upper layer protocol. The protocol determination unit determines the protocol in this way, and
The lower protocol information of P is output to the field division unit 2 and the reference data unit 3. The field dividing unit 2 divides the IP header into a fixed partial field and a variable partial field based on the determination result from the protocol determining unit 1. That is, when the protocol is determined by the protocol determining unit 1, the operating environment on the network becomes clear, and therefore, some of the parameters of the IP header can be grasped without reading the field data of the unit. Specifically, among the fields of the IP header shown in FIG. 2, the version, service, flag, and fragment are fixed partial fields because the operating environment does not change during the operation of communication. Other parts are variable part fields. Field division unit 2
Outputs data to the next-stage first adder 4 only for the variable portion after the division,
Is cumulatively added. The accumulative addition procedure is the same as in the past, in which 16 bits including the checksum field are added, and
When the carry is added to the addition result, 1 is added to the result, and the final addition result is output to the second adder 5.
Further, the reference data section 3 stores in advance the total value of the data of the fixed partial field for each type of protocol,
The corresponding fixed partial sum is output to the second adder 5 based on the determination result from the protocol determination unit 1. The second adder 5 adds the addition result from the first adder 4 and the fixed partial sum value from the reference data section 3, and outputs the result as checksum determination data.

If the checksum determination data is 0XFFFF, the checksum is established, and it is assumed that there is no bit error, and the process proceeds to the next stage. If the data for checksum determination is not 0XFFFF, the checksum is not satisfied, the packet is discarded, and a retransmission request is made. In this case, no bit error occurs because the data stored in the reference data section 3 is used for the fixed part, so that the bit error has occurred in the variable part field of the IP header. I understand.

As described above, if the checksum calculator having the configuration shown in FIG. 1 is used, it is not necessary to perform the cumulative addition operation on the value of the fixed partial field, so that the presence or absence of a bit error can be confirmed in a short time.

In the above-described embodiment of the present invention, a configuration example in which only the presence / absence of a bit error in a received packet is confirmed, but the present invention is not limited to this example. For example, when a packet is transferred in a host or the like, the calculation processing time can be reduced by using the fixed partial sum value of the reference data section 3 when newly calculating a checksum after changing a variable part field. Can be.

As described above, the checksum calculator according to the present invention can reduce the time required for checking the checksum for a bit error during data transmission and calculating the checksum value.

[0016]

As described above, the checksum calculator according to the present invention stores the total value of the fixed partial field in the reference data unit 3 in advance, and based on the result determined by the protocol determining unit 1, The configuration is such that the partial sum value is directly supplied to the second adder 5, and based on the result determined by the protocol determination unit 1, the field division unit 2 extracts only the variable partial field data and
, The result of the cumulative addition by the first adder 4 is supplied to the second adder, and the second adder 5 calculates the sum of the fixed partial sum and the variable partial sum. Therefore, it is possible to perform a checksum calculation without performing cumulative addition on all fields targeted by the checksum, thereby realizing a checksum calculator capable of performing a checksum calculation in a short time.

[Brief description of the drawings]

FIG. 1 is a functional block diagram showing a configuration example of a checksum calculator according to the present invention.

FIG. 2 is a functional block diagram showing another configuration example of the checksum calculator according to the present invention.

FIG. 3 is a diagram showing a configuration example of an Internet Protocol (IP) header.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Protocol determination part 2 ... Field division part 3 ... Reference data part 4 ... 1st adder 5 ... 2nd adder 6 ... 3rd adder 7. ..Fourth adder 8 ... one's complement operation unit

Claims (1)

[Claims] 1. A checksum calculator for performing data communication, comprising: a protocol determining unit for determining a protocol of an input data packet; and a fixed part and a variable part based on a result from the protocol determining unit. A field division unit that divides a field and extracts only a variable part; a reference data part that stores a result value obtained by adding data values of the fixed part in advance; and a variable part data that is divided by the field division unit. And a second adder for summing a result of addition from the adder and a fixed partial sum value from the reference data section. Calculator.