JP5308193B2 - Serial communication method - Google Patents

Description

The present invention relates to an asynchronous serial data communication method between different devices, and more particularly, to a method for correcting a synchronization error in real time in order to prevent an error in received data caused by a synchronization error between devices.

For example, in the case of a computer system in which a plurality of devices are used, these devices are operated independently in normal times, but necessary signals such as monitor data between the devices are always transmitted to the computer. It is customary to take in and prepare for the occurrence of an abnormality.

By the way, when transmitting such signals for monitoring, when there are a plurality of signals, a so-called parallel data communication method in which signals to be transmitted are transmitted independently has been conventionally used. In this case, as many communication channels as the number of data are required.

Therefore, there is an asynchronous serial data communication method as a communication method that requires only one communication channel regardless of the number of data. Further, there is an asynchronous serial data communication method as a kind of asynchronous serial data communication method.

This method performs reception processing according to the start bit and stop bit and can synchronize in units of characters. Therefore, even when a communication error occurs, there is little possibility that an error will occur over the entire transmitted information. There is an advantage.

For example, Patent Document 1 describes means for accurately detecting these bits for synchronization.

Also, unlike the start-stop synchronization method that synchronizes in units of characters, the method of synchronization in units of sentences is called an independent synchronization method. This method is suitable for high-speed data communication because the number of synchronization signals is smaller than that of the start-stop synchronization method. However, this method also has a drawback that the influence when a communication error occurs becomes large.

JP 2001-177583 A

However, in both the start-stop synchronization method and the independent synchronization method, when a communication error occurs with respect to the synchronization signal, it is inevitable that an error in information or a request for retransmission will occur. In addition, the amount of processing related to error correction for dealing with complicated communication errors also increases.

Therefore, an object of the present invention is to show a method of correcting a synchronization shift in serial communication regardless of the synchronization method and without causing a further increase in transmission signal rate due to addition of an error correction signal.

The serial communication system according to claim 1 samples a plurality of data serially transmitted in a start-stop synchronization system or an independent synchronization system at a constant k times the serial transmission rate, and obtains the data in order of time series under the sampling. Among the 0th to (k-1) th sample values to be obtained, a central sample value substantially in the middle, a preceding sample value preceding the central sample value, and a subsequent sample value following the central sample value are obtained. Pre-sampling means, first sampling means using the central sample value as received data when the central sample value, the preceding sample value, and the subsequent sample value are equal, and the central sample value and the preceding sample value Are different from each other, the second sampling means using the (k−1) th sample value as received data, and the central sampling And the third sampling means using the 0th sample value as received data when the sample value is different from the subsequent sample value, and the 0th sample value or the (k−1) th sample as the received data And synchronization control means for regarding the value as the central sample value.

The serial communication system according to claim 2 is the serial communication system according to claim 1, wherein the sampling unit is configured to specify the received data until the received data is specified by any one of the first to third sampling units. It is constructed by storing the 0th to (k-1) th sample values.

According to the present invention, it is possible to eliminate reception errors of received data due to synchronization deviation regardless of asynchronous serial communication such as start-stop synchronization method and independent synchronization method, and more accurate information communication can be performed. It becomes possible.

Further, at that time, since processing such as an increase in transmission rate that prevents synchronization loss is not performed, there is no need to perform special processing on the transmission side.

Hereinafter, a serial data communication method according to the present invention will be described in detail with reference to the drawings. Note that the meaning of “representative value” used in the description determines that the value of received data is D from a plurality of sampling points with respect to one received data value D having a certain width in serial communication. It shows the sampling position. For example, in FIG. 2, if the center of the width is determined as received data, the representative value is e.

FIG. 1 shows a data string Dn (n = 0, 1, 2,...) Transmitted from the transmitting side A to the receiving side B. If A and B are synchronized, B can receive the data string without any problem. For example, B may sample anywhere within the range of X and Y shown in FIG.

For example, when determining the sampling position, considering the fluctuation of synchronization, it is desirable to determine the data value at the midpoint between X and Y in FIG. 1 as the received data. Therefore, if reception sampling is performed at a speed higher than the transmission clock on the transmission side and nine sampling points from a to i are included in one reception data, the position of e shown in FIG. In the case, it is the best representative value. Note that sampling position labels such as a to i are given by a sampling number assigning unit installed on the receiving side.

However, if the position of e in FIG. 2 is determined as the representative value and the value of the received data is determined, if the synchronization is lost, the position of e is no longer the center position of the data width as shown in FIG.

As shown in FIG. 3, in the case where the received data is sampled at a sampling period that is a constant multiple of the transmission clock as shown in FIG. Is used to correct the synchronization error.

Here, as an example, a description will be given of a means for determining that the synchronization is shifted when the sampling values at the two sampling positions in total, ie, the sampling position shifted by one from the central position are different.

FIG. 4 shows a case where the representative value e is different from the sampling values at two sampling positions, i.e., one sampling position f ahead of e.

As shown in FIG. 4, when the sampling values at the two sampling positions are different, the present invention determines that a synchronization error has occurred. Next, when it is determined that a synchronization shift has occurred in this way, the synchronization shift further proceeds by changing the representative value from e to a as shown in FIG. It is possible to prevent the representative value of near the boundary between data.

Similarly, FIG. 6 shows a case where the representative value e and the sampling values at the two sampling positions of the sampling position d just before e are different. In this case, by changing the representative value from e to i, the synchronization shift further prevents the representative value of the next incoming serial data from becoming near the boundary between the data.

FIG. 7 is a flowchart regarding the processing described so far according to the present invention.

The transmission side performs serial data transmission at n Hz in the transmission unit 701. The reception side samples reception data at the reception unit 703 at a sampling frequency n kHz that is k times that of the transmission side. Next, the sampling number assigning unit 705 gives k labels to the sampling positions of the received data, and the sampling values at the respective positions are stored in the sampling value storage unit 707.

Referring to the value stored in 707, the first sampling value comparison unit 709 compares the k / 2 and k / 2-1th sampling values. Similarly, the second sampling value comparison unit 711 compares the k / 2 and k / 2 + 1th sampling values.

When it is determined that the two sampling values are not different in both 709 and 711, the first representative value determining unit 713 determines that the current representative value is used as it is in the next time.

If it is determined in 709 that the two sampling values are different, the second representative value determining unit 715 determines to use the position of the (k−1) th sampling number as the next representative value. Further, the first sampling number correcting unit corrects the k sampling numbers so that the (k-1) th sampling number becomes the k / 2th sampling number.

When it is determined in 711 that the two sampling values are different, the third representative value determining unit 719 determines to use the position of the 0th sampling number as the next representative value. Further, the second sampling number correction unit corrects the k sampling numbers so that the 0th sampling number becomes the k / 2th sampling number.

The representative sampling values determined in 713, 715, and 719 are finally determined sampling values and transmitted from the data transmission unit 723 to the data processing unit in the receiver. Thereafter, when the data is received, the series of processes are repeatedly executed.

An overall processing block diagram is shown in FIG. When data is received from the reception terminal 801, the k-times sampling unit 803 samples the received data at k times the transmission frequency on the transmission side, and the representative value determination unit 805 determines which of the k sampling values is received data. The determined sampling value is transmitted to the data processing device 807.

A more detailed processing block diagram of the representative value determining unit 805 is shown in FIG. The sampling number assigning unit 901 gives labels to the k sampling positions, and the sampling value comparison unit 903 compares the k centers and the sampling values before or after the proximity.

The synchronization deviation determination unit 905 refers to the comparison result of 903, and determines that a synchronization deviation has occurred if the two values are different. Next, the synchronization error correction unit 907 determines the representative value depending on whether the k / 2th and k / 2−1th values are different, or whether the k / 2nd and k / 2 + 1th values are different. And is transmitted by the representative value output unit 909.

As described above, according to the present invention, it is possible to prevent the synchronization deviation from gradually increasing, resulting in erroneous reception data by sequentially correcting the representative value.

In the present embodiment, for example, in the first sampling value comparison unit 709 and the second sampling value comparison unit 711, the two points to be compared are described as the central portion and one around it, but from the central portion, Is not limited to 1.

In this embodiment, the second representative value determining unit 715 and the first sampling number correcting unit pick up the (k−1) th sampling position, but it is the extreme end of the k sampling points. It is not limited.

In the present embodiment, the third representative value determining unit 719 and the second sampling number correcting unit pick up the 0th sampling position, but it is limited to being the end of the k sampling points. Not what you want.

Serial communication transmission / reception data string A piece of data in serial communication Example when synchronization loss occurs Effect of sampling loss on sampling position and sampling value Correction of representative value 1 Correction of representative value 2 Processing flowchart Processing block diagram 1 Processing block diagram 2

701 ... Transmitting unit, 703 ... Receiving unit, 705 ... Sampling number assigning unit,
707 ... Sampling value storage unit, 709 ... First sampling value comparison unit,
711 ... second sampling value comparison unit, 713 ... first representative value determination unit,
715, a second representative value determining unit, 717, a first sampling number correcting unit,
719 ... a third representative value determining unit, 721 ... a second sampling number correcting unit,
723 ... a data transmission unit,
801 ... receiving terminal, 803 ... k times sampling unit,
805 ... representative value determining unit, 807 ... data processing device,
901: Sampling number assigning unit, 903 ... Sampling value comparing unit,
905 ... synchronization deviation determination unit, 907 ... synchronization deviation correction unit,
909: Representative value output unit.

Claims (2)

A plurality of data serially transmitted in an asynchronous manner or an independent synchronous manner are sampled at a constant k times the rate of the serial transmission, and the 0th to (k−1) th data obtained in order of time series under the sampling Pre-sampling means for obtaining a central sample value substantially in the middle of the sample values, a preceding sample value preceding the central sample value, and a subsequent sample value following the central sample value;
First sampling means using the central sample value as received data when the central sample value, the preceding sample value, and the subsequent sample value are equal;
When the central sample value and the preceding sample value are different, the second sampling means using the (k-1) th sample value as received data;
When the central sample value and the subsequent sample value are different, a third sampling means using the 0th sample value as received data;
Synchronization control means for regarding the 0th sample value or the (k−1) th sample value as the received data as the central sample value;
Serial communication method comprising the. In the serial communication system according to claim 1,
The sampling means includes
The 0th to (k-1) th sample values are stored until the received data is specified by any of the first to third sampling means.
A serial communication system characterized by this .