JP2001156861A - Data communication method, two-way data communication method, data communication system, two-way data communication system, data transmitter and data receiver - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve a problem such that a mechanism is required for securing synchronization of codes and a simple constitution cannot comply with this requirement when the communication of data is carried out via a communication path that has a high error rate. SOLUTION: The data are represented at time intervals set between specific time series and the communication of data is carried out. Thereby the correlation is secured between the specific and receiving time series and the specific series are detected at the receiving side. Thus, the time intervals representing the data are decided and the data can be restored. Under such conditions, the specific series uses the series that have sharp self-correlation coefficients and the threshold of correlation is properly selected. As a result, the specific series can be detected despite an error of the receiving series, that is, the data can be restored and also the synchronous communication is attained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a data communication method,
The present invention relates to a two-way data communication method, a data communication system, a two-way data communication system, a data transmitting device, and a data receiving device.

[0002]

2. Description of the Related Art The following attempts have been made to realize bidirectional full-duplex communication using a single wavelength using a single-core optical fiber. (A) Communication is performed using a device that has sufficiently low reflection of transmission light at the fiber end face and electromagnetic noise from the transmission unit to the reception unit. (B) Communication is performed using an error correction code having a redundancy necessary for achieving a target error rate.

However, each of these attempts has the following problems. (A) To reduce the error occurrence rate to a level that does not hinder communication, the cost of the communication device increases. (B) In order to perform communication using an error correction code, code synchronization must be established, so a mechanism for accurately achieving synchronization is separately required.

Therefore, it is necessary to devise a method for accurately and efficiently synchronizing data transmission on a serial communication path having a high error rate. One example is a spread spectrum communication method applied to multiplex communication.

When the spread spectrum communication method is applied to the serial communication path, data communication is performed by expressing one bit using the inversion state of a specific sequence. Here, although the bit synchronization can be achieved by detecting the specific sequence, a mechanism for synchronizing symbols (inseparable minimum units of information) is required.

In the spread spectrum communication method, as a method of avoiding separately performing symbol synchronization, a method of transmitting symbols collectively using a plurality of multiplexed channels can be considered. In other words, by transmitting bits constituting a symbol simultaneously from a plurality of channels on the transmission side, the bits constituting the symbol can be received simultaneously on the reception side. Therefore, if bit synchronization is established, the symbol is determined. be able to.

As a method of realizing multiplex communication using a spread spectrum communication method in a serial communication path, it is conceivable to apply a communication method disclosed in Japanese Patent Application Laid-Open No. 8-130526, "Multiplexing method in spread spectrum communication".

Japanese Patent Laid-Open Publication No. Hei 8-130526, "Multiplexing Method in Spread Spectrum Communication", is characterized by a spread code sequence used for spread modulation and a method of determining the spread code sequence. A spreading code sequence used for spreading modulation is represented by a pattern of a combination of a plurality of specific sequences shorter than the spreading code sequence and a time interval inserted therebetween. The receiving side detects the specific sequence and the time interval between the specific sequences by correlating a received signal with the specific sequence, and detects the spread code sequence.

With this, in the spread spectrum communication, the transmitting station of the received data is specified by assigning a specific pattern to each transmitting station based on a combination of time intervals inserted between a plurality of specific sequences. And multiplexing can be realized.

In this case, by selecting a plurality of the specific sequences used for each channel and a pattern of the time interval so that the specific sequences do not interfere with each other, for example, in a time-shift manner, between a plurality of channel signals. Multiplexing is also possible on a serial communication path.

Here, by applying each bit constituting one data symbol as an object of multiplexing and performing communication, there is no special mechanism separately in the serial communication path, and communication in which symbol synchronization is achieved is achieved. It can be performed.

[0012]

Hereinafter, in this specification, a continuous signal sequence received by a receiver is referred to as a reception sequence.
A sequence obtained by cutting out a continuous portion having a length equal to the specific sequence from the received sequence is referred to as a partial reception sequence. In addition, information treated as one group, for example, a character code is described as a symbol.

Here, when applying the above-mentioned Japanese Patent Laid-Open Publication No. Hei 8-130526 “Multiplexing method in spread spectrum communication” to a serial communication channel, a pattern of a combination of a plurality of specific sequences and time intervals inserted therebetween is described. It is assumed that each bit is assigned to each transmitting station, and one bit is represented by the inverted state of the specific sequence.

Therefore, in the case where the object of multiplex communication is applied not to each data transmitted by each transmitting station but to the transmission of each bit constituting one symbol, the receiver is configured as shown in FIG. Configuration. As shown in the figure, the receiver receives a transmitted sequence by detecting whether or not the correlation between the newest partial received sequence in the received sequence and the specific sequence exceeds a threshold value. A correlator 2 for detecting the specific sequence and generating a detection signal, pattern determining units 3a to 3d for determining the inversion state of a pattern of a time interval between detection signals corresponding to each bit of a received symbol, and corresponding to each bit. In the case where the lengths of the patterns at the respective time intervals are different and the output values of the respective pattern determiners are output at different timings, in order to accurately determine the symbols, the synchronization of the output values of the respective pattern determiners is synchronized. Devices 4a to 4d, and a decoder 5 that determines the transmitted symbol from the output value of each pattern determiner and transmits the determined symbol to the upper layer.

As described above, in the data communication system to which the above-mentioned prior art is applied, the number of pattern decision units 3a to 3d and the synchronizers 4a to 4d corresponding to each bit constituting the symbol are required in the receiver. However, there is a disadvantage that the configuration becomes large.

In the case where the pattern of each time interval corresponding to each bit is of the same length and the receiving side selects all bit values at the same timing, the synchronizers 4a to 4d Is unnecessary, but the pattern of each time interval corresponding to each bit has a fixed length, and it is not possible to select a symbol having a high appearance frequency and a pattern of each time interval that is short. There is a disadvantage that it goes down.

An object of the present invention is to solve these problems,
In particular, it is an object of the present invention to provide a data communication method suitable for accurately transmitting a symbol using a communication path in which an error occurs when the type of the symbol is small.

[0018]

Means for Solving the Problems (1) A symbol is represented by the length of a time interval between specific sequences. In this communication method, the time interval between the specific sequences representing the symbols to be communicated can be detected at the receiver by correlating the partial reception sequence cut out from the received sequence with the specific sequence. . That is, as shown in FIG. 1, the time interval representing a symbol is detected from the time interval between detection signals generated by detecting whether the correlation exceeds a threshold and detecting the specific sequence from a received sequence. it can. Here, by selecting an appropriate value for the threshold value, accurate and synchronized communication can be performed even on a communication path with a high error rate.

Furthermore, compared with the method of applying the above-mentioned Japanese Patent Application Laid-Open No. Hei 8-130526, "Multiplexing method in spread spectrum communication", a plurality of pattern decision units are not required, so that data communication can be performed with a small configuration. it can. Further, the data transfer rate can be increased by associating a symbol with a high appearance frequency with a symbol having a short time interval.

(2) Except when the partial reception sequence is the specific sequence, transmission is performed in a period corresponding to a time interval between the specific sequences so that the correlation between the partial reception sequence and the specific sequence is low. By selecting a specific sequence, the specific sequence can be accurately detected from the received sequence as compared with the case where nothing is transmitted during this period. Hereinafter, in this specification, a time interval between the specific sequences is referred to as a gap, and a sequence transmitted in the gap is referred to as a gap sequence. In the claims, the specific sequence is referred to as a first specific sequence, and the gap sequence is referred to as a second specific sequence.

(3) By using a binary sequence for the specific sequence and the gap sequence and setting the gap interval in bit units, the present invention can be applied to binary communication in an optical fiber or the like. The present invention is not limited to this, and can be applied to, for example, multi-level communication in which quaternary data can be transmitted. It is also conceivable to transmit an analog quantity by continuously changing the time interval.

Further, a specific device for accurately and efficiently communicating data will be described below.

(4) A symbol is represented by using the presence or absence of inversion (inverting 1, 0 of each bit) of the specific sequence in addition to the time interval between the specific sequences. As a result, the maximum bit length and the average bit length necessary for transmitting one symbol can be reduced, and the data transfer rate can be improved. Hereinafter, in the present specification, inverting 1, 0 of each bit of the specific sequence is referred to as inversion for the specific sequence. Further, when the specific sequence that has not been inverted on the transmission side is transmitted, the value of the detection signal generated from the correlation with the specific sequence on the reception side is +1. When transmitted, the value of the detection signal generated from the correlation with the specific sequence on the receiving side is set to -1.

(5) In the case where the present communication method is applied to two-way communication, the specific sequence used by one communication path is inverted and the other communication path is used as the specific sequence. Accordingly, even when the two communication paths are likely to affect each other, that is, when crosstalk occurs, a detection signal is erroneously generated as compared with the case where the same specific sequence is used in both communication paths. Can be reduced.

In the case where bidirectional communication is performed using the data communication method of the present invention, if the same specific sequence is used in both communication paths, if there is a high possibility that the two communication paths will affect each other, one of them is used. Is more likely to erroneously generate a detection signal due to the specific sequence communicated on the other communication path. Therefore, using a different sequence as the specific sequence is advantageous in reducing the occurrence rate of data determination errors. Here, the sequence having the lowest correlation with the specific sequence is a sequence obtained by inverting its own sequence, and it is effective to use these two sequences as the specific sequence in each communication channel.

(6) As the specific sequence, a pseudo-noise sequence (PN sequence), which is a sequence having a sharp autocorrelation function, that is, one of the sequences having a property of reducing the correlation with the sequence rotated by itself, is applied. This makes it possible to accurately detect the specific sequence in a received sequence.

(7) As the specific sequence, a Barker sequence which is one of sequences having a sharp autocorrelation function is applied. Since the Barker sequence has a property in which the correlation between the Barker sequence and its own sequence has a remarkable value, the Barker sequence represents a combination of a time interval between symbols and the presence or absence of inversion of the specific sequence. In a communication method, it is suitable for use as the specific series.

(8) As the correlation between the partial reception sequence and the specific sequence, a value obtained by subtracting the number of non-matching bits from the number of bits where the value of each bit at the corresponding position in the partial reception sequence and the specific sequence matches Is used. Thereby, the calculation of the correlation can be easily performed, and the mounting by the digital circuit is enabled.

(9) In the present invention, a communication method for performing communication of 10 types of symbols is described as follows.
Using a ker sequence, transmitting a specific gap sequence in a period corresponding to a time interval sandwiched by the specific sequence,
Specific examples of the specific sequence and the gap sequence that can be applied when a method of representing a symbol by a time interval sandwiched by the specific sequence and an inverted state of the specific sequence are used. This is a sequence that is particularly selected so as to be advantageous in detecting the specific sequence from the received sequence by the correlation between the partial received sequence and the specific sequence. This allows
IEEE Std 13 for communication in 9 types of states
94-1995 and arbitration of high-speed serial bus communication conforming thereto.

(10) As a communication system using the communication method according to the present invention, in a transmitting device, means for converting data to be transmitted into the time interval, means for storing the specific sequence, and conversion into the time interval Means for transmitting the obtained data sandwiched by the specific sequence, the receiving device detects whether the correlation between the partial received sequence and the specific sequence exceeds a threshold, thereby detecting the specific sequence from the received sequence. A data communication system comprising: means for generating a detection signal at a corresponding timing; and means for restoring data from a time interval obtained by subtracting the specific sequence length from a time interval between the detection signals.

(11) As a two-way communication system using the communication method according to the present invention, two-way data communication in which a specific sequence used in one communication channel is inverted and used as the specific sequence used in the other communication channel. Configure the system. Thereby, a highly reliable two-way communication system can be constructed.

(12) As a data transmitting apparatus in the communication system according to the present invention, means for converting data to be transmitted into a time interval, means for storing the specific sequence, and specifying the data converted into the time interval into the specific And a means for transmitting the data in series.

(13) As a data receiving device in the communication system according to the present invention, by detecting whether the correlation between the partial reception sequence and the specific sequence exceeds a threshold value,
Means for detecting the specific sequence from a received sequence and generating a detection signal at a corresponding timing; and means for restoring data from a time interval obtained by subtracting the specific sequence length from a time interval between the detection signals. Construct a receiving device.

(14) In the data receiving apparatus of the present invention, as the means for detecting the specific sequence from the received sequence, means for detecting the specific sequence having a function of changing a threshold value of the correlation with the specific sequence is used. Thereby, the probability of erroneously detecting the specific sequence in the received sequence can be reduced. Here, if the specific sequence is correctly detected in the received sequence, the next specific sequence is not detected at least during a period shorter by one bit than the specific sequence length. That is, in the received sequence, the probability that the specific sequence is detected changes with the passage of time from the detection of the specific sequence. By changing the threshold used for the correlation in response to this, it is possible to reduce the probability of erroneously detecting the specific sequence in the received sequence.

(15) In a receiving apparatus in a data communication system using the data communication method described in claim 9 and using the correlation calculation method described in claim 8, the threshold value is set as follows:
After the detection signal is generated, it is set to 10 and -10 during a period shorter by one bit than the specific sequence length, and is set to 6 and -6 otherwise. Thus, in a sequence obtained by cutting out any 11 bits from the received sequence, if the occurrence of an error is 2 bits or less, data can be accurately communicated.

(16) In a receiving apparatus in a data communication system using the data communication method according to claim 10 and using the correlation calculation method according to claim 8, the threshold is set to the specific sequence after a detection signal is generated. It is set to 6 and -6 during a period of one bit shorter than the length, and is set to 4 and -4 otherwise. Thus, in a sequence obtained by cutting out arbitrary 7 bits from the received sequence, if the occurrence of an error is 1 bit or less, data can be accurately communicated.

[0037]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS (Embodiment 1) As an embodiment of a communication apparatus according to the present invention, a system for performing bidirectional full-duplex communication with a single wavelength using a single-core optical fiber is IEEE.
Std 1394-1995 and a case where the present invention is applied to transmission of an arbitration signal of high-speed serial bus communication conforming thereto. IEEE Std 1394-19
In the arbitration period of the high-speed serial bus communication 95 and the compliant arbitration period, the terminals mutually transmit nine types of states. That is, there are nine types of symbols. Therefore, it is assumed that state data is represented by a gap interval of 0 to 8 bits.

After the arbitration is completed and data transmission is started, data transmission may be performed by a general coding method, for example, NRZ. Since data transmission is one-way, that is, half-duplex communication, it is easy to improve the error rate by devising an optical system.

Here, a detection signal is generated when the correlation between the specific sequence and the partial reception sequence is higher than a threshold, but a detection signal may be generated when the correlation is lower than the threshold.

Here, the transmitter is as shown in FIG.
It comprises an encoder 11, a control unit 12, a specific sequence generator 13, and a transmission unit 14. The function of each block constituting the transmitter will be described. The encoder 11
The state received from the upper layer is converted into a time interval and transmitted to the control unit. After counting the time until the gap coincides with the time interval, the control unit 12 informs the specific sequence generator 13 of the timing for transmitting the specific sequence. The specific sequence generator 13 generates the specific sequence at a timing designated by the control unit 12 and transmits the specific sequence to the transmission unit 14. The transmitting unit 14 transmits the specific sequence sent from the specific sequence generator 13 bit by bit according to a fixed clock.

As for the receiver, as shown in FIG. 3, a receiving section 21, a received sequence storage section 22, a correlator 23, a counter 2
4. Decoder 25. The function of each block constituting the receiver will be described. Receiver 21
Each time 1 bit of data transmitted by the transmitter is received, the received 1 bit is transmitted to the reception sequence storage unit 22. The received sequence storage unit 22 adds one bit received from the receiving unit 21, stores the most recently received sequence by the same length as the specific sequence, and transmits it to the correlator 23 as a partial received sequence. The correlator 23 correlates the partial reception sequence transmitted from the reception sequence storage unit 22 with the specific sequence, and when the correlation exceeds a preset threshold,
A detection signal is generated and transmitted to the counter 24. Counter 24
Counts the time interval during which the detection signal is not transmitted from the correlator 23, and transmits the value of the counter indicating the time interval between the detection signals to the decoder 25 when the detection signal is detected.
The decoder 25 determines a state from the value of the counter transmitted from the counter 24 and transmits the state to an upper layer.

When a detection signal is generated at a time interval shorter than the specific sequence, and when the detection signal is not generated for a longer time than the sum of the specific sequence and the longest gap sequence, an error occurs. It is also conceivable to inform the upper layer of the detection. Also, in this case, it is conceivable to transmit the state received last time to the upper layer instead of the error.

In this embodiment, the specific series is 1
A case where a value of 4, 2, 0 is sequentially communicated from a state in which nothing is communicated using a 1-bit sequence will be specifically described.

The operation of each component of the transmitter will be described with reference to FIG. Here, the transmission unit 14 is a specific sequence generator 13
It is assumed that the transmitted bits are transmitted according to the clock. In the figure, X represents an indefinite value, and portions indicated by broken lines in the transmission sequence and the reception sequence are periods during which the transmitter does not transmit anything, and indicate that the value is indefinite.

(1) It is assumed that the specific sequence generator 13 outputs a specific sequence, and that the control unit counts the elapsed time from the start of the output of the specific sequence. Also, the specific sequence generator 1
3 sequentially transmits the specific sequence to the transmitting unit 14 bit by bit. When the encoder 11 receives the first data value 4 transmitted from the upper layer, the encoder 11 notifies the control unit 12 of the output interval value 14 indicating the output timing of the specific sequence corresponding to the value. As soon as the counter value reaches 10,
The control unit 12 takes in the output interval value. Here, the fact that the counter value becomes 10 indicates that the output of the specific sequence from the specific sequence generator 13 is completed.

(2) When the value of the counter of the control unit 12 becomes 14 and coincides with the output interval value, the control unit 12 transmits a control signal to the specific sequence generator 13 and resets the counter value at the same time. Start counting time elapsed from. The specific sequence generator 13 that has received the control signal
The specific sequence is sequentially transmitted to the transmission unit 14 bit by bit.

(3) At the same time when the value of the counter of the control unit 12 becomes 10, the control unit 12 takes in the output interval value 12 corresponding to the next transmission data value 2.

(4) Next, as in (2), the control unit 1
At the timing when the value of the counter 2 matches the output interval value 12, the transmission of the specific sequence is started. Similarly to (3), the value of the counter of the control unit 12 becomes 10, and at the same time, the control unit 12 The output interval value 10 corresponding to the transmission value 0 is taken.

(5) Subsequent to (2) and (3), as in (4), the output interval value corresponding to the data value to be transmitted next
Is repeated to continuously transmit data.

The operation of each component of the receiver will be described with reference to FIG. Here, it is assumed that the reception sequence storage unit 22 stores the most recently received partial reception sequence by receiving data bit by bit.

(1) The reception sequence storage unit 22 receives data from an empty state until the data has the same length as the specific sequence length,
Thereafter, each time one bit is received, the partial reception sequence is
Tell The correlator 23 calculates the correlation between the partial reception sequence and the specific sequence, generates a detection signal when the correlation exceeds the threshold value, and transmits the detection signal to the counter 24. When the detection signal is detected, the counter 24 resets the counter value, and starts counting the lapse of time from that point.

(2) Next, the correlator 23 detects that the correlation has exceeded the threshold, generates a detection signal, and
, The counter 24 reads the current value of the counter 1
4 is transmitted to the decoder, and counting of the elapsed time from that point is started again. Upon receiving the value 14 of the counter, the decoder transmits a data value 4 corresponding to the time interval to the upper layer.

(3) Next, when the correlator 23 detects that the correlation has exceeded the threshold value again, the data value 2 calculated from the counter value 12 of the counter 24 is transmitted to the upper layer as in (2). Can be

After (4), by repeating (2), every time a detection signal is generated in the correlator, the data value calculated from the counter value is continuously transmitted to the upper layer.

(Second Embodiment) In comparison with the first embodiment, a Barker sequence is used as the specific sequence in order to further reduce a determination error due to a random bit error in a received sequence. Further, a configuration is shown in which a mechanism for transmitting a specific sequence in the gap is added to the transmitter, and a mechanism for adaptively changing the threshold is added to the receiver.

Since the specific sequence needs to be accurately detected in the received sequence, the specific sequence is required to have a property that the self sequence has a low correlation with the rotated sequence. That is, it is necessary to use a sequence having a sharp autocorrelation function as the specific sequence. It is assumed that a Barker sequence which is a sequence having such properties is used as the specific sequence. There is also a method of using a pseudo noise sequence (PN sequence) as the specific sequence.

In the reception sequence, the partial reception sequence is Bar
In the case other than the case of the ker sequence, the partial reception sequence and B
The gap sequence is selected so that the correlation with the arcer sequence is low. This makes it possible to reduce the probability that an erroneous detection signal will be generated with respect to the occurrence of a random bit error in a received sequence, as compared to the case where nothing is transmitted in the gap. In addition, the fact that there is no bias in the appearance frequency of each binary in the binary data sequence, that is, that the DC balance is maintained, is advantageous in various communication paths including optical fibers. Therefore, a gap sequence in which the difference between the binary appearance frequencies is small can be selected. In the fourth embodiment, a Barker sequence and a gap sequence are specifically shown.

Here, in the transmitting apparatus, as shown in FIG. 6, in addition to the configuration of the first embodiment, a gap sequence generator 15
a, a selector 16 for selecting which of a gap sequence and a Barker sequence is to be transmitted is required. The specific sequence generator is a Barker sequence generator 13a that generates a Barker sequence, and the control unit 12a has a function of controlling the gap sequence generator 15a, the Barker sequence generator 13a, and the selector 16.

If the detection of the Barker sequence is not erroneous in the receiver, the next Barker sequence will not be detected in a period shorter than the Barker sequence length by one bit after the timing at which the Barker sequence is detected. Therefore, by increasing the threshold value during this period, the probability of generating a detection signal by mistake can be reduced,
By lowering the threshold value during periods other than the period, the probability that a detection signal is not erroneously generated can be reduced.

Further, after the detection of the Barker sequence,
The threshold value is set high in a short period,
By making it possible to detect the rker sequence, it is possible to generate the next correct detection signal when the previous generation of the detection signal is incorrect. In order to adaptively change the threshold, the receiving apparatus requires a threshold setting unit 26a in addition to the configuration used in the first embodiment, as shown in FIG. In the threshold setting device 26a, the counter 2
With reference to the value of 4a, the correlator 23a
A threshold used to detect a Barker sequence is set.

(Embodiment 3) When bidirectional full-duplex communication is performed in a single-core optical fiber using a single wavelength, an error may occur because one communication path affects the other. Therefore, it is advantageous to use a different sequence as the specific sequence instead of using the same sequence as the specific sequence in both communication paths in order to reduce the occurrence rate of data determination errors. Here, since the sequence having the lowest correlation with the specific Barker sequence is a sequence obtained by inverting itself, in the second embodiment, one of the communication channels is obtained by inverting the Barker sequence used as the specific sequence to the other communication channel. A road is used as the specific series.

(Embodiment 4) In Embodiment 2, Embodiment 1 and Embodiment 2 are described in that the data is further represented by combining the gap length and the inverted state of the Barker sequence.
Also, compared to the third embodiment, the maximum gap length required for transmitting one symbol is shorter, and the data transmission rate can be increased.

It is assumed that nine states are communicated in 1394 arbitration using the inverted state of the Barker sequence and a gap interval of 0 to 4 bits. Here, on the receiving side, as shown in FIG. 8, the time interval between Barker sequences and the presence / absence of inversion of the Barker sequence can be detected by correlating the partial reception sequence with the Barker sequence and using a threshold. it can.

Here, the inverted state of the Barker series is two states, ie, not inverted and inverted, and is equivalent to representing one bit. The following expressions are conceivable. (A) The two states of +1 and −1 of the detected detection signal are made to correspond to binary values. (B) A binary value indicates whether the value of the detected detection signal is the same as or different from the value of the previously detected detection signal.

Here, in the case of (A), the second embodiment
On the other hand, in the transmitting apparatus, as shown in FIG. 9, an inversion control unit 17b that controls the inversion of the gap sequence and the Barker sequence.
Is added, and the correlator 2 is used in the receiving apparatus as shown in FIG.
3b requires a function of transmitting the value of the detection signal to the decoder 25b.

In the case of (B), the second embodiment
On the other hand, as shown in FIG.
The inversion state of the ker sequence is stored, and the gap sequence and Ba are stored.
A previous inversion state storage and inversion control unit 17c for controlling the inversion of the rker sequence is added. In the receiving apparatus, as shown in FIG.
And a function of notifying the previously detected signal value storage unit 27,
Further, a previous detection signal value storage unit 27 for storing the value of the detection signal detected last time is required.

The case (A) has an advantage that the circuit is simpler than the case (B). In the case of (B),
Since it is possible to make the appearance probabilities of bits 1 and 0 in the series to be communicated uniform, it is more advantageous than (A) when performing communication over an optical fiber or the like.

In the following, particularly in the case of using the method (B), as a device for accurately transmitting an arbitration state in a communication path in which an error occurs, a specific method of (1) a Barker sequence and a gap sequence to be applied is specifically described. , (2) the calculation method of the correlation and the procedure of the adaptive variation of the threshold value.

(1) Barker sequence and gap sequence to be applied As the Barker sequence to be applied as the specific sequence, A
= 11100010010, and B = 00011101101, which is a series of the inverted sequence, is used. Also, corresponding to the Barker sequence, the received sequence has a Bar
The following is used as a gap sequence whose correlation is low except when a ker sequence is detected. (A) When A is continuous, the gap sequence of 1 to 4 bits interposed between them is 1, 11, 11
0,1101 is used. (B) When B follows A, the gap sequence of 1 to 4 bits interposed between them is 1, 10, 1
00 and 1001 are used. (C) When B is continuous, the gap sequence of 1 to 4 bits sandwiched between B is 0000, 00, 00, respectively.
Use 1,0010. (D) In the case where A follows B, 0, 01, 0 as gap sequences of 1 to 4 bits interposed therebetween.
11,0110 is used.

Note that the specific sequence and the gap sequence are not limited to these, and the correlation between the partial reception sequence including the gap sequence and the specific sequence is low. It can be selected within the range that satisfies. It is also conceivable to use a longer specific sequence on a communication channel having a high error occurrence probability.

The combination of these Baker sequences and gap sequences is a combination of two Barker sequences and a gap sequence sandwiched between them, and for 9 of them, the deviation of the number of occurrences of 0 and 1 for each bit is within 1 In the other one kind of combination, the deviation of the number of occurrences of 0 and 1 for each bit is 2, which is suitable for optical fiber communication. Here, it is assumed that a combination that becomes the same series when 0 and 1 of all bits are inverted represents the same data.

Here, in order to simplify the circuit for generating the gap sequence of the transmitter, the same sequence as in (a) is applied to the gap sequence used in case (b), and (d)
It is also possible to apply the same sequence as (c) to the gap sequence used in the case of (1).

(2) Correlation Calculation Method and Threshold Adaptive Variation Procedure The correlation between the partial reception sequence and the Barker sequence is calculated from the number of bits at which the value of each bit at the corresponding position of the partial reception sequence and the Barker sequence matches. It is assumed that a value obtained by subtracting the number of unmatched bits is used.

Using this correlation calculation method, the threshold value used in the receiving apparatus is set to the value of Barke after the detection signal is generated.
10 and -1 in a period 1 bit shorter than the r-sequence length
Set to 0, otherwise set to 6 and -6.

When the Barker sequence and the gap sequence shown in (1) are used, the correlation when no error occurs can be set to the range as shown in FIG. 13 for the sequence to be communicated. Further, when the threshold value is changed by the method shown in (2), the threshold value is set to a value as shown in FIG. 13 with respect to the correlation when no error occurs. here,
Since the value of the correlation changes by 2 each time a 1-bit error occurs, the correlation changes by 4 when a 2-bit error occurs. Therefore, if the threshold is changed as described above, a detection signal is erroneously generated or erroneously generated even if an error of 2 bits or less is included in a sequence obtained by cutting arbitrary 11 bits from the received sequence. Can be avoided. Further, even in a period in which the threshold value is set high, it is possible to detect a Barker sequence without any error. Therefore, even if a detection signal is erroneously generated, the next correct detection signal can be generated.

Here, after applying the ideas of (1) and (2), as further ideas for implementation, (3) allocation of a specific data sequence to a preamble, (4) pipelining of a correlation detection circuit, Is mentioned.

(3) Allocation of Specific Data Sequence to Preamble In IEEE Std 1394-1995 and arbitration of high-speed serial bus communication conforming to IEEE Std 1394-1995, nine types of states are communicated, so that ten types of symbols can be communicated. In the data communication method of the embodiment, one type of symbol is not used. Therefore, one type of data sequence not used in the arbitration is used as a preamble transmitted before arbitration communication of IEEE Std 1394-1995 and high-speed serial bus communication conforming thereto is started. The preamble is transmitted to stabilize the transmitting / receiving circuit.

In this case, as a data sequence used as a preamble, a case is used in which a Barker sequence having the same inverted state is continuously transmitted without a gap. This data series is a combination in which the deviation of the number of occurrences of 0 and 1 in the data series is 2, and is not very suitable for normal state transmission, so it is used as a preamble.

In this case, at the timing when there is a possibility of transmitting the preamble, that is, during the period until full-duplex communication starts, the partial reception sequence and the Barker
The threshold of the correlation with the sequence is set to +6, -6.

(4) Pipelining of Correlation Detection Circuit The correlation detection circuit compares the correlation with a threshold every time a bit is received for a received sequence, thereby obtaining Ba.
Detect rker sequence. Further, the threshold is changed based on the detection. This process can be broken down into two serial processing stages and processed by a pipeline. That is, since the comparison between the correlation and the threshold value and the change of the threshold value based on the detection are performed in parallel, it is possible to divide the process into two bits. As a result, it is possible to reduce a demand for a processing speed of a circuit that performs each process of comparing the correlation with the threshold and changing the threshold based on the detection.

By the above pipeline, the delay from the input being supplied to the output of the detection signal is a time corresponding to 2 bits. Therefore, the threshold value cannot be changed immediately after receiving the Barker sequence, and the threshold value changes after a delay of 2 bits. However, due to the idea of (1), the correlation between the partial reception sequence and the Barker sequence becomes 2 after the detection signal is generated, as shown in FIG.
In the bit period, the partial reception sequence and Barker
Since the correlation with the sequence is low, the timing for increasing the threshold is not immediately after the generation of the detection signal, but after two bits elapse, and even when an error occurs in any two bits out of 11 consecutive bits, accurate communication is performed. can do.

(Embodiment 5) By using the communication method shown in Embodiment 4, accurate communication can be performed even when an error occurs in any two of 11 consecutive bits in a reception sequence. Is possible. Here, when the error occurrence probability is lower than that, the data transfer rate can be increased by applying a shorter sequence than the Barker sequence used in the fourth embodiment as the specific sequence.

In the present embodiment, even when an error occurs in any one of the consecutive 7 bits in the received sequence, the 7-bit Barker sequence is converted to the specific sequence for the purpose of performing accurate communication. And the same contrivance as in the fourth embodiment is applied.

The procedure of (1) the Barker sequence and the gap sequence to be applied, and (2) the adaptive variation of the threshold value will be described below.

(1) Barker Sequence and Gap Sequence to be Applied A
= 1011000, and B =
0100111 is used. In addition, the following gap sequence is used as a gap sequence corresponding to the Barker sequence and having a low correlation except when the Barker sequence is detected in the received sequence. (A) When A is continuous, the gap sequence of 1 to 4 bits interposed between them is 1, 11, 11
1,1110 is used. (B) When B follows A, a gap sequence of 1 to 4 bits sandwiched between A, 0000, 0
Use 00,0001. (C) When B is continuous, the gap sequence of 1 to 4 bits sandwiched between B is 0000, 00, 00, respectively.
Use 0,0001. (D) When B is followed by A, the gap sequence of 1 to 4 bits interposed between them is 1, 11, 1
11, 1110 are used.

(2) Correlation Calculation Method and Procedure for Adaptive Variation of Threshold Value
6 in the period 1 bit shorter than the Barker sequence length
And -6, otherwise set to 4 and -4. The preamble period is set to 4 and -4.

[0087]

The present invention has the following effects as a method of performing communication on a serial communication path having a high error rate. 1. Communication can be performed accurately without a separate synchronization mechanism. 2. The configuration of a receiver for receiving data can be reduced. 3. Since the Barker sequence or the PN sequence whose frequency spectrum is spread is used as the specific sequence, the influence of electromagnetic noise generated in communication on other devices is small.

Further, since the present invention is particularly useful when the types of symbols to be communicated are small, IEEE SES is performed by mutually communicating nine types of states.
td is also suitable for arbitration of 1394-1995 and high-speed serial bus communication conforming thereto. That is, the present invention enables arbitration in full-duplex bidirectional communication having a high error rate in a communication path, thereby using a single-core optical fiber to use a single wavelength, thereby making an extremely expensive device. IEEE without using
Std 1394-1995 and a high-speed serial bus communication based on the standard can be realized.

[Brief description of the drawings]

FIG. 1 is a diagram showing a transmitted sequence and a correlation in a received sequence with the transmitted sequence in a preferred embodiment using the present invention.

FIG. 2 illustrates the invention as defined by IEEE Std 1394-199.
FIG. 5 is a diagram illustrating a configuration of a transmitter according to a first embodiment, applied to arbitration of high-speed serial bus communication conforming to No. 5;

FIG. 3 illustrates the invention as defined by IEEE Std 1394-199.
FIG. 5 is a diagram illustrating a configuration of a receiver according to the first embodiment, which is applied to arbitration of high-speed serial bus communication conforming to No. 5;

FIG. 4 illustrates the invention as defined by IEEE Std 1394-199.
5 is a timing chart showing the operation of the transmitter according to the first embodiment applied to arbitration of high-speed serial bus communication conforming to No. 5;

FIG. 5 illustrates the method of the present invention as IEEE Std 1394-199.
5 is a timing chart showing the operation of the receiver according to the first embodiment applied to arbitration of high-speed serial bus communication conforming to No. 5;

FIG. 6 illustrates the method of the present invention as IEEE Std 1394-199.
FIG. 5 is a diagram illustrating a configuration of a transmitter according to a second embodiment, which is applied to arbitration of high-speed serial bus communication conforming to No. 5;

FIG. 7 illustrates the invention as defined by IEEE Std 1394-199.
FIG. 5 is a diagram illustrating a configuration of a receiver according to a second embodiment, applied to arbitration of high-speed serial bus communication conforming to No. 5;

FIG. 8 is a diagram illustrating a correlation between a transmitted sequence and a received sequence with respect to the transmitted sequence according to the fourth embodiment of the present invention.

FIG. 9 illustrates the invention as defined by IEEE Std 1394-199.
FIG. 5 is a configuration diagram of a transmitter in a case where a data representation method (A) is used in Example 5 applied to arbitration of high-speed serial bus communication conforming to No. 5;

FIG. 10 illustrates the method of the present invention using IEEE Std 1394-19.
FIG. 25 is a configuration diagram of a receiver in the case where a data representation method (A) is used in Example 95 applied to arbitration of high-speed serial bus communication 95 and the compliant arbitration.

FIG. 11 illustrates the method of the present invention using IEEE Std 1394-19.
FIG. 25 is a configuration diagram of a transmitter in a case where a data representation method (B) is used in Example 95 applied to arbitration of high-speed serial bus communication 95 and compliant with the arbitration.

FIG. 12 illustrates the method of the present invention as IEEE Std 1394-19.
FIG. 25 is a configuration diagram of a receiver in the case where a data representation method (B) is used in Example 95 applied to arbitration of high-speed serial bus communication conforming to 95.

FIG. 13 illustrates the method of the present invention as IEEE Std 1394-19.
FIG. 25 is a diagram illustrating a change in correlation when a sequence is specifically set and a threshold value set corresponding to the sequence in the embodiment 95 applied to arbitration of high-speed serial bus communication conforming to 95.

FIG. 14 is a configuration diagram of a receiver when a communication method disclosed in Japanese Patent Application Laid-Open No. 8-130526 is applied to data communication on a serial communication path as a conventional example.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Receiving part in the conventional receiver 2 Correlator 3a, 3b, 3c, 3d in the conventional receiver Pattern determinator 4a, 4b, 4c, 4d in the conventional receiver 5 Synchronizer in the conventional receiver 5 Decoder 11 in Conventional Receiver 11 Encoder 11a in Transmitter of First Embodiment of Present Invention 11a Encoder 11b in Transmitter of Second Embodiment of Present Invention 11b Data Representation Method (A) in Fourth Embodiment of Present Invention
Encoder 11c in Transmitter in the Case of Using Data In the Fourth Embodiment of the Present Invention Data Representation Method (B)
Encoder in Transmitter in Case of Using Controller 12 Controller 12a in Transmitter of Embodiment 1 of Present Invention Controller 12b in Transmitter of Embodiment 2 of Present Invention Data Representation Method (A) in Embodiment 4 of Present Invention
Control unit 12c in transmitter in the case of using a data representation method (B) in the fourth embodiment of the present invention
Control unit 13 in transmitter in the case of using a specific sequence generator 13a in the transmitter according to the first embodiment of the present invention 13a Barke in the transmitter according to the second embodiment of the present invention
r-sequence generator 13b Data representation method (A) in Embodiment 4 of the present invention
13c of Barker sequence generator in transmitter in the case of using a data representation method (B) in the fourth embodiment of the present invention
Barker sequence generator in transmitter in the case of using .14 Transmitter 15a in transmitter of embodiments 1, 2 and 4 of the present invention 15a Gap sequence generator in transmitter of embodiment 2 of the present invention 15b Embodiment of the present invention 4. Data representation method (A)
Gap sequence generator 15c in transmitter in the case of using a data representation method (B) in the fourth embodiment of the present invention
Gap sequence generator in transmitter in the case of using 16 16 Selector in transmitter in embodiments 1, 2 and 4 of the present invention 17b Data representation method (A) in embodiment 4 of the present invention
Inversion control unit 17c in transmitter in the case of using a data representation method (B) in the fourth embodiment of the present invention
Last inversion state storage and inversion control unit in the transmitter in the case of using 21 Receiving unit in receiver of Embodiments 1, 2, and 4 of the present invention 22 Reception sequence in receiver of Embodiments 1, 2, and 4 of the present invention Storage unit 23 Correlator 23a in receiver of Embodiment 1 of the present invention 23a Correlator 23b in receiver of Embodiment 2 of the present invention 23b Data representation method (A) in Embodiment 4 of the present invention
Correlator 23c in Receiver in the Case of Using Data 23b Data Representation Method (B) in Embodiment 4 of the Present Invention
Correlator 24 in the receiver in the case of using the counter 24a in the receiver of the first embodiment of the present invention 24a Counter 24b in the receiver of the second embodiment of the present invention 24b Data representation method (A) in the fourth embodiment of the present invention
24c in the receiver in the case of using the data 24c in the fourth embodiment of the present invention (B)
The counter 25 in the receiver in the case of using the decoder 25 The decoder 25a in the receiver according to the first embodiment of the present invention 25a The decoder 25b in the receiver according to the second embodiment of the present invention The data representation method (A) in the fourth embodiment of the present invention
Decoder in receiver at the time of use of data 25c Data representation method (B) in the fourth embodiment of the present invention
Decoder 26a in receiver in the case of using a threshold setting unit 26b in the receiver according to the second embodiment of the present invention 26b Data representation method (A) in the fourth embodiment of the present invention
Threshold setting device 26c in the receiver using the data expression method (B) in the fourth embodiment of the present invention.
Threshold setting unit in the receiver in the case of using the data 27 The previous detection signal value storage unit in the receiver in the case of using the data representation method (B) in the fourth embodiment of the present invention

Continued on front page F term (reference) 5K018 AA03 BA01 5K022 EE02 EE22 EE32 5K029 AA01 CC04 GG03 5K047 AA15 BB02 CC01 GG34 GG37 HH01 HH03 HH15 HH42

Claims (17)

[Claims] 1. A data communication method, wherein data to be communicated is represented by a time interval, and a data portion represented by the time interval is communicated by being sandwiched by a first specific sequence having a sharp autocorrelation function. . 2. The data communication method according to claim 1, wherein the sequence of the data portion represented by the time interval is a second sequence having a specific relationship with the first specific sequence having the sharp autocorrelation function. A data communication method using a specific sequence. 3. The communication method according to claim 1, wherein a binary sequence is used as the first specific sequence and the second specific sequence, and the time interval is set in bit units. Characteristic data communication method. 4. The data communication method according to claim 3, wherein, as the first specific sequence, a certain specific sequence and a signal obtained by inverting 0 and 1 of each bit of the specific sequence are used. A data communication method comprising: performing communication in combination with the time interval representing data to be communicated. 5. A two-way data communication method using the data communication method according to claim 3, wherein 0 and 1 of each bit of the first specific sequence used in one communication path are inverted, and A two-way data communication method, wherein the method is performed as the first specific sequence used by the communication path of (1). 6. The data communication method according to claim 3, wherein the first specific sequence is a pseudo-noise sequence (Pseud).
o-Noise Sequence (PN sequence). 7. The data communication method according to claim 1, wherein a Barker sequence is used as the first specific sequence. 8. The data communication method according to claim 3, wherein a partial reception sequence obtained by cutting out a continuous portion having a length equal to the first specific sequence from a reception sequence and the first specific sequence are obtained. A data communication method using a value obtained by subtracting the number of bits that do not match from the number of bits that match the value of each bit at a position corresponding to each of the partial reception sequence and the first specific sequence as the correlation. . 9. The data communication method according to claim 7, wherein
A data communication method for communicating types of information, wherein A = 11100010010 as the first specific series.
And B = B
A data communication method using 00011101101 and partially or entirely using the following (1) to (4). (1) When A is consecutive, 1, 11, 4 bits are interposed between them as the second specific sequence of 1 to 4 bits, respectively.
110 and 1101 are used. (2) When B follows A, as the second specific sequence of 1 to 4 bits interposed between them, 1, 1
0, 100, and 1001 are used. (3) When B is continuous, 0000, 00, and 2 are respectively specified as the second specific sequence of 1 to 4 bits interposed therebetween.
001,0010 is used. (4) When B is followed by A, 0, 0, and 2 are respectively specified as the second specific sequence of 1 to 4 bits interposed therebetween.
1,011,0110 is used. 10. The method according to claim 7, wherein:
A data communication method for communicating 0 types of information, wherein A = 1011000 as the first specific sequence, and B = 0100 which is a sequence obtained by inverting 1,0 of each bit.
111. A data communication method using the following (1) to (4) partially or entirely. (1) When A is consecutive, 1, 11, 4 bits are interposed between them as the second specific sequence of 1 to 4 bits, respectively.
111 and 1110 are used. (2) In the case where B follows A, 0, 0, and 1- to 4-bit second specific sequences interposed between them, respectively.
Use 00000,0001. (3) When B is continuous, 0000, 00, and 2 are respectively specified as the second specific sequence of 1 to 4 bits interposed therebetween.
000,0001 is used. (4) When A follows B, 1 to 4 bits of the second specific sequence interposed therebetween are 1, 1
1,111,1110 are used. 11. A data communication system for performing data communication using the data communication method according to claim 1, wherein said transmitting device converts said data to be transmitted into said time interval, and said first specific sequence. And means for transmitting the data converted into the time interval with the first specific sequence interposed therebetween. In the receiving device, a continuation having a length equal to the first specific sequence from the received sequence is provided. By detecting whether the correlation between the partial received sequence obtained by cutting out the part to be extracted and the first specific sequence exceeds a threshold value, the first received sequence is extracted from the received sequence.
And a means for detecting a specific sequence and generating a detection signal at a corresponding timing; and a means for restoring data from a time interval obtained by subtracting the first specific sequence length from a time interval between the detection signals. A data communication system, comprising: 12. A data communication system using the communication method according to claim 5, comprising two communication paths for performing bidirectional communication, wherein each bit of the first specific sequence used in one communication path. A two-way data communication system characterized in that an inverted version of 1, 0 is used as the first specific sequence used by the other communication path. 13. A data transmitting apparatus for performing data communication using the data communication method according to claim 1, wherein said means for converting data to be transmitted into a time interval, and storing said first specific sequence. And a means for transmitting the data converted into the time interval with the first specific stream interposed therebetween. 14. A data receiving apparatus for performing data communication using the data communication method according to claim 1, wherein the data receiving apparatus is obtained by cutting out a continuous portion having a length equal to the first specific sequence from a received sequence. A partial reception sequence;
Means for detecting the first specific sequence from the received sequence by detecting that the correlation with the specific sequence exceeds a threshold, and generating a detection signal at a corresponding timing, Means for restoring data from a time interval obtained by subtracting the first specific sequence length from the time interval. 15. The receiving apparatus according to claim 14, wherein the partial reception sequence obtained by cutting out a continuous portion having a length equal to the first specific sequence from the reception sequence,
Detecting the first specific sequence from the received sequence by detecting that the correlation with the specific sequence exceeds the threshold value, and generating a detection signal at a corresponding timing by changing the threshold value. A data receiving device having a function. 16. The receiving apparatus according to claim 15, further comprising: calculating the correlation using the method according to claim 8, and defining the first specific sequence and the second specific sequence in claim 9. The threshold value is set to 10 and -10 during a period of one bit shorter than the first specific sequence length after the detection signal is generated, and is set to 6 and -6 otherwise. Data receiver. 17. The receiving device according to claim 15, further comprising: calculating the correlation using the method according to claim 8, and defining the first specific sequence and the second specific sequence as defined in claim 10. The threshold value is set to 6 and -6 during a period of one bit shorter than the first specific sequence length after the detection signal is generated, and is set to 4 and -4 otherwise. Data receiver.