JPH0813020B2 - Optical bus transmission system and transmitter-side encoder and receiver-side decoder for implementing the same - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical bus transmission system for converting a digital electrical signal of Manchester by phase into an optical signal and transmitting it through an optical bus, and an encoder on a transmission side and a decoder on a reception side for implementing the system.

[0002]

2. Description of the Related Art As a LAN (Local Area Network) introduced for communication of digital data between various electronic devices mounted on an aircraft, DATAC (Digital Au)
tomousTerminal Access Communication) method is known (for example, magazine "NEC technique" Vol. 39, No. 1).
2, 1986, 160-167). In this DATAC system, data is transmitted as a Manchester-by-phase digital signal. This Manchester-by-phase digital signal is transmitted as a pulse train of two trains of a TXO signal and a TXN signal, and therefore a TX transmission line is also used as a TX transmission line.
Two were required, one for transmitting the O signal and one for transmitting the TXN signal.

[0003]

It is possible to use an optical bus as the bus in this DATAC system.
An optical bus cannot transmit in both directions with a single bus because of the optical coupler with the terminal. Therefore, two optical transmission lines are required to transmit the TXO signal in both directions.
Two optical transmission lines are required to transmit the N signal, and the number of optical transmission lines is four, which increases the scale.

In addition, when many terminals are connected to the optical bus, the loss of the optical coupler is relatively large, so that the received optical signal from the near terminal and the received optical signal from the distant terminal are different from each other. If the levels are so different that even a small level optical signal can be received, the receiver may be saturated with a large level optical signal, causing large waveform distortion, and it may not be possible to obtain a correct pulse.

Similarly, in order for each terminal to be able to receive at a sufficient level, it is necessary to supply an optical signal with a large optical power to the optical bus, but conventional electrical signal pulses are used for that purpose. In that case, a signal with a large optical power cannot be output. If signals are output to the bus from multiple terminals at the same time, correct data cannot be obtained.
Therefore, each terminal starts transmission after detecting a state where there is no carrier on the bus (bus quiet). If some signal is sent to the bus at the same time, a so-called signal collision state, this is detected and the signal transmission is stopped. However, the conventional method of directly replacing an electric signal with an optical signal is to detect a collision between these two signals when the pulse width is relatively wide and a large level signal and a small level signal partially overlap each other. Was difficult.

It is an object of the present invention to provide an optical bus transmission system capable of reducing the scale of the optical bus and having a large dynamic range of reception level, and a transmission side encoder and a reception side decoder for implementing the same. .

[0007]

According to the present invention, in the terminal, the first digital signal TXO of the first and second digital signals TXO and TXN in the Manchester by phase is selected.
Two narrow pulses that are respectively synchronized with the rising and falling edges of the first pre-synchronization pulse provided at the beginning of the pulse, and a narrow pulse that is synchronized with the rising edge of each pulse following the first pre-synchronization pulse, and , A narrow pulse for each constant time during the duration of each pulse is created, these narrow pulses are converted into an optical signal and transmitted to the optical bus as a single optical pulse train. Converts the optical pulse received from the optical bus into an electrical signal,
The first two pulses of the electric signal and the subsequent pulses are separated, the first pre-synchronization pulse is made in synchronization with the former two pulses, and each pulse of the latter is a pulse whose pulse width is the above fixed time. To regenerate the digital signal TXO by converting it to
By reversing the polarity of the reproduced digital signal TXO, the digital signal TXN is reproduced.

The transmitter encoder which converts the first digital electric signal into the optical pulse train and sends the optical pulse train, confirms that one signal TXO of the Manchester-by-phase signal is kept at a low level for a predetermined first period or longer. Terminal gap detecting means for detecting and outputting a gap detection signal of a predetermined logic and holding the gap detection signal until the rising edge of the first pre-synchronization pulse, and the first pre-position in response to the gap detection signal. First pre-sync pulse extraction means for extracting and outputting a sync pulse, rising detection means for detecting a rise of the extracted first pre-sync pulse, and generating a first pulse, and the extracted first sync pulse 1 falling edge detecting means for detecting the falling edge of the pre-synchronization pulse and generating a second pulse; and the first edge in response to the gap detection signal.
First pre-sync pulse removing means for outputting the signal TXO from which the pre-sync pulse has been removed, and respective logic pulses of the signal TXO from which the first pre-sync pulse has been removed, at their respective rising times, and thereafter. While at a high level, the pulse generating means for generating a third pulse for each 0.5 bit length of the Manchester-by-phase signal and the first, second and third pulses are combined in a line and respond to the pulses respectively. And a pulse train generating means for outputting a train of narrow pulses shorter than 0.5 bit length of the Manchester by phase signal as a pulse train to be converted into the optical pulse train.

A receiving-side decoder that receives the optical pulse train and converts the optical pulse train into first and second digital electric signals detects that the electric pulse train signal remains at a low level for a predetermined period or longer, and then determines a predetermined level. A bus quiet signal that outputs a logical signal and holds the bus quiet signal until the rising edge of the first pulse of the electric pulse train signal; and a passage of the first pulse and the second pulse of the electric pulse train signal. A first gate means, a first front synchronizing pulse generating means for generating a first front synchronizing pulse which rises at the first pulse from the first gate means and falls at the second pulse; and After the gate means has passed the second pulse, the gate means for closing the first gate means in response to the trailing edge of the first pre-sync pulse. Gate control signal generating means for generating a control signal, and a second gate which is opened by being applied with the gate control signal from the gate control signal generating means and which passes the electric pulse train signal after the second pulse. Means, pulse generating means for generating a 0.5-bit length pulse of the Manchester-by-phase signal in response to each pulse that has passed through the second gate means, and the first pre-synchronization pulse generating means. Combining means for combining the first pre-sync pulse and the pulse from the pulse generating means into one pulse train to output as the one signal TXO of the demodulated Manchester-by-phase signal, and the demodulated signal TXO. Is inverted and output as the other signal TXN of the demodulated Manchester-by-phase signal.

[0010]

[Operation] Since the first and second digital electric signals in the Manchester-by-phase are converted into a series of optical pulses and transmitted, a required optical bus scale is half that of the conventional one.

[0011]

FIG. 1 shows an embodiment of the present invention. Optical bus 11
Is composed of an optical fiber transmission line 12 for rightward transmission and an optical fiber transmission line 13 for leftward transmission, and a plurality of terminals 14 are coupled to the transmission lines 12 and 13 by optical couplers 15 and 16, respectively. In each terminal 14, the conventional DATAC
The part connected to the bus in the method, that is, DATAC
Terminal 17 is provided and DATAC terminal 17
Data is transmitted and received between the subsystem 18 and the bus via the. DATAC terminal 17 is subsystem 1
Manchester-by-phase digital signals TXO and T in which the data to be transmitted from 8 has a predetermined format
The digital signals RXI and RXN of the Manchester by phase output as XN are also converted into data and supplied to the subsystem 18.

The message structure of the data before conversion to the digital signals TXO and TXN sent from the DATAC terminal 17 is the same as that shown in the above-mentioned document, and is as shown in FIG. The words to be transferred are classified into two, that is, a label word LW and a data word DW, as shown in lines C and D in the figure, each having a length of 20 bits, and the label word LW has 3 bits (high level 1. 5 bits + low level 1.5 bits) Synchronous SYN, 4
It consists of a bit label extension LEX, a 12-bit label LAB, and a 1-bit parity PB.
The data word consists of 3-bit synchronous SYN, 16-bit data DATA and 1-bit parity PB. As shown in lines B and C in the figure, the word string WS is constituted by the label word LW and the plurality of data words DW, and the word string WS is the minimum transmission unit. Each word string WS has a label word R of 1 word at the beginning
W, followed by 0-255 data words DW in succession. One terminal 17 has 1 to 31 word strings WS as shown in rows A and B of FIG.
Can be continuously transmitted, and a transmission unit composed of the plurality of word strings WS is called a message MS.
There is a 4-bit string gap SG between a word string WS and the next word string WS in the message MS. Message MS and next message M
A terminal gap TG of 8 bits or more is provided between S and S.

The DATAC terminal 17 outputs such a message MS as Manchester-by-phase digital signals TXO and TXN. To facilitate the understanding of the present invention, an example of these digital signals TXO and TXN described above in a similar invention for the same problem as the invention shown in Japanese Patent Application No. 2-165294 of the previous application is shown in FIG. A pre-synchronization pulse period PP is provided immediately before the start of the next message MS after the terminal gap TG, and during that period PP, the pre-synchronization pulse (P
SSP) is generated. Then, in synchronization with the falling edge of the pulse PSSP, the H-level is maintained for 1.5 bits on the rising edge and then the L-level is maintained for 1.5 bits on the synchronizing signal SYN.
Output to O. After that, “1” of each data bit,
A Manchester code in which the first half or the latter half of each bit is at a high level according to "0" follows, and finally the parity bit P
B is added. TXN has the polarity of TXO inverted immediately after PSSP, and finally a pulse of 0.5 bit width, which is the inverted parity bit PB of TXO, is added. The width of the pre-synchronization pulse (PSSP) existing immediately before each message MS can change in the range of 100 nS to 500 nS, but the other pulses have a pulse width of an integral multiple of 250 nS.

The DATAC terminal 17 extracts necessary data from the received Manchester-by-phase digital signals RXI (same as TXO) and RXN (same as TXN) and sends them to the subsystem 18. In the previous invention, DA
Manchester-by-phase digital signals TXO and TXN from TAC terminal 17 and 32 MHz (with a period of 3
Clock TICK of 1.25 nS) is input to the encoder 19, and a head pulse of TXN, that is, a pulse having a narrow width at the leading edge of the pre-synchronization pulse PSSP is output, as shown as the output ETX of the encoder 19 in FIG. Then T
A narrow pulse is output at the rising edge of each XO pulse, and the TXO pulse continues (during the pulse width),
A narrow pulse is output every 250 nS in this example for a fixed time after the rising edge. These are a single pulse train ETX
It is said.

Returning to the explanation of FIG. 1, the narrow pulse train ETX from the encoder 19 drives the light emitting elements 33 and 34 such as light emitting diodes through the drive circuits 31 and 32 to be converted into an optical pulse train. It is sent to the optical transmission lines 12 and 13 via the optical couplers 15 and 16, respectively. On the other hand, the optical pulse transmitted through the optical transmission lines 12 and 13 is branched by the optical couplers 15 and 16, respectively, and the terminal 14
Incident on a light receiving element 35 such as a photodiode and converted into an electric signal, the electric signal is amplified and shaped by a receiving circuit 36, and an output pulse train RP of the receiving circuit 36.
The S is reproduced by the decoder 37 into a signal RXI corresponding to the signal TXO on the transmission side and a signal RXN corresponding to the signal TXN. The reproduced signals RXI and RXN are DATAC.
It is supplied to the terminal 17.

The reproduction by the decoder 37 is performed with reference to each edge (for example, the leading edge) of the input pulse. In the absence of a received signal, that is, the narrow pulse first received from the bus quiet of 8 bits or more is the rising edge of the pre-synchronization pulse (PSSP) of the signal TXN, and the narrow pulse input thereafter is the narrow pulse of the signal TXO. It is a pulse corresponding to the rising edge of each pulse or the duration of the pulse width following the rising edge. Therefore, the narrow pulse after the pulse corresponding to the rising of the PSSP is used as a reference, and the interval of the narrow pulse generated between the pulse widths on the transmission side, the pulse having the pulse width of 250 nS in the above example. The signal RXI corresponding to the signal TXO on the transmission side is reproduced, and the signal RXN corresponding to the signal TXN is reproduced from the signal having the polarity inverted of the signal RXI and the narrow pulse at the head.

FIG. 4 is a time chart for explaining an embodiment of the optical bus transmission system according to the present invention. In this invention D
Two-phase signal TXO output from ATAC terminal 17
Pre-synchronization pulse PPSS is also applied to the signal TXO of TXN and TXN.
P is provided. That is, as the Manchester-by-phase signal, the pre-synchronization pulse P is added to the signal TXN as shown in FIG.
Not only the SSP is provided, but also the pre-sync pulse PPSSP (referred to as a main pre-sync pulse) which falls prior to this pre-sync pulse PSSP and coincident with the rising thereof is also provided to the signal TXO. . In such a case, the timing information at the time of rising of the pre-sync pulse PSSP of the signal TXN is the main pre-sync pulse PP of the signal TXO.
Since it can be obtained from the fall of SSP, in the present invention, the signal TXN is not used, a series of transmission pulses are generated and transmitted only from the signal TXO, and the receiving side receives the Manchester two-phase signal RXI, Decode RXN. That is, in the encoder 19 of FIG.
The main pre-synchronization pulse PPSSP of the signal TXO from the AC terminal 17 is detected and a narrow pulse indicating its rising and falling is generated, and thereafter, the rising of each pulse in the signal TXO and a certain period within its duration. Narrow pulse train ETX by generating a narrow pulse every time
Is output. On the reception side, the decoder 37 reproduces the main pre-synchronization pulse PPSSP and pre-synchronization pulse PSSP from the reception pulse train RPS (same as ETX), and further Manchester biphase signals RXI and RX following them.
Decode each N.

FIG. 5 shows a configuration example of the encoder 19 in the transmission system shown in FIG. 4, and its operation is shown in a time chart in FIG. The signal TXO is applied to the reset terminal of the bus quiet detection circuit 22 composed of a counter.
When the counter 22 counts 64 (2 μS) or more clocks TICK having a cycle of 31.25 nS (32 MHz), its output B
Q is set to a high level to show that the terminal gap is TG. FIG. 6 shows a state in which the flip-flop 21 is reset by the high level of the output BQ. In this state, the first pulse of signal TXO (PPSS
When P) is input, it is given to the flip-flop 24A through the gate 26B. At the rising edge of the pulse PPSSP, the H level is written in the flip-flop 24A, and the Q output immediately resets the flip-flop 24A. That is, the flip-flop 24A detects the rising of the main pre-synchronization pulse PPSSP and outputs a short pulse to its output D5. Similarly, the main pre-sync pulse PPSSP
Falling edge is detected by the flip-flop 24B, and a short pulse is output to the output D6. On the other hand, this pulse PPS
At the falling edge of SP, the Q output of the flip-flop 21 becomes high level and the gate 26A is opened and the gate 26B is closed.
Therefore, thereafter, the signal TXO is applied to the reset terminal of the 4-bit Johnson counter 28A through the gate 26A. 4-bit Johnson counter 28A is clock TI
Every time four CKs are counted, the logic level of the output Q1 is inverted, and the outputs Q2, Q3, and Q4 are sequentially delayed from the output Q1 by one clock, and the same level is inverted. The Q1 output and Q2 output of the counter 28A are the logic gate 28B.
When the signal D2 applied to the reset terminal of the counter 28A is at a high level, a pulse is generated every 8 clocks of the clock TICK at the output D4 of the gate 28B. The pulses in the outputs D4, D5, D6 of the gate 28B and the flip-flops 24A, 24B are given to the pulse generating circuit 29 through the gate 25, and the pulse generating circuit 29 responds to each input pulse by the clock TI.
A high level pulse is generated during the period of counting two CKs (62.5 nS). The pulse train generated thereby is output as the output signal ETX of the encoder 19. In this way, the encoder 19 generates a narrow pulse at the rising and falling of the main pre-synchronization pulse PPSSP (the latter corresponds to the rising of the pre-synchronization pulse PSSP), and thereafter the signal T
A narrow pulse is output every 250 nS (8 clocks TICK) while each rising edge of XO and the rising high level continues.

FIG. 7 shows a configuration example of the decoder 37 in the transmission system shown in FIG. 4, and its operation is shown in a time chart of FIG. In the example of this decoder, the bus quiet detection circuit 42 that constantly counts the clock RICK is reset by the signal RXI which is the decoding result, and output BQ when the signal RXI is not output for 2 μS or more.
To a high level. FIG. 8 shows a state in which the output BQ is at a high level after the bus quiet period of 2 μS or more has elapsed in the terminal gap TG. Therefore, the flip-flop 52 is in the reset state and its low level Q
The gate 44 is closed and the gate 46 is opened by the output. The receive pulse RPS is the first of the pulses labeled D1 through the gate 46 which causes the Q output of the flip-flop 47 to go high and inverts to a low level at the next pulse to cause the main presync pulse to go to D2. PPSSP is played. The main pre-synchronization pulse PPSSP is output through the gate 51, and the falling edge thereof causes the flip-flop 52 to read the H level, the Q output signal D3 becomes high level, and the gates 44 and 46 are inverted to close and open, respectively. To be done. Therefore, after that, the received pulse
S is supplied to the pulse generation circuit 45 as a signal D4 through the gate 44. The pulse generation circuit 45 outputs a pulse having a width of 250 nS each time a narrow pulse (signal D4) is input. Therefore, when the input pulse is sequentially given every 250 nS as shown in the signal D5, the number of pulses is increased. A pulse having a corresponding width is generated. The output signal D5 of the pulse generation circuit 45 is output as the signal RXI through the gate 51 and logically inverted by the NOR gate 49 to generate the signal RXN.
Is output as The output signal RXI of the gate 51 is given to the reset terminal of the counter 48 which constantly counts the clock RICK. The counter 48 has a clock RICK
When 32 or more (1 .mu.S) are counted, it is determined that the string gap SG or the terminal gap TG, and the output D6 is set to high level to force the signal RXN to low level. In this way, the decoder 3 shown in FIG.
Reference numeral 7 decodes the signal RXI having the first pre-synchronization pulse PPSSP at the beginning and the signal RXN having the pre-synchronization pulse PSSP at the beginning from one series of reception pulses RPS.

As described above, according to the present invention, the digital electric signal TXO having the main pre-synchronization pulse PPSSP as the timing information is transmitted as one optical signal, and the one optical signal is transmitted to the two Manchester signals. Bi-phase signals RXI and RXN can be reproduced, and thus 2
The scale of the optical bus can be reduced to one half as compared with the case where the book signals TXO and TXN are separately transmitted as optical signals.

Since the original signals TXO and TXN are transmitted in a pulse train having a narrower width than that of the original signals, and the receiving side processes each pulse as a mere timing signal at its rising edge,
Even if the pulse width is distorted due to the level of the input optical signal being large and the limiter amplifier in the receiving circuit 36 being saturated, if the pulse width is smaller than the pulse width of the signal TXO, the signal can be reproduced correctly and the input level The dynamic range is wide. That is, since the optical signal input to each optical coupler is split into two optical signals and output, the optical signals that have passed through the many optical couplers provided in the optical bus 11 (FIG. 1) and the few optical couplers are output. With the optical signal that has passed,
Although the levels of the input optical signals are significantly different, both of these signals can be reproduced correctly. When the pulse of each bit of the original signal TXO is converted into an optical signal having a pulse width (62.5 nS) that is a quarter of its pulse width (250 nS), the receiving circuit 3
The allowable distortion in 6 is 4 times the pulse width of the conventional one.

As described above, the duty ratio of the optical signal is considerably smaller than the duty ratio (about 50%) of the original signal TXO. Therefore, as compared with the case where the original signal TXO is directly converted into an optical signal, the mark ratio of the optical signal, that is, the ratio of the time during which the current is flowing through the light emitting element becomes smaller. When a light emitting diode is used as the light emitting element, the current value that can be passed is limited due to the limit of the junction temperature. However, for an extremely short time, it is possible to flow a current equal to or more than the current value at which continuous energization is possible. For example, if the rated current is 50mA with continuous energization, if the mark ratio of the optical signal is set to ⅛ and the pulse width is made sufficiently small, a current with a peak current of 400mA can flow and the optical power can be increased eight times. can do.
In the present invention, since the optical pulse train is output with a narrow width, the optical power can be increased correspondingly, and many optical couplers can be coupled.

When signals are simultaneously transmitted from the two terminals 14 on the optical bus and a signal collision occurs, it is necessary to detect this, stop the signal transmission, and discard the received signal as error data. For example, as shown in FIG. 9, it is assumed that the pulses of the original signal TXO1 of the incoming signal from the near terminal and the original signal TXO2 of the incoming signal from the far terminal have a timing relationship in which they partially overlap as shown in rows A and B. In this case, the received signal RPS obtained by directly converting the original signals TXO1 and TXO2 into optical signals without changing the pulse widths of them is a pulse of large level P ₁ from the near terminal to the pulse from the far terminal as shown in row C. A small level P ₂ pulse partially overlaps, and this P
The level difference between ₁ and P ₂ is, for example, about 1000: 1. It is extremely difficult to detect such signal overlap. However, in the present invention, the width of each pulse of the original signal TXO is set to, for example, 1/4. Therefore, in this example, as shown in row D, a large level optical pulse from the near terminal and a far terminal from the far terminal are provided. It is easy to detect each of these because they do not overlap the small level light pulses of. In Manchester code, there are no two pulses in one bit (500 nS), so the reception of the signal shown in row D indicates that a signal collision has occurred,
Therefore, the signal collision can be easily detected. If the two optical signals completely overlap, signal collision cannot be detected. It is clear that the lower the mark ratio of the optical signal, the easier the signal collision can be detected.

[0024]

As described above, according to the present invention, the digital signals TXO and TXN in the Manchester by phase are selected.
On the other hand, a series of pulses formed from TXO can be transmitted as an optical signal, and the original Manchester-by-phase digital signals TXO and TXN can be reproduced on the receiving side, and the scale of the optical bus can be reduced. Moreover, since the pulse is transmitted with a narrow pulse and processed at the timing of its rising edge on the receiving side, a wide dynamic range of the receiving level can be obtained. Further, since it is a narrow pulse, it is possible to output an optical pulse of strong power. Further, since the mark ratio of the light pulse is small, it is easy to detect a signal collision.

[Brief description of drawings]

FIG. 1 is a block diagram for explaining an embodiment of the present invention.

FIG. 2 is a diagram showing a message structure of the DATAC method,

FIG. 3 is a time chart showing an example in which bi-phase signals TXO and TXN are converted into a single pulse train by the encoder in the invention of the previous application.

FIG. 4 is a time chart for explaining an optical bus transmission system according to an embodiment of the present invention.

FIG. 5 is a block diagram showing a configuration example of an encoder for implementing the system of the present invention.

FIG. 6 is a time chart for explaining the operation of the encoder of FIG.

FIG. 7 is a block diagram showing a configuration example of a decoder for implementing the system of the present invention.

8 is a time chart for explaining the operation of the decoder of FIG.

FIG. 9 is a time chart showing an example of signal collision.

[Explanation of Codes] 11 Optical Bus 14 Terminal 15 Optical Coupler 16 Optical Coupler 17 DATAC Terminal 18 Subsystem 19 Encoder 21 Flip-Flop 22 Counter 24A Flip-Flop 24B Flip-Flop 28A 4-bit Johnson Counter 29 Pulse Generation Circuit 31 Drive Circuit 32 Drive Circuit 32 Drive Circuit 36 Reception Circuit 37 Decoder 42 Bus Quiet Detection Circuit 45 Pulse Generation Circuit 47 Flip Flop 48 Counter 52 Flip Flop

Claims (8)

[Claims] 1. Each of the terminals connected to the optical bus has a first and a second pre-synchronization pulse at the head thereof, respectively. Of the first and second digital electric signals of the Manchester by phase, the head of the first digital electric signal is included. A narrow pulse synchronized with the rising and falling of the first pre-synchronization pulse, a narrow pulse synchronized with the rising of each pulse following the head pulse, and a constant value during the continuation of each subsequent pulse. A narrow pulse for each time is created, the narrow pulse is converted into an optical signal and transmitted to the optical bus as a series of optical pulse trains, and each terminal receives the optical pulse received from the optical bus as an electrical pulse. It is converted into a signal and separated into the first two pulses of the electric pulse signal and the subsequent pulses, and a first pre-synchronization pulse whose rising and falling are synchronized with the above two pulses is made, The pulse after the pulse is converted into a pulse having a pulse width of the fixed time, reproduced as the first digital signal together with the first pre-synchronization pulse, and the polarity of the reproduced first digital electric signal is inverted to reproduce the first digital electric signal. 2 Optical bus transmission system that reproduces as a digital signal. 2. The optical bus transmission system according to claim 1, wherein the narrow pulse is selected to have a width equal to or less than a quarter of one bit of the Manchester by phase code. 3. The optical pulse train from each of the terminals is sent to both the first bus and the second bus of the optical bus via first and second optical couplers, respectively, and the first and second buses are provided. 2. The optical bus transmission system according to claim 1, wherein the optical signals are received from the respective optical signals via the first and second optical couplers. 4. A Manchester bi-phase signal having one first pre-sync pulse at the beginning TXO is converted into an optical pulse train and transmitted through an optical bus, and the received optical pulse train is converted into an electrical pulse train signal, which is then converted into Manchester. An encoder on the transmission side that implements an optical bus transmission method that demodulates the bi-phase signals RXI and RXN, and detects that one signal TXO of the Manchester bi-phase signal continues to be at a low level for a predetermined first period or longer. And outputs a gap detection signal of a predetermined logic and holds the gap detection signal until the rising of the first pre-synchronization pulse, and the first pre-synchronization in response to the gap detection signal. A first pre-synchronization pulse extracting means for extracting and outputting a pulse, and a rising edge of the extracted first pre-synchronization pulse. Rising edge detecting means for detecting a falling edge of the extracted first pre-synchronization pulse and generating a second pulse, and In response, a first pre-sync pulse removing means for outputting the signal TXO from which the first pre-sync pulse has been removed, and respective logic pulses of the signal TXO from which the first pre-sync pulse has been removed, A pulse generating means for generating a third pulse at every 0.5 bit length of the Manchester by phase signal between the rising time and the subsequent high level, and the first, second and third pulses are combined in a line, Pulse train generating means for outputting a train of narrow pulses shorter than 0.5 bit length of the Manchester by phase signal as a pulse train to be converted into the optical pulse train in response to each of the pulses; Encoder, including. 5. The first pre-sync pulse extraction means is a first gate means for passing the first pre-sync pulse, and the first gate means is responsive to a trailing edge of the first pre-sync pulse. Gate control means for generating a gate control signal for closing the signal TXO, and the first pre-synchronization pulse removal means receives the gate control signal from the gate control means to release the passage prohibition of the signal TXO. An encoder according to claim 4 including gating means. 6. The pulse train generating means combines the first pulse from the rising edge detecting means, the second pulse from the falling edge detecting means, and the third pulse from the pulse generating means into one row. The OR circuit that operates and the narrower than the minimum pulse width of the original signal of the Manchester by phase signal in response to each pulse of the output of the OR circuit, for example, 0.25.
An encoder according to claim 4 or 5, further comprising a narrow pulse generation circuit that outputs the pulse train by generating a narrow pulse having a bit length. 7. A Manchester-by-phase signal, one signal TXO having a first pre-sync pulse at the beginning is converted into an optical pulse train and transmitted through an optical bus, and the received optical pulse train is converted into a pulse train electric signal, and Manchester is then transmitted. The above-mentioned one signal RXI and the other signal RXN of the bi-phase signal
A decoder on the receiving side that implements an optical bus transmission system for demodulating, outputting a bus quiet signal of a predetermined logic by detecting that the electric pulse train signal continues to be at a low level for a predetermined period or more, A bus quiet detection means for holding the bus quiet signal until the rising of the first pulse of the electric pulse train signal; a first gate means for passing the first pulse and the second pulse of the electric pulse train signal; Rising on the first pulse from the gate means,
First pre-synchronization pulse generating means for generating a first pre-synchronization pulse falling at the second pulse, and the first pre-synchronization after the first gate means has passed the second pulse The first in response to the trailing edge of the pulse
A gate control signal generating means for generating a gate control signal for closing the gate means, and the gate control signal from the gate control signal generating means are provided to be opened so as to open the electric pulse train signal after the second pulse. Second gate means for passing, pulse generating means for generating a 0.5 bit long pulse of the Manchester by phase signal in response to each pulse passing through the second gate means, and the first pre-synchronization pulse Combining means for combining the first pre-synchronization pulse from the generating means and the pulse from the pulse generating means into one pulse train and outputting as the one signal RXI of the demodulated Manchester-by-phase signal; Inversion means for logically inverting the demodulated signal RXI and outputting it as the other signal RXN of the demodulated Manchester by phase signal. , Decoder, including the capital. 8. A gap detecting means is provided for detecting that the demodulated signal RXI is kept at a low level for a predetermined period or more and forcibly making the demodulated signal RXN a low level. The decoder according to claim 7.