JPS63120542A - Signal collision detection system - Google Patents

Abstract

PURPOSE:To automatically adjust the amplitude of a waveform to cancel an its own station sending signal by branching an electric pulse signal to drive an optical element at the sender side, using the identified high-order bit as a required reception digital signal and using the loworder bit as a voltage or the like for adjusting the amplitude of the branch signal automatically. CONSTITUTION:The electric pulse signal to drive the optical element 12 at the transmission side is branched to generate the signal whose polarity, phase and amplitude are adjusted. The signal is added to a pulse waveform signal at a remote end by an adder circuit at the near end reception circuit and waveform-equalized and 2-bit identification is applied and the high-order bit is used as a required reception digital signal, the polarity of low-order bit and the driven electric pulse at the sender side are collated to use as a voltage for the automatic amplitude adjustment of the branch signal. Moreover, the collation between the high-order bit and the low-order bit is taken to use the result as the voltage for gain control of the ACC amplifier 32 of the reception circuit and the presence of both a transmission clock and a reception clock is detected. Thus, even with the fluctuation of transmission optical signal level and transmission line loss, normal collision detection is attained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a collision detection method when performing pulse burst transmission in an optical transmission system using a star coupler.

[Conventional technology]

Conventionally, as shown in Fig. 2, terminal station 1 (T l, T 2 ,
In order to perform f-evening transmission using optical pulse berths between river TN), a method has been proposed in which the optical output of each terminal station is distributed to all terminal stations and received by using Star Kagura 2. . In this method, if one terminal station sends out a pulse burst while another terminal station sends out a pulse burst, there is a risk that these will be superimposed and cause errors in the data. need to be resent. For this reason, it is necessary to recognize the collision sound of the pulse burst at the own terminal station, or to recognize it at the receiving terminal station and notify the transmitting side of the reception failure.

The method of detecting a collision at the own terminal station has an advantage over Doma's efficiency because it can instantly recognize the collision. However, if the receiving circuit of the own terminal station has a received signal detection circuit, it can recognize the pulse burst transmission from other stations, but even if only the own station is transmitting, its own signal will be received, so the other station will not be able to transmit pulse bursts. There is a problem in that it is difficult to distinguish between collisions and collisions with other signals.

Therefore, in an optical transmission system using a star coupler, it is necessary to detect the collision of the sending signals between each terminal station that occurs when performing pulse burst transmission, distinguishing it from the signal sent from the own station and returning to the own station via the star coupler. A collision detection method that can detect collisions is required.

Conventionally, such a collision detection method uses a single star coupler as shown in Figure 3 to perform pulse burst transmission between multiple terminal stations that have an electro-optical conversion circuit in the transmitting section and an optical-electrical conversion circuit in the receiving section. In the system, the electric pulse signal that drives the optical element on the transmitting side of each terminal station is completely branched to generate a signal whose polarity, phase, and amplitude are adjusted, and this signal is optically converted and received by the receiving circuit of the own station. Add to the pulse signal waveform, equalize the waveform of the obtained signal, extract and identify the reception timing, obtain the desired reception pulse output, and confirm that both the transmission clock of the transmitter and the reception clock of the receiver exist. A collision detection method has been proposed (Japanese Patent Application No. 59-030690).

[Problem that the invention seeks to solve]

This method can be expected to be very effective when the local station sending signal level and transmission path loss are constant, but it cannot follow these fluctuations, so it may become unstable under poor conditions. be.

SUMMARY OF THE INVENTION An object of the present invention is to improve the above-mentioned drawbacks, and to provide a signal collision detection method that automatically adjusts the amplitude of a waveform for canceling the signal transmitted by the own station.

[Means for solving problems]

The invention is a system that uses one star coupler to perform pulse burst transmission between a plurality of terminal stations having an electro-optical conversion circuit in the transmitting section and an opto-electric conversion circuit in the receiving section, and an optical element on the transmitting side. and its driving circuit, a receiver (g element, automatic gain control (AGC) amplifier, adder circuit, reception equalization circuit, 2-pit identification circuit, and timing extraction circuit are provided on the receiving side, and an electric pulse that drives the optical element on the transmitting side. The signal is branched to generate a signal whose polarity and phase are manually adjusted, the amplitude is manually adjusted, and the amplitude is automatically adjusted, and this signal is added to the pulse waveform signal from the far end in the adder circuit of the receiving circuit at the near end. After equalizing the waveform of the signal, all 2 bits are identified, and the identified upper bits are used as the required received digital signal, and at the same time, the polarity of the lower bits is correlated with the driving electric pulse on the transmitting side to generate the branch signal. A voltage for automatic amplitude adjustment of the voltage, and a voltage for gain control of the AGC amplifier in the receiving path by correlating the upper bit and the lower bit.
This collision detection method is characterized by detecting the existence of both the transmitting clock of the transmitting section and the receiving clock of the receiving section.

〔Example〕

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1A and FIG. 1B are diagrams showing an example of an optical transmitting/receiving circuit and waveforms of each part thereof when performing misfire 8A. The transmission digital signal 101 is converted into a drive current signal 102 by the drive circuit 11 to drive the light emitting element 12,
This becomes a transmission optical signal 103. This signal is the star coupler 2
1 to the far end. Transmission digital signal 101
is further branched into a phase adjustment circuit 23 and a polarity inversion circuit 2.
Pass through 4. This signal undergoes coarse adjustment of the pulse amplitude in a manual amplitude adjustment circuit 25 and fine adjustment of the remaining pulse amplitude in an automatic amplitude adjustment circuit 26 to become a crosstalk cancellation signal 201.

On the other hand, a received optical signal 301 sent from the far end via the star coupler 21 is supplied to the light receiving element 31. It is assumed that a part of the transmitted optical signal 103 is further superimposed on this signal due to the branching action of the star coupler. Light receiving element 3
The signal converted into photocurrent by 1 is subjected to automatic gain control (AGC).
) is amplified by the amplifier 32, and the received electrical waveform signal 302
becomes.

The adder circuit 33 adds the crosstalk cancellation signal 20 to this signal.
1 is added, and the obtained addition output signal 303 is band-limited by the reception equalization circuit 34 and becomes an equalized waveform signal 304.

This waveform signal is supplied to a 2-bit identification circuit 35 and a timing extraction circuit 36. The 2-bit identification circuit 35 uses the received clock signal 3 generated by the timing extraction circuit 36.
Two-bit identification is performed at timing 08, and an upper bit (B1) 305 and a lower bit (Bo) 306 are generated. The upper bits 305 become the required received digital signal 307 as they are. FIG. 4 shows the relationship between the discrimination level and the amplitude of the equalized waveform signal. The correlation circuit 37 has Bl and B
The correlation between O and fi generates the AGC voltage 309, and the AGC
Controls the gain of amplifier 32. When BO and B1 have the same sign, it is determined that the amplitude is large, and when they have different signs, it is determined that the amplitude is small, and as a result, the AGC voltage 309 is obtained by total integration. FIG. 5 shows an example of a correlator circuit. Correlation circuit 27 receives transmission pulse information (D) 2
03 and BO to generate an amplitude control voltage 202 and control the automatic amplitude adjustment circuit 26. D shows the output of the phase adjustment circuit 23 read by the latch circuit 28 in the same phase as the discrimination dimming of the 2-bit discrimination circuit. D,!
: If the BOs have the same sign, it is determined that the amplitude of the crosstalk cancellation signal 201 is insufficient, and if they have different signs, it is determined that the amplitude is excessive, and the result is integrated to obtain the amplitude control voltage 202.

Finally, the collision detection circuit 4 is connected to the transmission clock 1 as shown in Figure 6.
04 and the reception clock 308 are present at the same time, the collision detection signal 401 is sent out. With this configuration, the signal component of the own terminal included in the received electrical signal 302 is
Since the signal can be automatically adjusted so that the peaks are eliminated, the presence or absence of the reception clock has no relation to the presence or absence of the burst signal sent from the transmitter of the local station.

〔Effect of the invention〕

As described above, according to the present invention, normal collision detection can be performed in optical burst transmission using all star couplers even when there are fluctuations in the transmitted optical signal level or transmission path loss.

[Brief explanation of the drawing]

Figures 1A and 1B are a block diagram 2 and waveform diagrams of various parts of an embodiment of the present invention, Figure 2 is a block diagram showing a conventional optical transmission system, and Figure 3 is a signal at each terminal station in Figure 2. A block diagram showing the collision detection method, FIG. 4 is a diagram showing the relationship between the discrimination level and the amplitude of the equalized waveform signal in FIG. 1A, and FIG. 5 is a circuit diagram showing an example of the correlation circuit used in FIG. 1A. , FIG. 6 is a waveform diagram showing the operation of FIG. 1A. 11...Drive circuit, 12...Light emitting element, 21...Star coupler, 23...
Phase adjustment circuit, 24...Polarity inversion circuit, 25.
...Manual amplitude adjustment circuit, 26...Automatic amplitude adjustment circuit, 27.37...Correlation circuit, 28.
...Latch circuit, 31... Light receiving element, 3
2...AGC amplifier, 33...Addition circuit, 34...Reception equalization circuit, 35......
2-bit identification circuit, 36... timing extraction circuit, 41... collision detection circuit. I
] 70g Sadness 30B Kaya 2 A

Claims (1)

[Claims] In a system that uses one star coupler to perform pulse burst transmission between multiple terminal stations that has an electro-optical conversion circuit in the transmitter and an opto-electric converter in the receiver, the transmitter has an optical element and its driving circuit, and the receiver has an optical element and its drive circuit. It has a receiving element, an automatic gain control (AGC) amplifier, an addition circuit, a receiving equalization circuit, a 2-bit identification circuit, and a timing extraction circuit on the side, and branches the electric pulse signal that drives the optical element on the transmitting side to determine the polarity. ,
Generate a signal with manual phase adjustment, manual amplitude adjustment, and automatic amplitude adjustment, add this signal to the pulse waveform signal from the far end in the adder circuit of the near-end receiving circuit, and waveform equalize the obtained signal. After that, the two bits are identified, and the identified upper bit is used as the required reception digital signal, and at the same time, the polarity of the lower bit is correlated with the driving electric pulse on the transmitting side, and the voltage for automatic amplitude adjustment of the branch signal is determined. Further, the correlation between the upper bit and the lower bit is determined to obtain a voltage for gain control of the AGC amplifier of the receiving circuit, and the presence of both the transmitting clock of the transmitting section and the receiving clock of the receiving section is detected. A collision detection method featuring: