Optical transceiver and method for controlling optical transceiver

Abstract

To provide an optical transceiver capable of decoding a code value of Manchester code if a synchronous clock period used in Manchester coding is unknown or even if the synchronous clock period fluctuates.SOLUTION: An optical transceiver for receiving an optical signal on which a Manchester-encoded surveillance signal is superimposed includes a decoding circuit that decodes the surveillance signal from an electric signal generated from the optical signal. The decoding circuit includes: an interval measuring unit that detects only either rising edges or falling edges of waveforms of the electric signal, measures a first time interval between a detected first edge and a second edge detected immediately after the first edge, and measures a second time interval between the second edge and a third edge detected immediately after the second edge; and a decoding unit that decodes the Manchester-encoded surveillance signal based on a ratio of the first time interval and the second time interval.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to an optical transceiver and a method for controlling an optical transceiver. [Background technology]

[0002] Manchester code is a code in which each of the binary logical values ​​to be transmitted is assigned to a rising edge and a falling edge of a signal waveform. For example, a receiving device that generates a code value and a synchronous clock from Manchester code data measures the time between either the rising edge or the falling edge of the received signal. The receiving device then determines from the measured time whether the edge at that point is a bit midpoint or bit endpoint, and generates a code value and a synchronous clock (for example, see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2005-142615 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, the receiver that decodes the Manchester code described above cannot generate a code value unless it knows the bit width, which is the period of the synchronous clock included in the Manchester code data. Therefore, if the bit width fluctuates due to temperature changes or aging of the receiver and exceeds the allowable error of the bit width, the code value of the Manchester code may not be decoded correctly.

[0005] In addition, Manchester codes are ambiguous in that it is difficult to distinguish between a pattern of consecutive identical code values ​​and a pattern of consecutive opposite code values. For this reason, if you try to receive Manchester codes from the middle of a stream, the code value decoded may be the opposite of the expected code value.

[0006] Therefore, an object of the present disclosure is to provide an optical transceiver that can decode the code value of a Manchester code even when the period of the synchronous clock used in the Manchester encoding is unknown or when the period of the synchronous clock fluctuates. [Means for solving the problem]

[0007] An optical transceiver according to the present disclosure is an optical transceiver that receives an optical signal on which a Manchester-encoded supervisory signal is superimposed, and includes a decoding circuit that decodes the supervisory signal from an electrical signal generated from the optical signal, the decoding circuit including an interval measuring unit that detects only a rising edge or a falling edge of a waveform of the electrical signal, measures a first time interval between a detected first edge and a second edge detected immediately after the first edge, and measures a second time interval between the second edge and a third edge detected immediately after the second edge, and a decoding unit that decodes the Manchester-encoded supervisory signal based on a ratio between the first time interval and the second time interval. the decoding unit includes a clock number estimation unit that estimates a clock number that is a ratio of the second time interval to a period of a synchronous clock used in the Manchester encoding of the supervisory signal based on a ratio between the first time interval and the second time interval, and a code value determination unit that decodes the Manchester encoded supervisory signal based on the estimated clock number and a logical value of the supervisory signal that was decoded immediately before; Equipped with. Effect of the Invention

[0008] According to the present disclosure, it is possible to provide an optical transceiver that can decode the code value of a Manchester code even when the period of the synchronous clock used in the Manchester encoding is unknown or the period of the synchronous clock varies. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram illustrating an example of a circuit configuration of an optical transceiver according to an embodiment. [Diagram 2] FIG. 2 is a signal waveform diagram showing an example of the relationship between the Manchester code and the edge interval. [Diagram 3] FIG. 3 is an explanatory diagram showing an example of a ratio indicating a change in the interval between two successive edges in the Manchester code. [Figure 4]FIG. 4 is an explanatory diagram showing an example of a technique for estimating an edge interval according to the ratio between two successive edges of a low-speed received data signal. [Diagram 5] FIG. 5 is an explanatory diagram showing an example of a technique for decoding a code value based on an edge interval of a low-speed received data signal and a code value decoded immediately before. [Figure 6] FIG. 6 is a circuit diagram showing an example of the single-edge interval counter of FIG. [Figure 7] FIG. 7 is a circuit diagram illustrating an example of the decoding unit of FIG. [Figure 8] FIG. 8 is a state transition diagram showing an example of the operation of the Manchester decoding circuit of FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] [Description of the embodiments of the present disclosure] First, the embodiments of the present disclosure will be listed and described.

[0011] [1] An optical transceiver according to an embodiment of the present disclosure is an optical transceiver that receives an optical signal on which a Manchester-encoded supervisory signal is superimposed, and includes a decoding circuit that decodes the supervisory signal from an electrical signal generated from the optical signal, the decoding circuit including an interval measuring unit that detects only a rising edge or a falling edge of a waveform of the electrical signal, measures a first time interval between the detected first edge and a second edge detected immediately after the first edge, and measures a second time interval between the second edge and a third edge detected immediately after the second edge, and a decoding unit that decodes the Manchester-encoded supervisory signal based on a ratio between the first time interval and the second time interval. This allows the optical transceiver to decode the Manchester-encoded supervisory signal based on a ratio between successive first time intervals and second time intervals even when the period of a synchronous clock used in the Manchester encoding is unknown or varies.

[0012] [2] In the above [1], the decoding unit may include a clock number estimation unit that estimates the number of clocks in the second time interval relative to the period of the synchronous clock used in the Manchester encoding of the supervisory signal based on the ratio between the first time interval and the second time interval, and a code value determination unit that decodes the Manchester encoded supervisory signal based on the estimated number of clocks and the logic value of the supervisory signal decoded immediately before. This makes it possible to decode the Manchester encoded supervisory signal even when the period of the synchronous clock is unknown, by measuring the first time interval and the second time interval and determining the number of clocks in the second time interval from the ratio of the time intervals. Furthermore, the bit width, which is the period of the synchronous clock, can be determined from the number of clocks in the second time interval.

[0013] [3] In the above [2], the clock count estimation unit may include a storage unit that stores the first time interval, a plurality of multipliers that multiply the first time interval by each of a plurality of thresholds, and a plurality of comparators that compare each of a plurality of multiplication results by the plurality of multipliers with the second time interval, and estimate the clock count based on a plurality of comparison results by the plurality of comparators. In this way, by performing multiplication by a multiplier instead of division to obtain a ratio between successive first time intervals and second time intervals, the clock count of the second time interval can be estimated and the supervisory signal can be decoded. As a result, the circuit scale of the decoding unit can be suppressed, and the cost of the optical transceiver can be reduced.

[0014] [4] In the above [1], the interval measurement unit may further measure a third time interval between the third edge and a fourth edge detected immediately after the third edge, and the decoding unit may decode a first logic value corresponding to the third edge of the supervisory signal based on a ratio between the first time interval and the second time interval, and decode a second logic value corresponding to the fourth edge of the supervisory signal based on a ratio between the second time interval and the third time interval. This allows the decoding unit to sequentially determine successive code values ​​of the Manchester code by sequentially using pairs of two successive time intervals.

[0015] [5] In the above [1], when the decoding unit detects that the ratio between the first time interval and the second time interval is out of a preset range, the interval measurement unit may further measure a third time interval between the third edge and a fourth edge detected immediately after the third edge, and a fourth time interval between the fourth edge and a fifth edge detected immediately after the fourth edge, and the decoding unit may decode the supervisory signal based on the ratio between the third time interval and the fourth time interval. This allows the decoding unit to restart decoding the Manchester-encoded supervisory signal without continuing to detect erroneous code values.

[0016] [6] In any one of [1] to [5] above, the Manchester-encoded supervisory signal received by the decoding circuit may include a correction pattern in which a first logic value, a first logic value, a second logic value, and a first logic value appear in sequence, and the first logic value may be indicated by a rising edge or a falling edge detected by the interval measurement unit. In this way, even if the decoding unit continues to erroneously detect consecutive codes of the first logic value as consecutive codes of the second logic value, for example, the decoding unit can return the reception operation of the code value to normal by receiving the correction pattern, and can then detect the correct code value.

[0017] [7] According to another aspect of the present disclosure, there is provided a method for controlling an optical transceiver, which receives an optical signal on which a Manchester-encoded supervisory signal is superimposed, and decodes the supervisory signal from an electrical signal generated from the optical signal, by detecting only a rising edge or a falling edge of a waveform of the electrical signal, measuring a first time interval between a first edge detected and a second edge detected immediately after the first edge, measuring a second time interval between the second edge and a third edge detected immediately after the second edge, and decoding the Manchester-encoded supervisory signal based on a ratio between the first time interval and the second time interval. This allows the optical transceiver to decode the Manchester-encoded supervisory signal based on a ratio between successive first time intervals and second time intervals even when the period of a synchronous clock used in the Manchester encoding is unknown or varies.

[0018] [Details of the embodiment of the present disclosure] A specific example of the optical transceiver of the present disclosure will be described below with reference to the drawings. In the following description, the same or corresponding elements are given the same reference numerals, and the description of those elements may be omitted. The reference numerals of the input terminals, output terminals, and each node are also used to indicate signals.

[0019] [One embodiment] [Circuit configuration of optical transceiver] Fig. 1 is a block diagram showing an example of a circuit configuration of an optical transceiver according to an embodiment. For example, the optical transceiver 100 shown in Fig. 1 is connected to a communication device such as a host device that transmits and receives optical signals. The optical transceiver 100 has a function of converting a digital signal into an optical signal and transmitting it to another optical transceiver, and a function of converting an optical signal received from another optical transceiver into a digital signal. The optical transceiver 100 has a DSP (Digital Signal Processor) 10, a Manchester encoding circuit 20, an opto-electrical conversion unit 30, a low pass filter (LPF) 40, and a Manchester decoding circuit 50.

[0020] The photoelectric conversion unit 30 has a laser diode (LD) driver 32, a laser diode (LD) 34, a photodiode (PD) 36, and a transimpedance amplifier (TIA) 38. Hereinafter, the laser diode driver 32, the laser diode 34, the photodiode 36, and the transimpedance amplifier 38 will also be referred to as the LD driver 32, the LD 34, the PD 36, and the TIA 38, respectively. The low-pass filter 40 will also be referred to as the LPF 40. The Manchester decoding circuit 50 has a single-edge interval counter 60 and a decoding unit 70. The Manchester decoding circuit 50 is an example of a decoding circuit.

[0021] The DSP 10 receives parallel high-speed transmission data signals (digital signals) including information to be transmitted to other optical transceivers 100 from a host device. The DSP 10 converts the received high-speed transmission data signals into analog transmission signals such as PAM (Pulse Amplitude Modulation) 4 signals. The DSP 10 outputs the converted analog transmission signals to the LD driver 32.

[0022] The DSP 10 also receives, from the TIA 38, analog reception signals such as PAM4 signals converted from optical signals received from other optical transceivers 100. The DSP 10 converts the analog reception signals into parallel high-speed reception data signals (digital signals). The DSP 10 outputs the converted high-speed reception data signals to the host device. The DSP 10 includes a high-pass filter (not shown). The high-pass filter removes supervisory control information superimposed on the analog reception signals received from the TIA 38.

[0023] For example, the high-speed transmission data signal and the high-speed reception data signal are NRZ (Non-Return-to-Zero) signals. Note that the high-speed transmission data signal and the high-speed reception data signal may be converted into a PAM4 signal or vice versa by a conversion circuit connected between the DSP 10 and the LD driver 32 and the TIA 38.

[0024] The Manchester encoding circuit 20 Manchester-encodes the supervisory control information for transmission (transmission supervisory control information) and outputs the encoded supervisory control information as a supervisory low-speed transmission data signal to the LD driver 32. The transfer rate of the supervisory low-speed transmission data signal is lower than the transfer rate of the data signal of the communication target transmitted and received by the DSP 10. For example, the PAM4 signal transmitted and received by the DSP 10 is 25 Gbaud, whereas the frequency of the low-speed transmission data signal output from the Manchester encoding circuit 20 is about several tens of kHz to 1 MHz.

[0025] The LD driver 32 superimposes a low-speed transmission data signal including supervisory control information on an analog transmission signal received from the DSP 10. The LD driver 32 outputs the superimposed analog transmission signal to the LD 34 to drive the LD 34. The LD 34 generates an optical transmission signal from the analog transmission signal (electrical signal) on which the low-speed transmission data signal is superimposed, and outputs the generated optical transmission signal to an optical cable.

[0026] For example, the optical signal output from the LD 34 is a PAM4 signal. Note that the superposition of the monitoring low-speed transmission data signal on the analog transmission signal may be performed by changing the bias current supplied to the LD driver 32 according to the low-speed transmission data signal.

[0027] The PD36 receives an optical reception signal on which supervisory control information is superimposed from another optical transceiver 100 via an optical cable. The PD36 generates a current signal (electrical signal) from the received optical reception signal, and outputs the generated current signal to the TIA38. For example, the optical signal received by the PD36 is a PAM4 signal. The TIA38 amplifies the minute current signal received from the PD36, and outputs the amplified current signal to the DSP10 and the LPF40 as a voltage signal (analog reception data signal).

[0028] The LPF 40 extracts the low-speed received data signal (Manchester-encoded supervisory control information) superimposed on the serial received data signal by cutting off high-frequency components of the received data signal received from the TIA 38. The LPF 40 outputs the extracted low-speed received data signal to a Manchester decoding circuit 50. The low-speed received data signal is an example of a Manchester-encoded supervisory signal.

[0029] The Manchester decoding circuit 50 decodes the Manchester-encoded low-speed received data signal using the single-edge interval counter 60 and the decoding section 70. The Manchester decoding circuit 50 outputs the decoded low-speed received data signal to the host device as reception monitoring control information.

[0030] Examples of the single-edge interval counter 60 and the decoder 70 are described in Figures 6 and 7. For example, the transmission monitoring control information and the reception monitoring control information include information for controlling on / off of the transmission of the optical signal, information for setting the transmission speed or data rate of the optical signal, information for setting the intensity of the optical signal, information indicating the power supply voltage or temperature of the optical transceiver 100, and the like.

[0031] [Relationship between Manchester code and edge spacing] 2 is a signal waveform diagram showing an example of the relationship between the Manchester code and the edge interval D. For example, the Manchester encoding circuit 20 in FIG. 1 assigns a rising edge (code value P) to the midpoint of the clock cycle when data generated in synchronization with the synchronous clock has a logical value of 1. Also, the Manchester encoding circuit 20 assigns a falling edge (code value N) to the midpoint of the clock cycle when data generated in synchronization with the synchronous clock has a logical value of 0. The code value P is an example of a first logical value, and the code value N is an example of a second logical value. Note that the Manchester code may have a falling edge as the code value P and a rising edge as the code value N.

[0032] Then, the Manchester encoding circuit 20 assigns a rising edge or a falling edge to a position corresponding to the edge (falling edge) of the synchronous clock according to the edge assigned to the midpoint of the clock cycle. In the Manchester code generated in this way, when the period of the synchronous clock, i.e., the bit width, is "T", the edge interval D between two adjacent rising edges is either "T", "1.5T", or "2T". Note that the edge interval D may be the interval between adjacent falling edges, but an example of the edge interval D between two adjacent rising edges will be described below.

[0033] In FIG. 2, the numerical value attached to the circular arc arrow connecting two consecutive edge intervals D corresponds to the number of the ratio R(n) of change of the edge interval D shown in FIG. 3 and FIG. 4. Of the two consecutive edge intervals D, the earlier edge interval D(n-1) is an example of the first time interval between the first edge and the second edge immediately after the first edge. Of the two consecutive edge intervals D, the later edge interval D(n) is an example of the second time interval between the second edge and the third edge immediately after the second edge. Furthermore, the edge interval D(n+1) next to the edge interval D(n) is an example of the third time interval between the third edge and the fourth edge immediately after the third edge. The first edge, the second edge, and the third edge are rising edges adjacent to each other in the signal waveform of the Manchester code. In the Manchester code of the signal waveform at the bottom of FIG. 2, the code values ​​P, P, N, and P shown in the dashed line frame are the same pattern as the correction pattern inserted at a predetermined interval. The correction pattern is illustrated in FIG.

[0034] [Edge spacing change and ratio] 3 is an explanatory diagram showing an example of a ratio R(n) indicating a change in the distance between two successive edges D(n-1) and D(n) in the Manchester code. The ratio R(n) is calculated by the following formula (1). R(n) = D(n) / D(n-1) (1) There are seven different ratios R(n) of change in edge spacing D, as indicated by numbers (1) to (7).

[0035] When the edge spacing D changes from "2T" to "T", the ratio R(n) of change in the edge spacing D is "0.5". When the edge spacing D changes from "1.5T" to "T", the ratio R(n) of change in the edge spacing D is "0.67". When the edge spacing D changes from "2T" to "1.5T", the ratio R(n) of change in the edge spacing D is "0.75".

[0036] When the edge spacing D does not change (from "T" to "T", "1.5T" to "1.5T" or "2T" to "2T"), the ratio R(n) of the change in the edge spacing D is "1". When the edge spacing D changes from "1.5T" to "2T", the ratio R(n) of the change in the edge spacing D is "1.33". When the edge spacing D changes from "T" to "1.5T", the ratio R(n) of the change in the edge spacing D is "1.5". When the edge spacing D changes from "T" to "2T", the ratio R(n) of the change in the edge spacing D is "2".

[0037] [Bit width estimation and code value decoding] Fig. 4 is an explanatory diagram showing an example of a technique for estimating an edge interval D according to the ratio between two successive edge intervals D of a low-speed received data signal. The numbers and ratios R(n) shown in parentheses on the left side of Fig. 4 correspond to the numbers and ratios R(n) shown in Fig. 3. The ratio R(n) is the ratio D(n) / D(n-1) of the two successive edge intervals D(n-1) and D(n). The edge interval D(n-1) is the preceding edge interval D, and the edge interval D(n) is the succeeding edge interval D.

[0038] In Fig. 4, the median values ​​of two adjacent ratios R(n) are set as the thresholds VT1, VT2, VT3, VT4, VT5, and VT6, respectively. The threshold VT0 is set to a value (0.42) obtained by subtracting the difference 0.08 between the threshold VT1=0.58 and the ratio R(n)=0.5 from the ratio R(n)=0.5. The threshold VT7 is set to a value (2.25) obtained by adding the difference 0.25 between the ratio R(n)=2 and the threshold VT6=1.75 to the ratio R(n)=2. Note that the thresholds VT0 to VT7 may be collectively referred to as the thresholds VT.

[0039] For example, the decoding unit 70 obtains a ratio R(n) from the edge interval D(n-1) and edge interval D(n) that are actually measured. The decoding unit 70 compares the obtained ratio R(n) with each threshold value VT. Then, based on the comparison result, the decoding unit 70 estimates how many times (1, 1.5, or 2) the subsequent edge interval D(n) of the two edge intervals D(n-1) and D(n) is as compared to the bit width T. Here, the bit width T is the period of the synchronous clock used in Manchester encoding. Hereinafter, the ratio of the edge interval D to the bit width T (clock period) is also referred to as the number of clocks. For example, if the edge interval D is 1.5 times the bit width T, the number of clocks for the edge interval D is "1.5".

[0040] By estimating the number of clocks of the edge interval D, it is possible to decode the Manchester-encoded code value contained in the low-speed received data signal LDT, as described in Fig. 5. Furthermore, since the number of clocks of the edge interval D can be estimated, it is possible to decode the code value even when the period of the synchronous clock used in the Manchester encoding is unknown or when the period of the synchronous clock used in the Manchester encoding has changed. In other words, the period of the synchronous clock can be estimated and updated every time the ratio R(n) is obtained.

[0041] For example, when the ratio R(n) is 0.75, the period of the synchronous clock, i.e., the bit width T, can be estimated by dividing the estimated edge interval D(n) by 1.5. When the ratio R(n) is 1.33, the period of the synchronous clock, i.e., the bit width T, can be estimated by dividing the estimated edge interval D(n) by 2. The period of the synchronous clock, i.e., the bit width T, may be estimated based on the average value obtained by dividing the sum of the edge intervals D(n-1) and D(n) by 2.

[0042] In addition, in (3) of Fig. 4, the bit width T may be estimated by multiplying the edge interval D(n) by 1 / 2, rather than by multiplying the edge interval D(n) by 1.5. In (6) and (7) of Fig. 4, the bit width T may be estimated not using the edge interval D(n), but by estimating the edge interval D(n-1) as the bit width T. In this case, the period of the synchronous clock can be calculated without calculation or by shifting the binary bit string indicating the edge interval D(n-1) to the right by one bit, thereby reducing the amount of calculation.

[0043] When the ratio R(n) is greater than "0.88" and equal to or less than "1.17", the decoding unit 70 determines that the edge intervals D(n-1) and D(n) are equal to each other. In this case, the edge intervals D(n-1) and D(n) are, for example, any of "T", "1.5T" and "2T" and are not determined. Therefore, the decoding unit 70 reserves the estimation of the edge interval D(n). Note that even when the decoding unit 70 reserves the estimation of the edge interval D, it can decode the code value included in the low-speed reception data signal LDT by using the method described in FIG. 5. For example, when the edge interval D(n-1) is determined, the edge interval of the edge interval D(n-1) may be set as the edge interval D(n).

[0044] If the ratio R(n) is equal to or smaller than the threshold value VT0, the decoding unit 70 determines that the edge interval D(n) is too small and therefore cannot be estimated. If the ratio R(n) is greater than the threshold value VT7, the decoding unit 70 determines that the edge interval D(n) is too large and therefore cannot be estimated.

[0045] When the decoding unit 70 determines that the estimation is impossible, it starts over from the beginning to obtain the ratio R(n) of the edge intervals D. In this case, the decoding unit 70 obtains the ratio R(n+2)=D(n+2) / D(n+1) of the edge interval D(n) next to the edge interval D(n) and the edge interval D(n+2) next to the edge interval D(n). The edge interval D(n+1) is an example of a third time interval between the third edge and the fourth edge detected immediately after the third edge. The edge interval D(n+2) is an example of a fourth time interval between the fourth edge and the fifth edge detected immediately after the fourth edge. The third edge, the fourth edge, and the fifth edge are rising edges adjacent to each other in the signal waveform of the Manchester code.

[0046] In this way, by setting the upper and lower limits of the ratio R(n) as the thresholds VT0 and VT7, it is possible to detect edges erroneously generated due to noise or the like, changes in the transmission rate of the Manchester code, or interruptions in the transmission of the low-speed reception data signal LDT. In this case, the decoding unit 70 makes the single-edge interval counter 60 restart measuring the initial edge interval D(n-1). This allows the decoding unit 70 to resume decoding the low-speed reception data signal LDT based on the new edge interval D(n-1) and the ratio R(n) of D(n).

[0047] The codes ERR1=1, BW1=1, BW2=1, BW3=1, BW4=1, BW5=1, BW6=1, BW7=1, and ERR2=1 shown on the right side of Fig. 4 are values ​​of signals used in the decoding unit 70 shown in Fig. 7, and only one of them is set to a logical value of 1. The codes ERR1 and ERR2 are error signals, and each of the codes BW1 to BW7 is a bit width signal. Note that each of the bit width signals BW1 to BW7 may be collectively referred to as the bit width signal BW.

[0048] For example, bit width signal BW1 of logical value 1 and bit width signal BW2 of logical value 1 indicate that the number of clocks in edge interval D(n) is "1," indicating that edge interval D(n) is approximately equal to bit width T (clock period). Bit width signal BW3 of logical value 1 and bit width signal BW6 of logical value 1 indicate that the number of clocks in edge interval D(n) is "1.5," indicating that edge interval D(n) is approximately equal to 1.5 times the bit width T (clock period).

[0049] The bit width signal BW5 of logic 1 and the bit width signal BW7 of logic 1 indicate that the number of clocks of the edge interval D(n) is "2," and indicate that the edge interval D(n) is approximately equal to twice the bit width T (clock period). Thus, the bit width signal BW indicates an estimate of how many times the edge interval D(n) is the bit width T of the synchronous clock used in Manchester encoding.

[0050] [Method of decoding code value from edge interval] Fig. 5 is an explanatory diagram showing an example of a method for decoding a code value based on an edge interval D of a low-speed reception data signal LDT and a code value decoded immediately before. Fig. 5 shows a decoding operation by the decoding unit 70 of Fig. 1. In Fig. 5, the edge interval D indicates the interval between two adjacent rising edges in the signal waveform of the low-speed reception data signal LDT. As shown in Fig. 2, the edge interval D is any one of "T", "1.5T", and "2T".

[0051] The "reference code value P" shown in a dashed rectangular frame indicates that the previously decoded code value is "P". The "reference code value N" shown in a dashed rectangular frame indicates that the previously decoded code value is "N". The dashed circular frame indicates the last code value determined by decoding. If the last code value determined is "P", it is shown as "LastP", and if the last code value determined is "N", it is shown as "LastN". Note that if there is only one code value determined, the determined code value becomes "LastP" or "LastN". "LastP" becomes the reference code value P when the next code value is determined. "LastN" becomes the reference code value N when the next code value is determined.

[0052] When the reference code value is "P" and the edge interval D is "T" (FIG. 5(a)), the decoding unit 70 determines that it has received a code value P. When the reference code value is "N" and the edge interval D is "T" (FIG. 5(b)), the decoding unit 70 determines that it has received a code value N.

[0053] When the reference code value is "P" and the edge interval D is "1.5T" (FIG. 5(c)), the decoding unit 70 determines that it has received code value N. Note that the code value N shown in parentheses following the received code value N will be decoded next time. When the reference code value is "N" and the edge interval D is "1.5T" (FIG. 5(d)), the decoding unit 70 determines that it has received code value N and code value P.

[0054] When the reference code value is "P" and the edge interval D is "2T" (FIG. 5(e)), the decoding unit 70 determines that it has received code values ​​N and P. When the reference code value is "N" and the edge interval D is "2T" (FIG. 5(f)), the decoding unit 70 determines that a phase shift has occurred because this is a pattern that does not exist in the Manchester code, and does not determine the received code value. However, the decoding unit 70 can determine the latter rising edge as "LastP", making it possible to determine the code value at the next edge interval D that appears.

[0055] As shown by the waveform in FIG. 2, the decoding unit 70 uses two successive edge intervals D to decode the Manchester code. For example, the decoding unit 70 determines one or two code values ​​based on the ratio R(n) between the edge interval D(n-1) and the edge interval D(n), and determines one or two code values ​​based on the ratio R(n+1) between the edge interval D(n) and the edge interval D(n+1). In this way, of the edge intervals D(n-1) and D(n) used to determine the code value, the latter edge interval D(n) is used to determine the next code value. As a result, the decoding unit 70 can sequentially determine the successive code values ​​of the Manchester code by sequentially using pairs of two successive edge intervals D.

[0056] [Circuit configuration of single edge interval counter 60] Fig. 6 is a circuit diagram showing an example of the single-edge interval counter 60 of Fig. 1. The single-edge interval counter 60 has a counter 62 and a register 64. The single-edge interval counter 60 is an example of an interval measurement unit that repeatedly measures the edge interval D between adjacent rising edges of the low-speed reception data signal LDT. Note that the single-edge interval counter 60 detects rising edges, but may also detect falling edges. However, the single-edge interval counter 60 detects only either a predetermined rising edge or a predetermined falling edge.

[0057] The counter 62 has a reset terminal RST, a clock terminal CK for receiving the clock signal CLK, and an output terminal OUT for outputting a count value CNT. The reset terminal RST receives the Manchester-encoded low-speed reception data signal LDT output from the LPF 40 in Fig. 1. The counter 62 resets the count value CNT to "0" in synchronization with the rising edge of the low-speed reception data signal LDT, and increments the count value CNT by "1" in synchronization with the clock signal CLK.

[0058] As a result, the counter 62 outputs the time interval from the previous rising edge as the count value CNT for each rising edge of the low-speed reception data signal LDT. That is, the counter 62 repeatedly measures the edge interval D between two adjacent rising edges as the count value CNT.

[0059] The register 64 has a load terminal for receiving the low-speed reception data signal LDT, an input terminal IN for receiving the count value CNT, and an output terminal OUT for outputting the loaded count value CNT. The register 64 latches the count value CNT in synchronization with the rising edge of the low-speed reception data signal LDT, and outputs the latched count value CNT from the output terminal OUT with an edge interval D.

[0060] The signal waveforms shown in parentheses at the bottom of Fig. 6 indicate the operation of the single-edge interval counter 60. The vertical stripes of the clock signal CLK and the count value CNT in the waveform diagram indicate that the frequency of the clock signal CLK is sufficiently higher than the frequency of the low-speed received data signal LDT. For example, the clock signal CLK may be a divided clock signal obtained by dividing the frequency of the clock signal used in the DSP 10 by a predetermined division ratio, or another clock signal may be used.

[0061] As shown in the signal waveform, two rising edges of the low-speed reception data signal LDT are required to obtain one edge interval D. Furthermore, two consecutive edge intervals D (for example, D(n-1) and D(n)) are required to decode the reception monitor control information in the decoder 70.

[0062] The single-edge interval counter 60 may repeatedly measure the edge interval D between adjacent falling edges of the low-speed reception data signal LDT. In this case, the counter 62 is reset at the falling edge of the low-speed reception data signal LDT. The register 64 latches the count value CNT in synchronization with the falling edge of the low-speed reception data signal LDT, and outputs the latched count value CNT from the output terminal OUT as the edge interval D. However, the configuration and operation of the decoding unit 70 when measuring the edge interval D between falling edges differ from the configuration and operation of the decoding unit 70 described below.

[0063] [Circuit configuration of the decoding unit 70] Fig. 7 is a circuit diagram showing an example of the decoding unit 70 of Fig. 1. The decoding unit 70 has a clock count estimation unit 80 and a code value determination unit 90. The clock count estimation unit 80 has a previous value memory 82, eight multipliers 84, eight comparators 86, eight inverters IV, and seven AND circuits AND. The numbers of the multipliers 84, comparators 86, inverters IV, and AND circuits AND are set according to the number (types) of ratios R(n) shown in Fig. 3.

[0064] The previous value memory 82 receives the edge interval D(n) from the single-edge interval counter 60 in synchronization with the rising edge of the low-speed received data signal LDT received at the clock terminal CK, and outputs it as the edge interval D(n-1). The edge interval D(n-1) output by the previous value memory 82 is supplied to each multiplier 84.

[0065] The edge interval D(n) input to the previous value memory 82 is the edge interval D between the two slower rising edges of the three rising edges of the low-speed reception data signal LDT, as shown in the signal waveform of Fig. 6. In other words, the edge interval D(n) is the subsequent edge interval D of the two edge intervals D(n-1), D(n).

[0066] The edge interval D(n-1) output from the previous value memory 82 is the edge interval D between the two rising edges on the earlier side among the three rising edges of the low-speed reception data signal LDT, as shown in the signal waveform of Fig. 6. In other words, the edge interval D(n-1) is the preceding edge interval D between two successive edge intervals D(n-1), D(n). The previous value memory 82 is an example of a storage unit that stores the preceding edge interval D(n-1) between two successive edge intervals D(n-1), D(n) of the low-speed reception data signal LDT.

[0067] Each multiplier 84 multiplies the edge interval D(n-1) by a different multiplier and outputs the multiplication result to the comparator 86. The multipliers used in the eight multipliers 84 are "0.42", "0.58", "0.71", "0.86", "1.17", "1.42", "1.75", and "2.25", respectively, which are the thresholds VT0, VT1, ..., VT7 described in FIG.

[0068] Each comparator 86 compares the multiplication result output by the corresponding multiplier 84 with the edge interval D(n) and outputs the comparison result. Each comparator 86 outputs a logical value of 1 if the edge interval D(n) is greater than the multiplication result, and outputs a logical value of 0 if the edge interval D(n) is equal to or less than the multiplication result.

[0069] The circuit shown in FIG. 7 is configured using equation (2), which is a modification of equation (1) explained in FIG. D(n) = R(n) D(n-1) (2) Each multiplier 84 calculates the right side of equation (2), and comparator 86 compares the left and right sides of equation (2). That is, comparator 86 compares edge interval D(n) with threshold value R(n)·D(n−1).

[0070] This makes it possible to compare the ratio R(n) with the threshold value VT by multiplying the edge interval D(n-1) with each threshold value VT without performing the division shown in equation (1). As a result, it is possible to reduce the circuit scale of the decoding unit 70 and the cost of the optical transceiver 100. Note that the circuit configuration of the clock count estimation unit 80 is not limited to that shown in FIG.

[0071] Furthermore, the threshold value R(n)·D(n-1) varies depending on the edge interval D(n-1). That is, the clock count estimator 80 can update the threshold value each time according to the previously measured edge interval D(n-1) and compare it with the edge interval D(n). Therefore, even if the frequency of the synchronous clock used in Manchester encoding changes or synchronous clock jitter or the like occurs, the threshold value can be updated to track the frequency change, jitter, or the like. As a result, the decoding accuracy of the Manchester decoding circuit 50 can be improved compared to when the threshold value is fixed, and the reliability of the optical transceiver 100 can be improved.

[0072] Each AND circuit AND receives a logical value output by a corresponding comparator 86 and a logical value obtained by inverting, by an inverter IV, a logical value output by a comparator 86 that compares a multiplication result with a magnification factor one higher than the multiplication result compared by the corresponding comparator 86. Each AND circuit AND outputs the logical operation result to the code value determination unit 90 as a bit width signal BW (BW1, BW2, ..., BW7).

[0073] 7 is inverted by an inverter IV and output as an error signal ERR1 to a code value determination unit 90. The logical value output by the comparator 86 on the right side of FIG 7 is output as an error signal ERR2 to the code value determination unit 90.

[0074] For example, when both edge intervals D(n-1) and D(n) are "1", the comparators 86 corresponding to the four multipliers 84 with 0.42 times to 0.88 times output a logical value of 1, and the comparators 86 corresponding to the four multipliers 84 with 1.17 times to 2.25 times output a logical value of 0. In this case, the clock count estimation unit 80 outputs a bit width signal BW4 with a logical value of 1, and bit width signals BW1-BW3, BW5-BW7, and error signals ERR1 and ERR2 with a logical value of 0. That is, the bit width signal BW4 with a logical value of 1 indicates that the ratio R(n) is greater than 0.88 and equal to or less than 1.17.

[0075] Furthermore, when the edge interval D(n-1) is "1" and the edge interval D(n) is "2", the comparators 86 corresponding to the seven multipliers 84 with 0.42 times to 1.75 times output a logical value of 1, and the comparator 86 corresponding to the 2.25 times multiplier 84 outputs a logical value of 0. In this case, the clock number estimation unit 80 outputs a bit width signal BW7 with a logical value of 1, and bit width signals BW1-BW6 and error signals ERR1 and ERR2 with a logical value of 0. That is, the bit width signal BW7 with a logical value of 1 indicates that the ratio R(n) is greater than 1.75 and equal to or less than 2.25.

[0076] When the edge interval D(n-1) is "1" and the edge interval D(n) is "3", all comparators 86 output a logical value of 1. In this case, the clock count estimation unit 80 outputs an error signal ERR1 with a logical value of 1, and bit width signals BW1-BW7 and an error signal ERR2 with a logical value of 0. In other words, the error signal ERR1 with a logical value of 1 indicates that the ratio R(n) is equal to or less than 0.42.

[0077] When the edge interval D(n-1) is "3" and the edge interval D(n) is "1", all comparators 86 output a logical value of 0. In this case, the clock count estimation unit 80 outputs an error signal ERR2 of a logical value of 1, and an error signal ERR1 and bit width signals BW1-BW7 of a logical value of 0. In other words, the error signal ERR2 of a logical value of 1 indicates that the ratio R(n) is greater than 2.25.

[0078] The code value determination unit 90 has a code value extraction unit 92 and a last value holding unit 94. The code value extraction unit 92 extracts a code value included in the received low-speed reception data signal LDT based on the logical value of the bit width signal BW1-BW7 output by the clock count estimation unit 80 and the code value (P or N) held by the last value holding unit 94. The code value held by the last value holding unit 94 is the code value extracted by the code value extraction unit 92 immediately before.

[0079] The extraction of the code value by the code value extraction unit 92 is performed according to the method of decoding the code value shown in Fig. 5. The most recently decoded code value, shown in a dashed rectangular frame in Fig. 5, is held in the Last value holding unit 94 and is used to extract the code value in the code value extraction unit 92. LastP or LastN, shown in a dashed circular frame in Fig. 5, is the last code value extracted by the code value extraction unit 92 and is held in the Last value holding unit 94. Then, the decoding unit 70 outputs the code value extracted by the code value extraction unit 92 to the host device as reception monitoring control information.

[0080] 7, the Manchester-encoded code value can be decoded by comparing the ratio R(n) of two consecutive edge intervals D(n-1), D(n) in the low-speed received data signal LDT with a threshold value VT. In other words, the clock count estimation unit 80 can find the number of clocks for edge interval D(n) by comparing the edge interval D(n) with a threshold value R(n)·D(n-1), which is updated for each previously measured edge interval D(n-1).

[0081] Generally, when Manchester encoding information, a synchronizing bit string such as a preamble is inserted at an arbitrary timing. By inserting a synchronizing bit string, it is possible to prevent a large number of code values ​​P or N from appearing consecutively.

[0082] Therefore, by continuing to receive the low-speed reception data signal LDT, the Manchester decoding circuit 50 can sequentially estimate the number of clocks of the edge interval D, and sequentially decode the Manchester code. In other words, after successfully decoding the low-speed reception data signal LDT once, the low-speed reception data signal LDT can be continued to be decoded using the method shown in FIG.

[0083] [Manchester Decoding Circuit State Transition Diagram] Fig. 8 is a state transition diagram showing an example of the operation of the Manchester decoding circuit 50 of Fig. 1. The Manchester decoding circuit 50 has a start state, a measurement state, a LastP state, and a LastN state. In the LastP state, the decoding unit 70 shown in Fig. 7 judges the code value according to the judgment method for the reference code value P shown in Fig. 5. In the LastN state, the decoding unit 70 judges the code value according to the judgment method for the reference code value N shown in Fig. 5.

[0084] First, in a start state, the single-edge interval counter 60 starts receiving the Manchester-encoded low-speed reception data signal LDT. When the single-edge interval counter 60 detects the first rising edge, it transitions to a state of measuring the edge interval D(n-1) (FIG. 8(a)). If the single-edge interval counter 60 fails to measure the first edge interval D(n-1), it resets the count value CNT to remeasure the edge interval D(n-1) (FIG. 8(b)).

[0085] When the single-edge interval counter 60 detects the next rising edge of the low-speed reception data signal LDT and succeeds in measuring the first edge interval D(n-1), it determines the edge interval D(n-1) and transitions to the LastP state (FIG. 8(c)). The Last value holding unit 94 holds the code value P as the Last value.

[0086] As shown in Fig. 2, when the code value is "P", the rising edge of the signal is located at the midpoint of the bit width T (the falling edge of the synchronous clock). When the code value is "N", the falling edge of the signal is located at the midpoint of the bit width T (the falling edge of the synchronous clock). Also, when the rising edge of the signal is within the synchronous clock, the rising edge of the signal is located at the end point of the bit width T (the rising edge of the synchronous clock).

[0087] Therefore, when the edge interval D(n-1) is measured for the first time after transition to the measurement state, either the LastP state or the LastN state can be taken. In this embodiment, when the edge interval D(n-1) is measured for the first time after transition to the measurement state, the state is transitioned to the LastP state as a provisional state. Note that when the edge interval D(n-1) is measured for the first time after transition to the measurement state, the Manchester decoding circuit 50 may be transitioned to the LastN state.

[0088] In the LastP state or the LastN state, when the single-edge interval counter 60 detects the next rising edge of the low-speed reception data signal LDT, it stores the edge interval D(n-1) in the previous value memory 82. Then, as described in FIG. 7, the clock count estimation unit 80 sets either the bit width signal BW1-BW7 or the error signals ERR1, ERR2 to a logical value of 1 based on the ratio R(n) (=edge interval D(n) / D(n-1)). In other words, reception of the Manchester code is started.

[0089] In the LastP state, when the code value extraction unit 92 receives a bit width signal BW1 having a logical value of 1 or a bit width signal BW2 having a logical value of 1 indicating the edge interval D(n)="T", it determines that it has received the code value P and remains in the LastP state (FIG. 8(d)). The Last value holding unit 94 holds the code value P determined by the code value extraction unit 92 as the Last value.

[0090] In the LastP state, when the code value extraction unit 92 receives a bit width signal BW3 having a logical value of 1 or a bit width signal BW6 having a logical value of 1 indicating the edge interval D(n)="1.5T", it determines that it has received the code value N and transitions to the LastN state (FIG. 8(e)). The Last value holding unit 94 holds the code value N determined by the code value extraction unit 92 as the Last value.

[0091] In the LastP state, when the code value extraction unit 92 receives a bit width signal BW5 having a logical value of 1 indicating the edge interval D(n)="2T" or a bit width signal BW7 having a logical value of 1, it determines that it has received code values ​​N and P, and remains in the LastP state (FIG. 8(f)). The Last value holding unit 94 holds the code value P last determined by the code value extraction unit 92 as the Last value.

[0092] In the LastP state, the code value determining section 90 stops the operation of repeatedly measuring the edge interval D(n) when it receives the error signal ERR1 of logical value 1 or the error signal ERR2 of logical value 1. Then, the code value determining section 90 transitions to the measurement state to newly start measurement from the first edge interval D(n-1) (FIG. 8(g)).

[0093] An error signal ERR1 with a logical value of 1 is output when the edge interval D(n) is too short. An error signal ERR2 with a logical value of 1 is output when the edge interval D(n) is too long. This allows the Manchester decoding circuit 50 to restart decoding the Manchester code without continuing to detect erroneous code values.

[0094] On the other hand, in the LastN state, if the code value extraction unit 92 receives a bit width signal BW1 with a logical value of 1 or a bit width signal BW2 with a logical value of 1 indicating the edge interval D(n)="T", it determines that it has received the code value N and remains in the LastN state (FIG. 8(h)). The Last value holding unit 94 holds the code value N determined by the code value extraction unit 92 as the Last value.

[0095] In the LastN state, when the code value extraction unit 92 receives a bit width signal BW3 having a logical value of 1 or a bit width signal BW6 having a logical value of 1 indicating the edge interval D(n)="1.5T", it determines that it has received the code values ​​N and P, and transitions to the LastP state (FIG. 8(i)). The Last value holding unit 94 holds the code value P last determined by the code value extraction unit 92 as the Last value.

[0096] In the LastN state, when the code value extraction unit 92 receives a bit width signal BW5 having a logical value of 1 indicating the edge interval D(n)="2T" or a bit width signal BW7 having a logical value of 1, it determines that the measurement of the edge interval D(n) is incorrect. Then, the code value extraction unit 92 transitions to the LastP state without determining whether a code value has been received (FIG. 8(j)).

[0097] In the LastN state, when the code value determination unit 90 receives the error signal ERR1 with a logical value of 1 or the error signal ERR2 with a logical value of 1, it stops the operation of repeatedly measuring the edge interval D(n). Then, the code value determination unit 90 transitions to the measurement state to start measurement again from the first edge interval D(n-1) (FIG. 8(k)). This allows the Manchester decoding circuit 50 to restart decoding of the Manchester code without continuing to detect erroneous code values.

[0098] In the LastP state or the LastN state, the code value extraction unit 92 of the code value determination unit 90 suspends the estimation of the edge interval D(n) when the bit width signal BW4 has a logical value of 1. However, even when the decoding unit 70 suspends the estimation of the edge interval D(n), it can decode the Manchester code using the method shown in Fig. 5. For example, when the edge interval D(n-1) is determined, the edge interval of the edge interval D(n-1) may be set as the edge interval D(n).

[0099] 2 (codes P, P, N, P) into the bit string of the Manchester code, the decoding unit 70 detects edge intervals D(n-1), D(n) with a ratio R(n) of "2". The decoding unit 70 then outputs a bit width signal BW7 with a logical value of 1 to determine reception of the code value N. Therefore, even if the decoding unit 70 continuously receives bit strings with all the bits except for the correction pattern having the code value P and continues to reserve estimation based on the same edge intervals D(n-1), D(n), it can estimate the edge interval D(n) by receiving the correction pattern.

[0100] Furthermore, even if the decoding unit 70 continues to erroneously detect consecutive code values ​​P as consecutive code values ​​N and repeats the LastN state, it transitions to the LastP state upon receiving the correction pattern. This allows the code value reception operation to return to normal, and the correct code value can then be detected.

[0101] As described above, in the first embodiment, the optical transceiver 100 can decode the Manchester-encoded supervisory signal even when the period of the synchronous clock used in the Manchester encoding is unknown or varies. Specifically, the optical transceiver 100 can decode the Manchester-encoded supervisory signal based on the ratio between two successive edge intervals D(n-1), D(n).

[0102] In other words, the optical transceiver 100 can decode Manchester-encoded monitoring information regardless of the period of the synchronous clock used in the Manchester encoding. Therefore, a common optical transceiver 100 can be used to decode Manchester codes with various synchronous clock periods used in various host devices.

[0103] This reduces the development cost, manufacturing cost, and maintenance cost of the optical transceiver 100 compared to the case where an optical transceiver 100 is prepared for each system. For example, the monitoring information is communicated by being redundantly added to the signal that is originally to be communicated. For this reason, the period of the synchronous clock used in Manchester encoding may be changed depending on the installation environment and operation environment of the system to avoid affecting the communication quality of the PAM4 signal, which is the original communication target. Even in this case, the Manchester-encoded monitoring and control information can be decoded by automatically following the change in the period of the synchronous clock.

[0104] Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above embodiments. Various changes, modifications, substitutions, additions, deletions, and combinations are possible within the scope of the claims. These naturally fall within the technical scope of the present disclosure. [Explanation of symbols]

[0105] 10 DSP 20 Manchester Encoding Circuit 30 Photoelectric conversion unit 32 Laser Diode (LD) Driver 34 Laser Diode (LD) 36 Photodiode (PD) 38 Transimpedance Amplifier (TIA) 40 Low Pass Filter (LPF) 50 Manchester Decoding Circuit 60 Single Edge Interval Counter 62 Counter 64 Registers 70 Decoding section 80 Clock Number Estimation Unit 82 Previous value memory 84 Multiplier 86 Comparator 90 Code value determination section 92 Code value extraction unit 94 Last value storage section 100 Optical Transceiver BW(BW1-BW7) Bit width signal CNT Count value D, D(n), D(n-1) Edge Spacing ERR1, ERR2 Error signals IV Inverter LDT Low speed receive data signal R(n) ratio T bit width VT(VT0-VT7) Threshold

Claims

1. 1. An optical transceiver for receiving an optical signal on which a Manchester encoded supervisory signal is superimposed, comprising: a decoding circuit for decoding the supervisory signal from an electrical signal generated from the optical signal; the decoding circuit includes an interval measurement unit that detects only a rising edge or a falling edge of a waveform of the electrical signal, measures a first time interval between a detected first edge and a second edge detected immediately after the first edge, and measures a second time interval between the second edge and a third edge detected immediately after the second edge; a decoder for decoding the Manchester-encoded supervisory signal based on a ratio of the first time interval and the second time interval; Equipped with The decoding unit is a clock number estimation unit that estimates a clock number that is a ratio of the second time interval to a period of a synchronous clock used in the Manchester encoding of the supervisory signal, based on a ratio between the first time interval and the second time interval; a code value determination unit that decodes the Manchester-encoded supervisory signal based on the estimated clock number and a logic value of the supervisory signal that was decoded immediately before; Equipped with Optical transceiver.

2. 1. An optical transceiver for receiving an optical signal on which a Manchester encoded supervisory signal is superimposed, comprising: a decoding circuit for decoding the supervisory signal from an electrical signal generated from the optical signal; the decoding circuit includes an interval measurement unit that detects only a rising edge or a falling edge of a waveform of the electrical signal, measures a first time interval between a detected first edge and a second edge detected immediately after the first edge, and measures a second time interval between the second edge and a third edge detected immediately after the second edge; a decoder for decoding the Manchester-encoded supervisory signal based on a ratio of the first time interval and the second time interval; Equipped with the interval measurement unit further measures a third time interval between the third edge and a fourth edge detected immediately after the third edge; the decoder decodes a first logic value corresponding to the third edge of the supervisory signal based on a ratio between the first time interval and the second time interval, and decodes a second logic value corresponding to the fourth edge of the supervisory signal based on a ratio between the second time interval and the third time interval. Optical transceiver.

3. 1. An optical transceiver for receiving an optical signal on which a Manchester encoded supervisory signal is superimposed, comprising: a decoding circuit for decoding the supervisory signal from an electrical signal generated from the optical signal; the decoding circuit includes an interval measurement unit that detects only a rising edge or a falling edge of a waveform of the electrical signal, measures a first time interval between a detected first edge and a second edge detected immediately after the first edge, and measures a second time interval between the second edge and a third edge detected immediately after the second edge; a decoder for decoding the Manchester-encoded supervisory signal based on a ratio of the first time interval and the second time interval; Equipped with When the decoding unit detects that the ratio between the first time interval and the second time interval is out of a predetermined range, the interval measurement unit further measures a third time interval between the third edge and a fourth edge detected immediately after the third edge, and a fourth time interval between the fourth edge and a fifth edge detected immediately after the fourth edge, the decoder decodes the monitoring signal based on a ratio of the third time interval and the fourth time interval. Optical transceiver.

4. The decoding unit is a clock number estimation unit that estimates a clock number that is a ratio of the second time interval to a period of a synchronous clock used in the Manchester encoding of the supervisory signal, based on a ratio between the first time interval and the second time interval; a code value determination unit that decodes the Manchester-encoded supervisory signal based on the estimated clock number and a logic value of the supervisory signal that was decoded immediately before; 4. The optical transceiver according to claim 2, further comprising:

5. The clock count estimation unit A storage unit that stores the first time interval; a plurality of multipliers for multiplying the first time interval by a plurality of threshold values, respectively; a plurality of comparators for comparing each of a plurality of multiplication results by the plurality of multipliers with the second time interval; Equipped with estimating the number of clocks based on a plurality of comparison results by the plurality of comparators; 5. The optical transceiver according to claim 1 or 4.

6. When the decoding unit detects that the ratio between the first time interval and the second time interval is out of a predetermined range, the interval measurement unit further measures a third time interval between the third edge and a fourth edge detected immediately after the third edge, and a fourth time interval between the fourth edge and a fifth edge detected immediately after the fourth edge, the decoder decodes the monitoring signal based on a ratio of the third time interval and the fourth time interval.

3. The optical transceiver according to claim 2.

7. the Manchester encoded supervisory signal received by the decoding circuit includes a correction pattern in which a first logic value, a first logic value, a second logic value, and a first logic value appear in sequence; the first logic value is represented by a rising edge or a falling edge detected by the interval measurement unit; The optical transceiver according to any one of claims 1 to 6.

8. 1. A method for controlling an optical transceiver that receives an optical signal on which a Manchester-encoded supervisory signal is superimposed, and decodes the supervisory signal from an electrical signal generated from the optical signal, comprising the steps of: Detecting only a rising edge or a falling edge of the waveform of the electrical signal; measuring a first time interval between a detected first edge and a detected second edge immediately following the first edge; measuring a second time interval between the second edge and a third edge detected immediately after the second edge; decoding the Manchester encoded supervisory signal based on a ratio of the first time interval and the second time interval; estimating a number of clocks that is a ratio of the second time interval to a period of a synchronous clock used in the Manchester encoding of the supervisory signal based on a ratio between the first time interval and the second time interval; The Manchester-encoded supervisory signal is decoded based on the estimated clock number and the logic value of the supervisory signal decoded immediately before. A method for controlling an optical transceiver.

9. 1. A method for controlling an optical transceiver that receives an optical signal on which a Manchester-encoded supervisory signal is superimposed, and decodes the supervisory signal from an electrical signal generated from the optical signal, comprising the steps of: Detecting only a rising edge or a falling edge of the waveform of the electrical signal; measuring a first time interval between a detected first edge and a detected second edge immediately following the first edge; measuring a second time interval between the second edge and a third edge detected immediately after the second edge; decoding the Manchester encoded supervisory signal based on a ratio of the first time interval and the second time interval; further measuring a third time interval between the third edge and a fourth edge detected immediately after the third edge; decoding a first logic value corresponding to the third edge of the supervisory signal based on a ratio of the first time interval and the second time interval, and decoding a second logic value corresponding to the fourth edge of the supervisory signal based on a ratio of the second time interval and the third time interval; A method for controlling an optical transceiver.

10. 1. A method for controlling an optical transceiver that receives an optical signal on which a Manchester-encoded supervisory signal is superimposed, and decodes the supervisory signal from an electrical signal generated from the optical signal, comprising the steps of: Detecting only a rising edge or a falling edge of the waveform of the electrical signal; measuring a first time interval between a detected first edge and a detected second edge immediately following the first edge; measuring a second time interval between the second edge and a third edge detected immediately after the second edge; decoding the Manchester encoded supervisory signal based on a ratio of the first time interval and the second time interval; When it is detected that the ratio between the first time interval and the second time interval is out of a predetermined range, further measuring a third time interval between the third edge and a fourth edge detected immediately after the third edge, and a fourth time interval between the fourth edge and a fifth edge detected immediately after the fourth edge; decoding the supervisory signal based on a ratio of the third time interval and the fourth time interval; A method for controlling an optical transceiver.

Images