Optical Transmitter and Optical Transceiver

Abstract

To realize a high speed of an optical transmitter transmitting a multilevel optical signal.SOLUTION: An optical transmitter comprises: a signal generation circuit; a Mach-zehnder interferometer; and an M*N movement phase segments, and transmits a modulation optical signal for transmitting an M bit by each symbol. The signal generation circuit generates M*N binary electric signals each having the same bit rate from transmission data when the optical transmitter multiples N optical signals by a time division multiplexing. The M*N movement phase segments are provided along an optical path of the Mach-zehnder interferometer, and makes an optical phase transmitting each optical path in accordance with the M*N binary electric signals shift. The M*N movement phase segments are constructed by N electrode groups. Each electrode group contains M or more electrodes to which the M binary electric signals corresponded to them from the M*N binary electric signals are applied.SELECTED DRAWING: Figure 10

Description

Technical Field

[0001] The present invention relates to an optical transmitter that transmits multi-valued optical signals. and an optical transceiver It is related to.

Background Art

[0002] An optical modulator is one of the key devices for realizing long-distance / high-capacity optical transmission. The optical modulator generates a modulated optical signal, for example, by modulating continuous light with an electrical signal corresponding to transmission data generated by a digital signal processor (DSP: Digital Signal Processor). An example of an optical transmitter including an optical modulator is shown in FIG. 1.

[0003] In the configuration shown in FIG. 1(a), transmission data (digital signal) generated by the DSP is converted into an analog signal by a digital / analog converter (DAC). Then, the output signal of the DAC is amplified by an analog driver (linear driver) and applied to the optical modulator. The optical modulator includes an optical waveguide that forms a Mach-Zehnder interferometer, and electrodes are formed in the vicinity of the optical waveguide. Continuous light is input to the Mach-Zehnder interferometer. When the output signal of the driver is applied to the electrodes, the phase of the light propagating through the waveguide changes according to the signal, and a modulated optical signal representing the transmission data is output. In the following description, the electrode to which an electrical signal representing transmission data is applied (i.e., the electrode used as a phase shifter) may be referred to as a "phase shift segment" or simply a "segment".

[0004] In this configuration, when generating an optical signal in which each symbol transmits 2-bit data, the DSP outputs 2-bit parallel data. Then, since a 4-level analog signal is output from the DAC, a PAM4 (4-level Pulse Amplitude Modulation) optical signal is generated. However, in order to obtain a sufficient optical amplitude in this configuration, as the baud rate increases, a larger amplitude analog signal is required, so the power consumption of the driver increases.

[0005] This problem is alleviated, for example, by the configuration shown in FIG. 1(b). In the configuration shown in FIG. 1(b), the optical modulator includes electrodes for each of the plurality of bits transmitted by each symbol. That is, when each symbol transmits 2-bit data, the optical modulator includes an electrode for the lower bit (LSB segment) and an electrode for the upper bit (MSB segment). Here, it is assumed that signals with the same voltage amplitude are input to each segment. In this case, the length of the MSB segment is twice that of the LSB segment. Then, when transmission bits corresponding to each segment are given, a PAM4 optical signal is generated. According to this configuration, compared with the configuration shown in FIG. 1(a), it is not necessary to increase the amplitude of the electrical signal applied to the optical modulator, and a binary driver in which current flows only during data transitions can be used, so power consumption is reduced. Note that the modulation method shown in FIG. 1(b) may be called an "optical DAC" because a digital signal is applied to a Mach-Zehnder interferometer and an analog signal is generated in the optical domain.

[0006] Note that an optical communication device or an optical transmission circuit including an optical modulator is described, for example, in Patent Documents 1 to 4.

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0008] As described above, a configuration for generating a multi-valued optical signal using a DSP and an optical DAC is known. However, there are limits to the high-speed operation of the DSP and the electrical circuit. For this reason, the operating speed of the DSP or the electrical circuit may become a bottleneck in the high-speed operation of the transmission data.

[0009] An object according to one aspect of the present invention is to achieve high-speed operation of an optical transmitter that transmits a multi-valued optical signal and an optical transceiver

Means for Solving the Problem

[0010] An optical transmitter according to one aspect of the present invention is An optical transmitter that transmits a modulated optical signal in which each symbol transmits M (M is an integer of 2 or more) bits. When the optical transmitter multiplexes N (N is an integer of 2 or more) optical signals by time-division multiplexing, a signal generation circuit that generates M×N binary electrical signals having the same bit rate from transmission data, a Mach-Zehnder interferometer, and M×N phase shifter segments provided along the optical path of the Mach-Zehnder interferometer and shifting the phase of light propagating through the optical path according to the M×N binary electrical signals, respectively. The M×N phase shifter segments are composed of N electrode groups, the N electrode groups are provided in series along the optical path, and the distance between an electrode included in a first electrode group among the N electrode groups and a corresponding electrode included in a second electrode group adjacent to the first electrode group is the distance that light propagates through the optical path in a time corresponding to one time slot of the time-division multiplexing, and each electrode group includes M or more electrodes to which corresponding M binary electrical signals among the M×N binary electrical signals are applied. Also, an optical transmitter according to one aspect of the present invention is an optical transmitter that transmits a modulated optical signal in which each symbol transmits M (M is an integer of 2 or more) bits. When the optical transmitter multiplexes N (N is an integer of 2 or more) optical signals by time-division multiplexing, a signal generation circuit that generates M×N binary electrical signals having the same bit rate from each other from transmission data, a Mach-Zehnder interferometer, and M×N phase shift segments provided along the optical path of the Mach-Zehnder interferometer and shifting the phase of light propagating through the optical path according to the M×N binary electrical signals, respectively. The M×N phase shift segments are composed of N electrode groups, the N electrode groups are provided in series along the optical path, and the distance between an electrode included in a first electrode group among the N electrode groups and a corresponding electrode included in a second electrode group adjacent to the first electrode group is a distance that light propagates through the optical path during a period M / B when the bit rate of the transmission data is B. Each electrode group includes M or more electrodes to which corresponding M binary electrical signals among the M×N binary electrical signals are applied.

Effect of the Invention

[0011] According to the above aspect, high-speed operation of an optical transmitter that transmits a multi-valued optical signal and an optical transceiver is achieved.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

[0013] FIG. 2 shows an example of an optical transmitter using an optical DAC. In this example, the optical transmitter 1 includes a digital signal processor (DSP) 2 and an optical modulator 3. Then, the optical transmitter 1 generates a multi-valued optical signal in which each symbol transmits M bits. For example, when M = 2, a PAM4 (4-level Pulse Amplitude Modulation) optical signal is generated, and when M = 3, a PAM8 optical signal is generated. In the following description, an optical transmitter using an optical DAC may be referred to as an "optical DAC transmitter".

[0014] DSP2 generates M data sequences from the transmission data. For example, when M = 3, as shown in FIG. 3, three data sequences bit0 to bit2 are generated. In this example, the bit sequence representing the transmission data is sequentially distributed to the data sequences bit0 to bit2 one bit at a time. However, the method of distributing the transmission data to each data sequence is not limited to the example shown in FIG. 3.

[0015] The optical modulator 3 includes a Mach-Zehnder interferometer. The Mach-Zehnder interferometer includes an input optical waveguide, a P-arm optical waveguide, an N-arm optical waveguide, and an output optical waveguide. The input ends of the P-arm optical waveguide and the N-arm optical waveguide are coupled to the input optical waveguide. Therefore, the input light to the optical modulator 3 propagates through the P-arm optical waveguide and the N-arm optical waveguide. The output ends of the P-arm optical waveguide and the N-arm optical waveguide are coupled to the output optical waveguide. Therefore, the output light of the P-arm optical waveguide and the output light of the N-arm optical waveguide are combined and output. In the following description, the P-arm optical waveguide and the N-arm optical waveguide may be referred to as "arm waveguides".

[0016] Each arm is provided with an electrode used as a phase shifter. Specifically, an electrode is provided for each data sequence. For example, in the case where M = 3, electrodes S0 to S2 are provided on each arm. And electrical signals representing the corresponding data sequences are applied to the electrodes S0 to S2 respectively. Specifically, the electrical signal representing the data sequence bit0 is applied to the electrode S0, the electrical signal representing the data sequence bit1 is applied to the electrode S1, and the electrical signal representing the data sequence bit2 is applied to the electrode S2.

[0017] In the optical modulator 3 configured as described above, when an electrical signal is applied to the electrode, the refractive index of the arm waveguide changes according to the electrical signal. When the refractive index of the arm waveguide changes, the phase of the light passing through the arm waveguide changes. That is, the electrodes S0 to S2 act as phase shifters. Note that the electrical signals representing each data sequence are applied to the P-arm optical waveguide and the N-arm optical waveguide as differential signals.

[0018] Here, the length of electrode S1 is twice that of electrode S0, and the length of electrode S2 is twice that of electrode S1. Therefore, the phase shift caused by the signal applied to electrode S1 is approximately twice that caused by the signal applied to electrode S0. Similarly, the phase shift caused by the signal applied to electrode S2 is approximately twice that caused by the signal applied to electrode S1.

[0019] Also, in this example, it is assumed that the intensity of the output light of the Mach-Zehnder interferometer is proportional to the phase shift in the arm waveguide. Then, the intensity of the output light of the Mach-Zehnder interferometer will be controlled by the data sequence bit0~bit2.

[0020] For example, assume that the amplitude of the intensity of the output light controlled by the signal bit0 representing the data sequence bit0 is "1". In this case, the amplitude of the intensity of the output light controlled by the signal bit1 representing the data sequence bit1 is "2", and the amplitude of the intensity of the output light controlled by the signal bit2 representing the data sequence bit2 is "4". Then, for example, when the values of bit0~bit2 are "101", the intensity of the output light of the Mach-Zehnder interferometer is "5(=1 + 0 + 4)", and if the values of bit0~bit2 are "010", the intensity of the output light of the Mach-Zehnder interferometer is "2(=0 + 2 + 0)". That is, PAM8 is realized according to bit0~bit2.

[0021] In the following description, the electrode to which the electrical signal representing the transmission data is applied (i.e., the electrode used as the phase shifter) may be referred to as a "phase shift segment" or simply a "segment". That is, the optical modulator shown in FIG. 2 includes segments S0~S2.

[0022] In this way, the optical modulator 3 generates an optical signal representing transmission data by controlling the phase of the input light based on the signal bits 0 to 2. However, if the timings of the signal bits 0 to 2 given to each segment are not appropriately adjusted, the quality of the output optical signal deteriorates. For example, if the timings of the signal bits 0 to 2 given to each segment are not appropriately adjusted, the waveform of the output optical signal collapses. Therefore, it is preferable that the optical modulator 3 includes a delay circuit for appropriately adjusting the timings of the signal bits 0 to 2 given to each segment.

[0023] Also, in the configuration shown in FIG. 2, PAM is realized by appropriately setting the length of the electrode of each segment, but the present invention is not limited to this configuration. For example, PAM may be realized by appropriately setting the number of electrodes. In the example shown in FIG. 4, one electrode (S0) is provided for the data sequence bit0, two electrodes (S1a to S1b) are provided for the data sequence bit1, and four electrodes (S2a to S2d) are provided for the data sequence bit2. In this case, the lengths of the respective electrodes are the same as each other.

[0024] FIG. 5 shows an example of a configuration for realizing high speed of the optical DAC. In this example, each data sequence (bits 0 to 2) is parallelized into K sub-data sequences. K is an arbitrary integer of 2 or more. Also, the data lengths of the K sub-data sequences are the same as each other. That is, the bit rates of the K sub-data sequences are the same as each other. Although only the data sequence bit0 is drawn in FIG. 5, the other data sequences are similarly parallelized into K sub-data sequences.

[0025] DSP2 operates at a clock of frequency f DSP . In this case, the symbol rate of each sub-data sequence is f DSP . And each sub-data sequence is led to the serializer 4. Also, the clock generation circuit 5 generates a clock signal CLK2 from the clock signal CLK1 output from the DSP2. The frequency of the clock signal CLK1 is f DSPand the frequency of the clock signal CLK2 is fs. The frequency fs is, for example, K times the frequency f DSP . Alternatively, as an example, the frequency f DSP is 1 GHz and the frequency fs is 64 GHz.

[0026] The serializer 4 uses the clock signal CLK2 to convert K sub-data sequences into a serial bit sequence. As a result, a bit sequence with a symbol rate (or bit rate) of fs is obtained. Then, this bit sequence is amplified by a binary driver and then given to the corresponding segment provided in the optical modulator 3. Therefore, according to the configuration shown in FIG. 5, the operating speed of the DSP 2 can be reduced.

[0027] By the way, in recent years, ultra-high-speed optical transmission of about 1 Tbps has been demanded. And in order to realize such ultra-high-speed optical transmission, in addition to increasing the number of bits transmitted by each symbol, a sampling rate of about 100 G symbols / second is considered necessary. Therefore, the embodiment of the present invention provides both a modulation method in which each symbol transmits a plurality of bits and a high sampling rate by time-division multiplexing.

[0028] FIG. 6 shows an example of an optical transmitter according to an embodiment of the present invention. In this embodiment, in addition to the parallelization shown in FIG. 5, further speeding up of the optical transmitter is realized by using time-division multiplexing.

[0029] In the configuration shown in FIG. 6, data sequences bit0 to bit2 are generated from the transmission data, and N×K sub-data sequences are generated from each of the data sequences bit0 to bit2. Here, N represents the number of optical signals multiplexed by time-division multiplexing in the optical modulator 3. K is an arbitrary integer of 2 or more. The data lengths of the N×K sub-data sequences are the same as each other. In other words, the bit rates of the N×K sub-data sequences are the same as each other.

[0030] The N×K sub-data sequences are grouped into N groups of sub-data sequences. In this example, N is 2. That is, the 2K sub-data sequences are grouped into 2 groups of sub-data sequences. Here, assume that the 2K sub-data sequences are identified by serial numbers "1" to "2K". In this case, as shown in FIG. 7, one group of sub-data sequences is composed of sub-data sequences with odd serial numbers (1, 3, 5,..., 2K−1). Also, the other group of sub-data sequences is composed of sub-data sequences with even serial numbers (2, 4, 6,..., 2K). Therefore, in the following description, the group of sub-data sequences composed of sub-data sequences with odd serial numbers may be referred to as "sub-data sequence group odd". Also, the group of sub-data sequences composed of sub-data sequences with even serial numbers may be referred to as "sub-data sequence group even". And the sub-data sequences 1, 3, 5,... that make up the sub-data sequence group odd are led to the serializer 4a. Also, the sub-data sequences 2, 4, 6,... that make up the sub-data sequence group even are led to the serializer 4b.

[0031] The clock generation circuit 5 generates a clock signal CLK3 from the clock signal CLK1 output from the DSP2. However, the frequency of the clock signal CLK3 is 1 / N of the frequency of the clock signal CLK2 generated in the configuration shown in FIG. 5. In this embodiment, since N = 2, when the frequency of the clock signal shown in FIG. 5 is fs, the frequency of the clock signal CLK3 is fs / 2. Also, the clock generation circuit 5 outputs N clock signals CLK3 whose phases are shifted by 2π / N from each other. In this embodiment, since N = 2, two clock signals (CLK3_0, CLK3_180) whose phases are shifted by π from each other are output.

[0032] The serializer 4a serializes the sub-data sequences d1, d3, d5, ... using the clock signal CLK3_0. That is, the serializer 4a sequentially selects and outputs 1-bit data from the sub-data sequences d1, d3, d5, ... in synchronization with the clock signal CLK3_0. Thereby, the bit sequence A shown in FIG. 6 is generated. Similarly, the serializer 4b serializes the sub-data sequences d2, d4, d6, ... using the clock signal CLK3_180. That is, the serializer 4b sequentially selects and outputs 1-bit data from the sub-data sequences d2, d4, d6, ... in synchronization with the clock signal CLK3_180. Thereby, the bit sequence B shown in FIG. 6 is generated. The symbol rate (or bit rate) of the bit sequence A and the bit sequence B is the same as each other and is fs / 2 respectively.

[0033] As described with reference to FIG. 2 or FIG. 4, the optical modulator 3 includes electrodes that act as phase shifters for each data sequence bit0 to bit2. However, in the configuration shown in FIG. 6, electrodes to which the bit sequence A is applied and electrodes to which the bit sequence B is applied are provided for each data sequence bit0 to bit2. For example, for the data sequence bit0, an electrode S0a to which the bit sequence A is applied and an electrode S0b to which the bit sequence B is applied are provided. The same applies to the other data sequences bit1 and bit2. Therefore, when each transmission symbol transmits M bits and the number of optical signals multiplexed by time-division multiplexing in the optical modulator 3 is N, the optical modulator 3 includes M×N phase shift segments. Each phase shift segment includes 1 electrode in the configuration shown in FIG. 2 and includes 1 or more electrodes in the configuration shown in FIG. 4. Thus, the optical modulator 3 includes M×N or more electrodes.

[0034] When the first symbol of bit sequence A (i.e., data d1) is applied to electrode S0a, the light passing through the Mach-Zehnder interferometer is modulated according to data d1. Also, when the first symbol of bit sequence B (i.e., data d2) is applied to electrode S0b, the light passing through the Mach-Zehnder interferometer is modulated according to data d2. Subsequently, when the second symbol of bit sequence A (i.e., data d3) is applied to electrode S0a, the light passing through the Mach-Zehnder interferometer is modulated according to data d3. Also, when the second symbol of bit sequence B (i.e., data d4) is applied to electrode S0b, the light passing through the Mach-Zehnder interferometer is modulated according to data d4. Similarly hereinafter, a modulated optical signal is generated according to bit sequence A and bit sequence B.

[0035] Here, the phase of clock signal CLK3_180 for generating bit sequence B is shifted by 180 degrees with respect to the phase of clock signal CLK3_0 for generating bit sequence A. That is, the timing at which serializer 4b outputs bit sequence B using clock signal CLK3_180 is shifted by half of the period of clock signal CLK3 with respect to the timing at which serializer 4a outputs bit sequence A using clock signal CLK3_0. Specifically, the timing at which serializer 4b outputs bit sequence B is shifted by 1 / fs with respect to the timing at which serializer 4a outputs bit sequence A. Therefore, as shown in FIG. 6, in optical modulator 3, modulation components by bit sequence A (d1, d3, d5,...) and modulation components by bit sequence B (d2, d4, d6,...) are generated alternately. As a result, time-division multiplexing (or time-division interleaving) of bit sequence A and bit sequence B is realized.

[0036] Note that in FIGS. 6 to 7, time-division multiplexing of data sequence bit0 has been described, but the same time-division multiplexing is also realized for other data sequences (bit1, bit2). Here, the timings of the electrical signals representing data sequences bit0 to bit2 are appropriately adjusted by a delay circuit (not shown). As a result, amplitude multiplexing of data sequences bit0 to bit2 is realized, and a PAM8 optical signal is generated.

[0037] Thus, when compared with the configuration shown in FIG. 5, in the configuration shown in FIG. 6, the frequency of the clock signal generated by the clock generation circuit 5 is 1 / 2. In the configuration shown in FIG. 6, two sub-data sequence groups A and B are multiplexed. When N sub-data sequence groups are multiplexed, when compared with the configuration shown in FIG. 5, in the configuration shown in FIG. 6, the frequency of the clock signal generated by the clock generation circuit 5 is 1 / N. Therefore, the requirement for the operating speed required for the clock generation circuit 5 is relaxed. Alternatively, when the upper limit operating frequency of the clock generation circuit 5 is fs, a transmission rate of N×fs is achieved.

[0038] Then, by using the above-described time-division multiplexing, further speeding up of the optical transmitter is achieved. However, in order to implement time-division multiplexing, it is necessary to accurately set the timing at which each electrical signal (here, the binary electrical signal representing bit sequence A and the binary electrical signal representing bit sequence B) is applied to the corresponding electrode. In this case, it is necessary to consider the optical propagation time within the Mach-Zehnder interferometer.

[0039] FIG. 8 shows an example of an optical transmitter including a delay circuit that takes into account the optical propagation time. In this embodiment, the distance between electrode S0b and electrode S0a is LX. And an optical propagation delay Topt occurs between electrode S0b and electrode S0a.

[0040] FIG. 9 shows an example of an optical signal generated in the optical modulator of the optical transmitter shown in FIG. 8. Here, when the delay circuit 6 shown in FIG. 8 is not provided, as shown in FIG. 9(a), it is assumed that the optical signal representing the bit sequence B is delayed by 47.6 ps with respect to the optical signal representing the bit sequence A. On the other hand, in order to realize time-division multiplexing of the bit sequence A and the bit sequence B, as shown in FIG. 9(c), a state is required in which the optical signal representing the bit sequence B is delayed by "1 / fs" with respect to the optical signal representing the bit sequence A. fs represents the sampling frequency of the optical signal output from the optical modulator 3. Here, when the sampling frequency fs is 64 GHz, the length of the time slot of time-division multiplexing (i.e., 1 / fs) is 15.6 ps. Therefore, in this case, a state is required in which the optical signal representing the bit sequence B is delayed by 15.6 ps with respect to the optical signal representing the bit sequence A.

[0041] Therefore, in the configuration shown in FIG. 8, the optical transmitter includes a delay circuit 6. In this example, the clock signal CLK_0 supplied to the serializer 4a is delayed by the delay circuit 6. On the other hand, the clock signal CLK_180 is supplied to the serializer 4b without passing through the delay circuit.

[0042] The delay time of the delay circuit 6 is designed to satisfy the above conditions. Thus, as shown in FIG. 9(b), the delay circuit 6 delays the clock signal CLK_0 supplied to the serializer 4a by 32 ps. Then, in the optical modulator 3, a state is realized in which the optical signal representing the bit sequence B is delayed by 15.6 ps with respect to the optical signal representing the bit sequence A. As a result, the time-division multiplexing shown in FIG. 9(c) is realized.

[0043] In this way, by appropriately adjusting the timing of the clock signal supplied to the serializer using a delay circuit, accurate time-division multiplexing can be achieved. However, the delay circuit 6 is realized by, for example, an amplifier circuit. Therefore, implementing the delay circuit 6 increases the power consumption of the optical transmitter. Also, when the delay circuit 6 is implemented, the quality of the optical signal output from the optical transmitter (e.g., jitter characteristics) may deteriorate. Therefore, it is preferable that the optical transmitter does not include the delay circuit 6 for adjusting the timing of the clock signal supplied to the serializer.

[0044] FIG. 10 shows another example of an optical transmitter according to an embodiment of the present invention. In this embodiment, the optical transmitter 10 includes a DSP 2, an optical modulator 3, serializers 4a and 4b, and a clock generation circuit 5, in the same configuration as shown in FIG. 6 or FIG. 8. Note that FIG. 10 shows a configuration for processing the data sequence bit0 of the least significant bit, and the configuration for processing the data sequence of the higher-order bits (data sequences bit1 to bit2 in FIGS. 2 to 4) is omitted. Hereinafter, the configuration for processing the data sequence bit0 of the least significant bit will be described.

[0045] In this embodiment, the clock signal CLK3 generated by the clock generation circuit 5 is supplied to the serializers 4a and 4b. That is, the serializers 4a and 4b output data in synchronization with the same clock signal.

[0046] Assume that the signal propagation time between DSP2 and electrode S0a and the signal propagation time between DSP2 and electrode S0b are the same as each other. That is, the wiring is designed such that the signal propagation time between DSP2 and electrode S0a and the signal propagation time between DSP2 and electrode S0b are the same as each other. Also, assume that the clock signal propagation time between clock generation circuit 5 and serializer 4a and the clock signal propagation time between clock generation circuit 5 and serializer 4b are the same as each other. That is, the wiring is designed such that the clock signal propagation time between clock generation circuit 5 and serializer 4a and the clock signal propagation time between clock generation circuit 5 and serializer 4b are the same as each other.

[0047] In optical transmitter 10, instead of providing delay circuit 6 shown in FIG. 8, time division multiplexing is achieved by appropriately determining the interval between electrodes that act as a phase shifter. That is, time division multiplexing is achieved by appropriately determining the segment interval.

[0048] FIG. 11 shows an example of an optical signal generated in optical modulator 3 of optical transmitter 10 shown in FIG. 10. Assume that the sampling frequency fs of the optical signal output from optical modulator 3 is 64 GHz.

[0049] FIG. 11(a) shows an example of an optical signal generated in a configuration where the segment interval is not appropriately determined. Here, the interval L between electrode S0a and electrode S0b is 2.5 mm. In this case, the optical propagation time Topt required for light to be transmitted from electrode S0b to electrode S0a via the optical waveguide is expressed by the following formula. ng represents the group refractive index of the optical waveguide constituting the Mach-Zehnder interferometer, and in this embodiment, it is 3.84. c represents the speed of light in a vacuum. Topt = L × ng / c

[0050] Therefore, in the case where the timing at which the signal of bit string A (e.g., d1) arrives at electrode S0a coincides with the timing at which the signal of bit string B (e.g., d2) arrives at electrode S0b, the optical signal representing bit string B will be delayed by Topt with respect to the optical signal representing bit string A. In this example, when the interval L is 2.5 mm, the optical propagation time Topt is 32 ps.

[0051] Here, in order to accurately implement time-division multiplexing of bit string A and bit string B, as shown in FIG. 11(c), a state is required in which the optical signal representing bit string B is delayed by "1 / fs" with respect to the optical signal representing bit string A. fs represents the sampling frequency of the optical signal output from optical modulator 3. Here, when the sampling frequency fs is 64 GHz, the length of the time slot for time-division multiplexing (i.e., 1 / fs) is 15.6 ps. Therefore, in this case, a state is required in which the optical signal representing bit string B is delayed by 15.6 ps with respect to the optical signal representing bit string A.

[0052] In optical transmitter 10, the segment interval is determined so as to satisfy the above conditions. Specifically, the interval L is determined such that the optical propagation time Topt required for light to be transmitted from electrode S0b to electrode S0a via the optical waveguide becomes 15.6 ps. In this embodiment, when the interval L is 1.23 mm, the optical propagation time Topt becomes 15.6 ps. Therefore, in optical transmitter 10, the interval L between electrode S0a and electrode S0b is 1.23 mm. Then, as shown in FIG. 11(b), when the timing at which the signal of bit string A (e.g., d1) arrives at electrode S0a coincides with the timing at which the signal of bit string B (e.g., d2) arrives at electrode S0b, the optical signal representing bit string B will be delayed by 15.6 ps (i.e., 1 / fs) with respect to the optical signal representing bit string A. As a result, the time-division multiplexing shown in FIG. 11(c) is realized.

[0053] FIG. 12 shows an example of time-division multiplexing of optical signals. The horizontal axis represents the position in the propagation direction of light within the Mach-Zehnder interferometer. "S0a" and "S0b" represent the positions where electrodes S0a and S0b are provided, respectively. The sampling frequency fs is 64 GHz, and the frequency of the clock signal CLK3 applied to the serializers 4a and 4b is "fs / 2 (32 GHz)". The distance L between electrode S0a and electrode S0b is 1.23 mm. Assume that at time zero, the signal d1 of bit sequence A arrives at electrode S0a, and the signal d2 of bit sequence B arrives at electrode S0b. In the following description, the optical signal generated by the signal di (i = 1, 2,...) may be referred to as optical signal di.

[0054] When one cycle time has elapsed from time zero, the optical signals d1 and d2 have propagated 1.23 mm from electrodes S0a and S0b, respectively. One cycle time means the length of the time slot for time-division multiplexing (or 1 / fs), which is 15.625 ps in this embodiment. When two cycle times have elapsed from time zero, the optical signals d1 and d2 have propagated 2.46 mm from electrodes S0a and S0b, respectively. At this time, signals d3 and d4 arrive at electrodes S0a and S0b, respectively, and optical signals d3 and d4 are generated. As a result, optical signals d1 to d4 are obtained. Similarly hereinafter, new electrical signals arrive at electrodes S0a and S0b every two cycle times, and corresponding optical signals are generated. As a result, time-division multiplexing of the two bit sequences A and B is realized.

[0055] As described above, the optical transmitter 10 shown in FIG. 10 does not include the delay circuit 6 shown in FIG. 8. Therefore, compared with the configuration shown in FIG. 8, the power consumption of the optical transmitter 10 is reduced. In addition, compared with the configuration shown in FIG. 8, the quality (e.g., jitter characteristics) of the optical signal output from the optical transmitter 10 is improved.

[0056] Figures 13 to 14 show an example of the simulation of the operation of the optical transmitter 10. Here, the results of the simulation based on the model shown in Fig. 13(a) are presented. Note that the optical transmitter 10 outputs a PAM4 optical signal. That is, each transmission symbol transmits 2-bit data. Therefore, the DSP2 generates two data sequences (bit0, bit1) from the transmission data. Also, as shown in Fig. 7, each data sequence (bit0, bit1) is divided into two sub-data sequence groups (odd, even).

[0057] In the configuration shown in Fig. 10, the optical DAC 11 corresponds to the serializer 4a and the electrodes to which the signals output from the serializer 4a are applied. Similarly, the optical DAC 12 corresponds to the serializer 4b and the electrodes to which the signals output from the serializer 4b are applied. Note that the optical DAC 11 and the optical DAC 12 output signals synchronized with clock signals having inverted phases with respect to each other. Also, the serializer and the electrodes are provided for each data sequence (bit0, bit1). The adder 13 represents a state in which the optical signals representing each data sequence are multiplexed on the optical waveguide. The Nyquist filter (NF) 14 performs Nyquist filtering. The Nyquist filter 14 is not particularly limited, but corresponds to, for example, an optical bandpass filter provided on the output side of the optical modulator 3. Then, the signal shown in Fig. 13(b) is input to the above simulation model. Fig. 13(b) represents the waveform and spectrum of the input signal.

[0058] Figs. 14(a) to 14(c) represent the waveforms and spectra of the optical signals at nodes A to C shown in Fig. 13(a). That is, Fig. 14(a) represents the waveform and spectrum of the optical signal generated by the optical DAC 11. Fig. 14(b) represents the waveform and spectrum of the optical signal generated by the optical DAC 12. Fig. 14(c) represents the waveform and spectrum of the optical signal obtained by multiplexing the optical signal generated by the optical DAC 11 and the optical signal generated by the optical DAC 12.

[0059] In the spectrum of the multiplexed optical signal, as shown in Fig. 14(c), the signal components in the frequency region separated from the center by fs are greatly suppressed. That is, similar to the case of oversampling at 2fs, the image components are suppressed. Therefore, by using the band-pass filter BPF shown in Fig. 14(c), unnecessary components can be removed. In this embodiment, by providing the Nyquist filter 14, the waveforms and spectra shown in Fig. 14(d) can be obtained.

[0060] Fig. 15 is a diagram for explaining the effect of deleting the delay circuit. That is, hereinafter, the effect of deleting the delay circuit 6 shown in Fig. 8 will be explained.

[0061] Fig. 15(a) shows the jitter of the optical signal output from the optical modulator 3. The horizontal axis of the graph represents the delay time due to the delay circuit 6 shown in Fig. 8. Here, in the examples shown in Figs. 8 to 9, the delay time due to the delay circuit 6 is 32 ps. In this case, jitter of about 7.1 ps occurs. On the other hand, the optical transmitter 10 shown in Fig. 10 does not include the delay circuit 6. Therefore, the characteristics of the optical transmitter 10 correspond to the state corresponding to "delay time = zero" in Fig. 15(a). That is, the jitter of the optical signal output from the optical transmitter 10 is about 4.8 ps. Thus, the jitter characteristics are improved by not using the delay circuit 6.

[0062] Fig. 15(b) shows the power consumption of the optical DAC. In this example, in the optical transmitter 10 without the delay circuit 6, the power consumption of the optical DAC is about 20 mW. On the other hand, as the delay time due to the delay circuit 6 increases, the power consumption in the delay circuit 6 also increases. As a result, in the case of generating a delay time of 32 ps, the power consumption of the optical DAC is about 45.6 mW. Thus, the power consumption can be reduced by not using the delay circuit 6.

[0063] FIG. 16 schematically shows amplitude multiplexing and time-division multiplexing according to an embodiment of the present invention. In this example, the optical transmitter 10 generates an optical signal in which each symbol transmits 3 bits. That is, M = 3. Also, the number of optical signals multiplexed by time-division multiplexing in the optical modulator 3 is 2. That is, N = 2. Therefore, six binary electrical signals (bit0_odd, bit0_even, bit1_odd, bit1_even, bit2_odd, bit2_even) are generated from the transmission data. Note that the binary electrical signal is generated by amplifying, by a driver, a bit sequence serialized by a serializer in the configuration shown in FIG. 10.

[0064] Three optical signals generated by the three binary electrical signals bit0_odd, bit1_odd, and bit2_odd are multiplexed and inserted into the time slot SL1. At this time, the amplitude of the optical signal generated by bit1_odd is twice the amplitude of the optical signal generated by bit0_odd, and the amplitude of the optical signal generated by bit2_odd is twice the amplitude of the optical signal generated by bit1_odd. Thereby, a PAM8 symbol, which is an example of 3-bit intensity modulation, is generated.

[0065] Subsequently, three optical signals generated by the three binary electrical signals bit0_even, bit1_even, and bit2_even are multiplexed and inserted into the time slot SL2, and a PAM8 symbol is generated. Similarly, symbols generated from three binary electrical signals are sequentially inserted into time slots. Thereby, time-division multiplexing is realized.

[0066] As described above, according to the embodiment of the present invention, an optical analog signal having a very high sampling rate or symbol rate and a large number of bits transmitted by each symbol can be generated from a binary electrical signal. Therefore, the embodiment of the present invention contributes to the realization of ultra-high-speed optical transmission.

[0067] Although an optical transmitter that generates a multi-level intensity modulation signal such as PAM4 and PAM8 has been described, the present invention is not limited to only multi-level intensity modulation signals. That is, by appropriately changing the configuration of the optical transmitter, the present invention can be applied to an optical transmitter that generates a multi-level coherent modulation optical signal such as a QPSK signal or a QAM signal. Specifically, by paralleling the optical transmitters shown in FIGS. 8 and 10 and adopting an IQ modulator configuration in which the optical outputs of the Mach-Zehnder interferometers are combined by a parent Mach-Zehnder interferometer, a coherent modulation optical signal is generated.

[0068] <First Embodiment> FIG. 17 shows a first embodiment of an optical transmitter. The optical transmitter 10 according to the first embodiment includes a DSP 2, an optical modulator 3, a clock generation circuit 5, an encoder 21, serializers 31 to 34, delay elements 35 to 36, and drivers 37 to 40.

[0069] In this embodiment, M = 2. That is, the optical transmitter 10 generates an optical signal in which each symbol transmits 2 bits. Therefore, the encoder 21 generates two data sequences (bit0, bit1). Also, in this embodiment, N = 2. Therefore, the encoder 21 generates two sub-data sequence groups (odd, even) from each data sequence. That is, four sub-data sequence groups (bit0_odd, bit0_even, bit1_odd, bit1_even) are generated in the encoder 21. Note that the encoder 21 is realized by, for example, a hardware logic circuit. Alternatively, the encoder 21 may be incorporated in the DSP 2.

[0070] As described with reference to FIG. 7, each sub-data sequence group is composed of a plurality of sub-data sequences. The plurality of sub-data sequences constituting each sub-data sequence group are output in parallel and led to the corresponding serializer.

[0071] The serializers 31 to 34 serialize the corresponding plurality of sub-data sequences respectively using the clock signal generated by the clock generation circuit 5. Specifically, the serializer 31 generates the bit sequence bit0_even by serializing the plurality of sub-data sequences belonging to the sub-data sequence group bit0_even. The serializer 32 generates the bit sequence bit1_even by serializing the plurality of sub-data sequences belonging to the sub-data sequence group bit1_even. The serializer 33 generates the bit sequence bit0_odd by serializing the plurality of sub-data sequences belonging to the sub-data sequence group bit0_odd. The serializer 34 generates the bit sequence bit1_odd by serializing the plurality of sub-data sequences belonging to the sub-data sequence group bit1_odd. Then, the bit sequences output from the serializers 31 to 34 are respectively led to the drivers 37 to 40. Note that the clock signals supplied to the serializers 32 and 34 are respectively delayed by the delay elements 35 and 36.

[0072] The drivers 37 to 40 generate binary electrical signals from the bit sequences output from the serializers 31 to 34 respectively. Specifically, the driver 37 generates the binary electrical signal bit0_even from the bit sequence bit0_even output from the serializer 31. The driver 38 generates the binary electrical signal bit1_even from the bit sequence bit1_even output from the serializer 32. The driver 39 generates the binary electrical signal bit0_odd from the bit sequence bit0_odd output from the serializer 33. The driver 40 generates the binary electrical signal bit1_odd from the bit sequence bit1_odd output from the serializer 34. Thus, M×N (i.e., 4) binary electrical signals are generated. Then, these binary electrical signals are supplied to the optical modulator 3.

[0073] In the optical modulator 3, a plurality of electrodes are provided along the optical path of the Mach-Zehnder interferometer. In this embodiment, M×N (i.e., 4) electrodes are provided. Although omitted for ease of viewing the drawings, the electrodes are provided on both the P arm and the N arm.

[0074] In this example, the electrode S0_ev, the electrode S1_ev, the electrode S0_od, and the electrode S1_od are provided in order from the input end to the output end of the Mach-Zehnder interferometer. And the binary electrical signals bit0_even, bit1_even, bit0_odd, and bit1_odd are applied to the electrode S0_ev, the electrode S1_ev, the electrode S0_od, and the electrode S1_od, respectively. Note that the lengths of the electrode S0_ev and the electrode S0_od are the same as each other, and the lengths of the electrode S1_ev and the electrode S1_od are the same as each other. Also, the lengths of the electrode S1_ev and the electrode S1_od are each twice the lengths of the electrode S0_ev and the electrode S0_od.

[0075] The plurality of electrodes are grouped based on time slots of time-division multiplexing. In this embodiment, the electrode S0_ev and the electrode S1_ev belong to the electrode group even, and the electrode S0_od and the electrode S1_od belong to the electrode group odd.

[0076] The distance L1 between the electrode included in the electrode group even and the corresponding electrode included in the electrode group odd is represented by the following formula. L1 = c / (ng×fs) That is, the distance between electrode S0_ev and electrode S0_od is L1, and the distance between electrode S1_ev and electrode S1_od is also L1. Note that fs corresponds to the length of one time slot when the optical signals generated according to the signals applied to electrode group even and the optical signals generated according to the signals applied to electrode group odd are multiplexed in the time domain. Therefore, the distance L1 corresponds to the distance that light propagates through the optical waveguide in the time corresponding to the time slot of time-division multiplexing. Also, in this embodiment, since each symbol transmits M bits, the distance L1 corresponds to the distance that light propagates through the optical path during the period M / B when the bit rate of the transmitted data is B.

[0077] The distance L2 between the electrodes within each electrode group is not particularly limited. However, in order to make the timings of bit0 and bit1 of the PAM4 optical signal coincide with each other, it is required that the time for light to propagate the distance L2 is the same as the delay times of delay elements 35 and 36. Therefore, the delay element 35 is designed such that the delay time of the delay element 35 is the same as the time for light to propagate from electrode S0_ev to electrode S1_ev through the optical waveguide. Similarly, the delay element 36 is designed such that the delay time of the delay element 36 is the same as the time for light to propagate from electrode S0_od to electrode S1_od through the optical waveguide.

[0078] Note that in FIG. 17, the encoder 21, serializers 31 to 34, delay elements 35 to 36, drivers 37 to 40, and clock generation circuit 5 are an example of a signal generation circuit that generates M×N binary electrical signals having the same bit rate from the transmitted data. Also, electrodes S0_ev, S1_ev, S0_od, and S1_od are an example of M×N phase shift segments that shift the phase of the light propagating through the optical path according to the M×N binary electrical signals, respectively.

[0079] FIG. 18 shows an example of amplitude multiplexing and time division multiplexing by the optical transmitter shown in FIG. 17. The horizontal axis represents the position in the optical propagation direction within the Mach-Zehnder interferometer. "S0_ev", "S1_ev", "S0_od", and "S1_od" represent the positions where the electrodes S0_ev, S1_ev, S0_od, and S1_od are provided, respectively.

[0080] At time T0, the electrical signal bit0_1 and the electrical signal bit0_2 arrive at the electrode S0_od and the electrode S0_ev, respectively. Then, an optical signal bit0_1 is generated at the electrode S0_od, and an optical signal bit0_2 is generated at the electrode S0_ev.

[0081] At time T1, the optical signal bit0_1 arrives at the electrode S1_od, and the optical signal bit0_2 arrives at the electrode S1_ev. Also, the electrical signal bit1_1 and the electrical signal bit1_2 arrive at the electrode S1_od and the electrode S1_ev, respectively. Then, an optical signal bit1_1 is generated at the electrode S1_od, and an optical signal bit1_2 is generated at the electrode S1_ev. Therefore, at the electrode S1_od, the optical signal bit0_1 and the optical signal bit1_1 are multiplexed, and the transmission symbol 1 is generated. Similarly, at the electrode S1_ev, the optical signal bit0_2 and the optical signal bit1_2 are multiplexed, and the transmission symbol 2 is generated. After that, from time T1 to time T4, the transmission symbols 1 and 2 propagate toward the output port of the Mach-Zehnder interferometer.

[0082] At time T4, the electrical signal bit0_3 and the electrical signal bit0_4 arrive at the electrode S0_od and the electrode S0_ev, respectively. Then, an optical signal bit0_3 is generated at the electrode S0_od, and an optical signal bit0_4 is generated at the electrode S0_ev.

[0083] At time T5, the optical signal bit0_3 arrives at electrode S1_od, and the optical signal bit0_4 arrives at electrode S1_ev. Also, the electrical signal bit1_3 and the electrical signal bit1_4 arrive at electrodes S1_od and S1_ev respectively. Then, the optical signal bit1_3 is generated at electrode S1_od, and the optical signal bit1_4 is generated at electrode S1_ev. Therefore, at electrode S1_od, the optical signal bit0_3 and the optical signal bit1_3 are multiplexed, and the transmission symbol 3 is generated. Similarly, at electrode S1_ev, the optical signal bit0_4 and the optical signal bit1_4 are multiplexed, and the transmission symbol 4 is generated. After that, the transmission symbols 1 to 4 propagate toward the output port of the Mach-Zehnder interferometer. In this way, amplitude multiplexing (bit0, bit1) and time-division multiplexing (odd, even) are realized simultaneously.

[0084] <Second Embodiment> FIG. 19 shows a second embodiment of the optical transmitter. In the second embodiment, M = 3. That is, the optical transmitter 10 generates an optical signal in which each symbol transmits 3 bits. Therefore, the encoder 21 generates three data sequences (bit0, bit1, bit2). Also, in this embodiment, N = 2. Therefore, the encoder 21 generates two sub-data sequence groups (odd, even) from each data sequence respectively. That is, six sub-data sequence groups (bit0_odd, bit0_even, bit1_odd, bit1_even, bit2_odd, bit2_even) are generated in the encoder 21. Therefore, six bit sequences are generated.

[0085] The optical modulator 3 includes phase shift segments for each bit sequence. In FIG. 19, phase shift segments S0_ev, S1_ev, S2_ev, S0_od, S1_od, S2_od are provided.

[0086] Each phase-shifting segment includes one or more electrodes. Specifically, the phase-shifting segments S0_ev and S0_od for generating the optical signal corresponding to bit0 each include one electrode. The phase-shifting segments S1_ev and S1_od for generating the optical signal corresponding to bit1 each include two electrodes. The phase-shifting segments S2_ev and S2_od for generating the optical signal corresponding to bit2 each include four electrodes. In this configuration, the lengths of all the electrodes are the same.

[0087] The phase-shifting segments S0_od, S1_od, and S2_od form one electrode group, and the phase-shifting segments S0_ev, S1_ev, and S2_ev also form one electrode group. And between the groups, the distance L1 between the corresponding phase-shifting segments (or electrodes) is represented by the following formula. L1 = c / (ng × fs) Specifically, the distance between the phase-shifting segment S0_ev and the phase-shifting segment S0_od is L1. Also, the distance between each electrode of the phase-shifting segment S1_ev and each electrode of the phase-shifting segment S1_od is also L1 respectively. Further, the distance between each electrode of the phase-shifting segment S2_ev and each electrode of the phase-shifting segment S2_od is also L1 respectively.

[0088] <Transceiver> FIG. 20 shows an example of an optical transceiver including an optical transmitter according to an embodiment of the present invention. This optical transceiver includes a DSP chip 101, a transmission circuit chip 102, an optical integrated circuit chip 103, a reception circuit chip 104, and a light source 105, and is mounted on a substrate 100.

[0089] The transmission circuit chip 102 includes a serializer, a clock generation circuit, a driver, an encoder, etc. shown in FIG. 10. However, the encoder may be realized by the DSP chip 101. The optical integrated circuit chip 103 includes the optical modulator 3 shown in FIG. 10, and generates a modulated optical signal using the continuous light generated by the light source 105. Also, the optical integrated circuit chip 103 generates an electrical signal representing the received optical signal, for example, by coherent reception. In this case, the optical integrated circuit chip 103 performs coherent reception using the continuous light generated by the light source 105. The DSP chip 101 includes the DSP2 shown in FIG. 10. Also, the DSP chip 101 regenerates data from the output signal of the reception circuit chip 104.

Explanation of Signs

[0090] 1 Optical transmitter 2 DSP 3 Optical modulator 4, 4a, 4b Serializer 5 Clock generation circuit 6 Delay circuit 10 Optical transmitter 21 Encoder 31~34 Serializer 35~36 Delay element 37~40 Driver

Claims

1. An optical transmitter that transmits a modulated optical signal in which each symbol transmits M (M is an integer of 2 or more) bits, When the optical transmitter multiplexes N (N is an integer of 2 or more) optical signals by time-division multiplexing, a signal generation circuit that generates M×N binary electrical signals having the same bit rate from transmission data, A Mach-Zehnder interferometer, M×N phase shift segments provided along the optical path of the Mach-Zehnder interferometer, and each shifting the phase of the light propagating through the optical path according to the M×N binary electrical signals, The M×N phase shift segments are composed of N electrode groups, The N electrode groups are provided in series along the optical path, The distance between the electrodes included in the first electrode group among the N electrode groups and the corresponding electrodes included in the second electrode group adjacent to the first electrode group is the distance that light propagates through the optical path in the time corresponding to one time slot of the time-division multiplexing, Each electrode group includes M or more electrodes to which the corresponding M binary electrical signals among the M×N binary electrical signals are applied An optical transmitter characterized by the above.

2. An optical transmitter that transmits a modulated optical signal in which each symbol transmits M (M is an integer of 2 or more) bits, When the optical transmitter multiplexes N (N is an integer of 2 or more) optical signals by time-division multiplexing, a signal generation circuit that generates M×N binary electrical signals having the same bit rate from transmission data, A Mach-Zehnder interferometer, M×N phase shift segments provided along the optical path of the Mach-Zehnder interferometer, and each shifting the phase of the light propagating through the optical path according to the M×N binary electrical signals, The M×N phase shift segments are composed of N electrode groups, The N electrode groups are provided in series along the optical path, The distance between the electrodes included in the first electrode group among the N electrode groups and the corresponding electrodes included in the second electrode group adjacent to the first electrode group is the distance that light propagates through the optical path during the period M / B when the bit rate of the transmission data is B, Each electrode group includes M or more electrodes to which the corresponding M binary electrical signals among the M×N binary electrical signals are applied An optical transmitter characterized by the above.

3. Each electrode group is composed of M electrodes with different lengths from each other The optical transmitter according to claim 1 or 2, characterized in that

4. Each electrode group is composed of more than M electrodes with the same length from each other The optical transmitter according to claim 1 or 2, characterized in that

5. The signal generation circuit adjusts the timing of the M×N binary electrical signals so that M optical signals corresponding to the M binary electrical signals applied to each electrode group overlap with each other in the time domain The optical transmitter according to claim 1 or 2, characterized in that

6. The signal generation circuit An encoder that generates M×N×K (K is an integer of 2 or more) sub-data sequences from the transmission data, M×N serializers, M×N drivers, A clock generation circuit that generates a clock signal, and includes The M×N serializers generate M×N bit sequences by sequentially selecting bits from the corresponding K sub-data sequences using the clock signal and outputting them, The M×N drivers generate the M×N binary electrical signals from the M×N bit sequences The optical transmitter according to claim 1 or 2, characterized in that

7. An optical transceiver including an optical receiver and an optical transmitter that transmits a modulated optical signal in which each symbol transmits M (M is an integer of 2 or more) bits, The optical transmitter A signal generation circuit that generates M×N binary electrical signals with the same bit rate from each other from the transmission data when the optical transmitter multiplexes N (N is an integer of 2 or more) optical signals by time-division multiplexing, A Mach-Zehnder interferometer, M×N phase shift segments provided along the optical path of the Mach-Zehnder interferometer, and each phase shift the phase of the light propagating along the optical path according to the M×N binary electrical signals, The M×N phase shift segments are composed of N electrode groups, The N electrode groups are provided in series along the optical path, The distance between the electrodes included in the first electrode group among the N electrode groups and the corresponding electrodes included in the second electrode group adjacent to the first electrode group is the distance that light propagates along the optical path in the time corresponding to one time slot of the time-division multiplexing, Each electrode group includes M or more electrodes to which the corresponding M binary electrical signals among the M×N binary electrical signals are applied An optical transceiver characterized by the above.

8. An optical transceiver including an optical receiver and an optical transmitter that transmits a modulated optical signal in which each symbol transmits M (M is an integer of 2 or more) bits, wherein the optical transmitter is a signal generation circuit that generates M×N binary electrical signals having the same bit rate from each other from transmission data when the optical transmitter multiplexes N (N is an integer of 2 or more) optical signals by time-division multiplexing; a Mach-Zehnder interferometer; M×N phase shift segments provided along the optical path of the Mach-Zehnder interferometer, each of which shifts the phase of light propagating through the optical path according to the M×N binary electrical signals; the M×N phase shift segments are composed of N electrode groups; the N electrode groups are provided in series along the optical path; the distance between an electrode included in a first electrode group among the N electrode groups and a corresponding electrode included in a second electrode group adjacent to the first electrode group is a distance that light propagates through the optical path during a period M / B when the bit rate of the transmission data is B; each electrode group includes M or more electrodes to which corresponding M binary electrical signals among the M×N binary electrical signals are applied An optical transceiver characterized by the above.

Images