JP5119492B2 - Optical flip-flop - Google Patents

Description

  The present invention relates to an optical flip-flop necessary for realizing optical digital signal processing.

  In an optical communication system, an optical signal is usually used for transmission and an electric signal is used for signal processing. However, with the increase in signal speed, it is expected that part of signal processing is performed as an optical signal.

  A logic circuit that performs digital signal processing generally includes an AND circuit, an OR circuit, and an inverting circuit as basic elements. A flip-flop is an important component of the logic circuit. There are several types of flip-flops, one of which is a set-reset (SR) flip-flop. FIG. 10 shows a circuit diagram and characteristic table of the SR flip-flop.

  The operation of the SR flip-flop will be described with reference to FIG. When the first input (S) called a set input becomes 1, the output (Q) becomes 1, and the output Q continues to hold 1 even after the input S becomes 0. Further, when the second input (R) called a reset input becomes 1, the output Q becomes 0, and the output Q continues to hold 0 even after the input R becomes 0. Note that the operation when both the input S and the input R are 1 is not guaranteed, and such an input state is prohibited.

  In a logic circuit that operates with an electrical signal, it is common to determine the binary value of 1 and 0 based on the voltage level. When the threshold voltage Vt is used as a reference, the voltage is higher than Vt and 1 is lower. 0. Various flip-flops that operate on electrical signals are actually used in processors and memories.

  In order to realize optical digital signal processing, proposals have been made for optical logic circuits and all-optical elements (for example, see Patent Document 1 or 2). In an optical logic circuit as well, it is common that binary values of 1 and 0 correspond to the light intensity level. The applicant of the present application has already proposed an optical SR flip-flop operating with an optical signal (Japanese Patent Application No. 2006-224165). The configuration of the optical SR flip-flop according to this proposal is shown in FIG.

  The optical SR flip-flop of FIG. 11 has Mach-Zehnder type optical switches 111 and 112.

  In the Mach-Zehnder type optical switch 111, nonlinear waveguides 61 and 62 are arranged on both arms, and a set light pulse (S) input port 63, a reset light pulse (R) input port 65, continuous (CW) light (B). An input port 64 and an optical output port 66 are provided. The output light from the optical output port 66 is output from the optical output port 68 as the output light (Q) through the optical branching unit 67, and a part thereof is used as the input light (A1) to the Mach-Zehnder type optical switch 112. It is done.

  In the Mach-Zehnder optical switch 112, the nonlinear waveguides 71 and 72 are arranged on both arms, and the optical input port 73 and the optical input port 75 for inputting the input light (A1) via the optical branching unit 77, CW light (A2) ) And an optical output port 76 for outputting output light (A3). Further, the output light A3 from the Mach-Zehnder type optical switch 112 is fed back to the set light pulse input port 63 of the Mach-Zehnder type optical switch 111 via the optical multiplexing unit 78.

  In the Mach-Zehnder optical switch 111, the set light pulse S excites the nonlinear optical effect in the nonlinear waveguide 61, and the reset light pulse R excites the nonlinear optical effect in the nonlinear waveguide 62. The CW light B is input to the continuous light input port 64, is once branched, passes through the nonlinear waveguides 61 and 62, and is then combined. The branched CW light B is given a phase shift (nonlinear phase shift) according to the excited nonlinear optical effect when passing through the nonlinear waveguide 61 or 62. As a result, output light Q having an intensity corresponding to the phase difference of the light when combined and interfered is generated and output from the optical output port 66. If the phase difference at the time of interference is 0, the output light Q is in a high intensity state, that is, 1 state. Conversely, if the phase difference at the time of interference is π, the output light Q is in a low intensity state, that is, in the 0 state.

  In the Mach-Zehnder type optical switch 112, the input light A1 is branched by the optical branching unit 77 and then given a time difference and an intensity difference, and one of them is inputted from the optical input port 73 and nonlinear optical effect in the nonlinear waveguide 71 is obtained. The other is input from the optical input port 75 and used for exciting the nonlinear optical effect in the nonlinear waveguide 72. The CW light A2 is input to the optical input port 74, branched once, passes through the nonlinear waveguides 71 and 72, is given a nonlinear phase shift, and then multiplexed. As a result, output light A3 having an intensity corresponding to the phase difference of the light when combined and interfered is generated and output from the optical output port 76. When the input light A1 is 0, the nonlinear waveguides 71 and 72 are not excited, so that the phase difference when the input light A2 passes through the nonlinear waveguide and interferes is π, and the optical output port The output light A3 from 76 is in a low intensity state, that is, 0. On the contrary, when the input light A1 is 1, excitation to the nonlinear waveguides 71 and 72 is performed, so that the phase difference when the input light A2 passes through the nonlinear waveguide and interferes is 0. The output light A3 from the output port 76 is in a high intensity state, that is, 1.

  Next, the operation of the optical SR flip-flop in FIG. 11 will be described with reference to the waveform diagram shown in FIG. The set light pulse light S and the reset light pulse R are pulse trains having a pulse width of 5 ps and a pulse repetition interval of 400 ps, as shown in FIGS. The interval between the set light pulse S and the reset light pulse R is set to 200 ps.

  First, in the initial state, as shown in FIG. 12D, the output light Q is in a low intensity state, that is, in the 0 state. That is, the phase difference when the input light B passes through the nonlinear waveguide 61 and the nonlinear waveguide 62 and interferes is π. At this time, since the input light A1 is in the 0 state in the Mach-Zehnder optical switch 112, the output light A3 is also in the 0 state as shown in FIG.

  Next, when the set light pulse S is input to the set light pulse input port 63, the nonlinear optical effect is excited in the nonlinear waveguide 61, and a nonlinear phase shift is caused when the input light B passes through the nonlinear waveguide 61. receive. Normally, this nonlinear phase shift is desirably a value close to π. In that case, the phase difference when the input light B interferes changes to 0, and the output light Q changes to the 1 state. At this time, in the Mach-Zehnder optical switch 112, the input light A1 changes to 1 state, and the output light A3 also changes to 1 state. The output light A3 is fed back as excitation light for the nonlinear waveguide 61, and the output light Q is kept in the 1 state.

  The time required to start feedback, that is, the time from when the set light pulse S is input to when the output light A3 of the Mach-Zehnder optical switch 112 starts to be input at the set light pulse input port 63 is 10 ps. . This time is determined by the optical path length of the portion used for forming the feedback light A3.

  Next, when the reset light pulse R is input from the reset light pulse input port 65 200 ps after the set light pulse S is input, the nonlinear optical effect is excited in the nonlinear waveguide 62. When the input light B passes through the nonlinear waveguide 62, it undergoes a nonlinear phase shift, so that the phase difference when the input light B interferes returns to π, which is the same value as the initial state, and the output light Q returns to the 0 state. Return. At this time, in the Mach-Zehnder optical switch 112, the input light A1 becomes 0 state, and the output light A3 also returns to 0 state.

  As described above, an optical SR flip-flop is realized.

JP 59-168421 A Japanese Patent Laid-Open No. 7-20510

  The optical SR flip-flop is provided with two input optical ports, and the state of the output light is controlled by inputting the set light pulse and the reset light pulse to the respective ports. The SR flip-flop is the basis of various flip-flops, but a delay (D) flip-flop is necessary to realize various functions such as waveform shaping, a shift register, and a counter. The D flip-flop is sometimes called a data flip-flop. A circuit diagram and a characteristic table of the D flip-flop are shown in FIGS. 13A and 13B, respectively. The characteristic table of FIG. 13B is based on the premise that the clock pulse has a sufficiently short width.

  The optical D flip-flop has two input optical ports. When data light and clock light are input, the optical D flip-flop controls the state of output light based on the state of the data light at the time when the clock light is input. . Realization of an optical logic circuit capable of obtaining an optical output synchronized with such a clock light is demanded.

  An object of the present invention is to provide an optical D flip-flop.

  Neither Patent Document 1 nor 2 discloses or suggests means for realizing an optical D flip-flop.

  An optical flip-flop according to the present invention receives an optical set-reset flip-flop having a set light input port and a reset light input port, binary intensity modulated data light and clock light or CW light, and is based on the data light Switching means for distributing a part or all of the clock light or CW light to the set light input port and the reset light input port, and is connected to the optical output port or the optical output port of the optical set reset flip-flop. A delayed data light that is substantially equal to the delayed data light is output to the output port of the optical circuit.

  The optical flip-flop according to the present invention includes two input optical ports, and the state of the output light is controlled by inputting the data light and the clock light pulse to the respective ports. Thereby, the D flip-flop operation is realized.

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

  The configuration of the optical flip-flop according to the first embodiment of the present invention is shown in FIG. This optical D flip-flop includes a data optical (D) input port 1, a clock optical pulse (Clk) input port 2, and an optical output (Q) port 3. A Mach-Zehnder optical switch 102 and an optical SR flip-flop 101 are provided. Consists of.

  In the Mach-Zehnder type optical switch 102, the nonlinear waveguides 11 and 12 are arranged in both arms, and the data light D is input to the optical input port 13 and the optical input port 15 via the data optical input port 1, The clock light pulse Clk is input to the light input port 14 via the clock light pulse input port 2, and output light is output from the light output port 16 and the light output port 17.

  The optical SR flip-flop 101 includes a set light pulse (S) input port 4 and a reset light pulse (R) input port 5, which are connected to the optical output port 16 and the optical output port 17, respectively. The light is output from the output port 6 and guided to the optical output port 3.

  In the Mach-Zehnder optical switch 102, the data light D excites the nonlinear optical effect in the nonlinear waveguides 11 and 12. On the other hand, the clock light pulse Clk is inputted to the optical input port 14, then once branched, passed through the nonlinear waveguides 11 and 12, and then multiplexed. The branched clock light pulse Clk is given a phase shift (nonlinear phase shift) corresponding to the excited nonlinear optical effect when passing through the nonlinear waveguide 11 or 12. As a result, output light (S, R) having an intensity corresponding to the phase difference of the light when combined and interfered is generated and output from the optical output port 16 and the optical output port 17.

  When the data light D is 0, excitation to the nonlinear waveguides 11 and 12 is not performed, so that the phase difference when the clock light pulse Clk passes through the nonlinear waveguides 11 and 12 and interferes is π. The output light intensity at the light output port 16 is low, and the output light intensity at the light output port 17 is high. On the other hand, when the data light D is 1, excitation to the nonlinear waveguides 11 and 12 is performed, so that the phase difference when the clock light pulse Clk passes through the nonlinear waveguides 11 and 12 and interferes is 0. In this state, the output light intensity of the light output port 16 is 1 state high, and the output light intensity of the light output port 17 is 0 state low.

  In the optical SR flip-flop 101, when the set light pulse S is input to the set light pulse input port 4, the light intensity of the output light Q from the optical output port 6 transitions to one state. The output light state is maintained until it is input to the reset light pulse input port 5. Conversely, when the reset light pulse R is input to the reset light pulse input port 5, the light intensity of the output light Q transitions to a low 0 state, and thereafter the set light pulse S is input to the set light pulse input port 4. Until the output light state is maintained.

  Next, the operation of the optical D flip-flop according to the first embodiment will be described with reference to the waveform diagrams shown in FIGS. 2A to 2E show waveforms of the data light D, the clock light pulse Clk, the set light pulse S, the reset light pulse R, and the output light Q, respectively. Here, the data light D is a non-zero return (NRZ) modulated optical signal having a bit rate of 40 Gb / s, and the clock light pulse Clk is an optical pulse train having a repetition frequency of 40 GHz and a pulse width of 6 ps.

  When the data light D is in the 0 state, the nonlinear waveguides 11 and 12 are not excited in the Mach-Zehnder optical switch 102. At this time, the clock light pulse Clk passes through the nonlinear waveguides 11 and 12. The phase difference at the time of multiplexing is π. For this reason, an optical pulse is not output from the optical output port 16, but an optical pulse is output from the optical output port 17. That is, in the optical SR flip-flop 101, the reset light pulse R is continuously input to the reset light pulse input port 5, and the light Q output via the light output port 6 and the light output port 3 is maintained in the 0 state. .

  Next, when the data light D changes to the 1 state, the Mach-Zehnder type optical switch 102 starts excitation of the nonlinear waveguides 11 and 12. At this time, the phase difference when the clock light pulse Clk is combined after passing through the nonlinear waveguides 11 and 12 becomes zero. For this reason, after the first clock light pulse input after the data light D is in the 1 state, the light pulse is output from the light output port 16 and the light pulse is not output from the light output port 17. That is, in the optical SR flip-flop 101, the set optical pulse S is input to the set optical pulse input port 4, and the light Q output via the optical output port 6 and the optical output port 3 transitions to the 1 state. , Retained.

  Further, when the data light D changes to the 0 state, the Mach-Zehnder optical switch 102 does not excite the nonlinear waveguides 11 and 12. At this time, the phase difference when the clock light pulse Clk is combined through the nonlinear waveguides 11 and 12 is π. For this reason, after the first clock light pulse Clk input after the data light D is in the 0 state, the light pulse is not output from the light output port 16 and the light pulse is output from the light output port 17. become. That is, in the optical SR flip-flop 101, the reset optical pulse R is input to the reset optical pulse input port 5, and the light Q output via the optical output port 6 and the optical output port 3 transitions to the 0 state. , Retained.

  As described above, the optical D flip-flop that controls the state of the output light corresponding to the state of the data light at the time when the clock light pulse is input is realized. In the description of the waveform diagram of FIG. 2, the case where the input data light D is an NRZ optical signal is taken as an example, but the same operation is realized when an RZ optical signal is used.

  In the above description, the case where the Mach-Zehnder optical switch 102 performs excitation with the data light D on both the nonlinear waveguides 11 and 12 has been described. Although it is possible to control the state of the output light by exciting either one of the nonlinear waveguides 11 or 12, the excitation time of the nonlinear optical effect can be reduced by exciting both arms and utilizing the time difference or the intensity difference. The influence is suppressed and high-speed operation becomes possible.

  For example, Tajima's “Japanese Journal of Applied Physics” describes a method for enabling high-speed operation by utilizing the time difference between excitations of the nonlinear waveguides of both arms when excited by RZ-modulated data light. (Japanese Journal of Applied Physics), Vol. 32, pages L1746 to L1749, 1993. In addition, when excited by NRZ-modulated data light, a method for enabling high-speed operation by utilizing the time difference and the intensity difference of excitation of the nonlinear waveguides of both arms is described in, for example, “Optical Fiber” by Nakamura et al. "Communication Conference" (Optical Fiber Communication Conference), paper number FD3, 2004.

  The configuration of the optical flip-flop according to the second embodiment of the present invention is shown in FIG. This optical D flip-flop includes a data optical input port 1 for data light D, a clock optical pulse input port 2 for clock optical pulse Clk, and an optical output port 3 for output light Q, and is a Mach-Zehnder optical switch. 103 and 104 and an optical SR flip-flop 101.

  In the Mach-Zehnder optical switch 103, the nonlinear waveguides 21 and 22 are arranged on both arms, and the data light D is input to the optical input port 23 and the optical input port 25 via the data optical input port 1, The clock light pulse Clk is input to the light input port 24 via the clock light input port 2, and output light (S) is output from the light output port 26.

  In the Mach-Zehnder type optical switch 104, the nonlinear waveguides 31 and 32 are arranged on both arms, and the data light D is input to the optical input port 33 and the optical input port 35 via the data optical input port 1. Then, the clock light pulse Clk is input to the light input port 34 via the clock light pulse input port 2, and the output light (R) is output from the light output port 36.

  The optical SR flip-flop 101 includes a set optical pulse input port 4 for the set optical pulse S and a reset optical pulse input port 5 for the reset optical pulse R, which are connected to the optical output port 26 and the optical output port 36, respectively. The output light Q is output from the optical output port 6 and guided to the optical output port 3.

  In the Mach-Zehnder optical switch 103, the data light D excites the nonlinear optical effect in the nonlinear waveguides 21 and 22. On the other hand, the clock light pulse Clk is input to the optical input port 24, branched once, passed through the nonlinear waveguides 21 and 22, and given a nonlinear phase shift, and then multiplexed. As a result, output light (S) having an intensity corresponding to the phase difference of the light when combined and interfered is output from the optical output port 26.

  When the data light D is 0, excitation to the nonlinear waveguides 21 and 22 is not performed, so that the phase difference when the clock light pulse Clk passes through the nonlinear waveguides 21 and 22 and interferes is π. The output light intensity of the light output port 26 is in a zero state. Conversely, when the data light D is in the 1 state, excitation to the nonlinear waveguides 21 and 22 is performed, so that the phase difference when the clock light pulse Clk passes through the nonlinear waveguides 21 and 22 and interferes is 0, and the output light intensity of the light output port 26 is in the 1 state.

  In the Mach-Zehnder optical switch 104, the data light D excites the nonlinear optical effect in the nonlinear waveguides 31 and 32. On the other hand, the clock light pulse Clk is input to the optical input port 34, branched once, passed through the nonlinear waveguides 31 and 32, and given a nonlinear phase shift, and then multiplexed. As a result, output light (R) having an intensity corresponding to the phase difference of the light when combined and interfered is output from the optical output port 36.

  When the data light D is 0, excitation to the nonlinear waveguides 31 and 32 is not performed, and therefore the phase difference when the clock light pulse Clk passes through the nonlinear waveguides 31 and 32 and interferes is 0. In this state, the output light intensity of the light output port 36 is high. Conversely, when the data light D is in the 1 state, excitation to the nonlinear waveguides 31 and 32 is performed, so that the phase difference when the clock light pulse Clk passes through the nonlinear waveguides 31 and 32 and interferes is as follows. π, and the output light intensity of the light output port 36 is in a low 0 state. That is, the output of the Mach-Zehnder optical switch 103 and the output of the Mach-Zehnder optical switch 104 are logically inverted from each other. The optical path amount of each light is determined so that the output of the Mach-Zehnder type optical switch 103 and the output of the Mach-Zehnder type optical switch 104 are logically inverted.

  In the optical SR flip-flop 101, when the set light pulse S is input to the set light pulse input port 4, the light intensity of the output light Q from the optical output port 6 transitions to one state. The output light state is maintained until it is input to the reset light pulse input port 5. Conversely, when the reset light pulse R is input to the reset light pulse input port 5, the light intensity of the output light Q transitions to a low 0 state, and thereafter the set light pulse S is input to the set light pulse input port 4. Until the output light state is maintained.

  The operation of the optical D flip-flop according to the second embodiment is described with reference to the waveform diagram shown in FIG. 2 in the same manner as the operation according to the first embodiment. Again, the data light D is an NRZ optical signal with a bit rate of 40 Gb / s, and the clock light pulse Clk is an optical pulse train with a repetition frequency of 40 GHz and a pulse width of 6 ps.

  When the data light D is in the 0 state, the Mach-Zehnder type optical switch 103 and the Mach-Zehnder type optical switch 104 do not excite the nonlinear waveguide. At this time, an optical pulse is output from the optical output port 26. Instead, an optical pulse is output from the optical output port 36. That is, in the optical SR flip-flop 101, the reset light pulse R is continuously input to the reset light pulse input port 5. The light Q output through the optical output port 6 and the optical output port 3 is kept in the 0 state.

  Next, when the data light D changes to the 1 state, the Mach-Zehnder type optical switch 103 and the Mach-Zehnder type optical switch 104 start excitation of the nonlinear waveguide. At this time, the optical pulse is output from the optical output port 26 after the first clock optical pulse Clk input after the data light D is in the 1 state, and the optical pulse is not output from the optical output port 36. . That is, in the optical SR flip-flop 101, the set optical pulse S is input to the set optical pulse input port 4, and the light Q output via the optical output port 6 and the optical output port 3 transitions to the 1 state. , Retained.

  Next, when the data light D changes to the 0 state, the Mach-Zehnder optical switch 103 and the Mach-Zehnder optical switch 104 do not excite the nonlinear waveguide. Therefore, after the first clock light pulse Clk input after the data light D becomes 0, the light pulse is not output from the light output port 26, and the light pulse is output from the light output port 36. become. That is, in the optical SR flip-flop 101, the reset optical pulse R is input to the reset optical pulse input port 5, and the light Q output via the optical output port 6 and the optical output port 3 transitions to the 0 state. , Retained.

  As described above, the optical D flip-flop that controls the state of the output light corresponding to the state of the data light at the time when the clock light pulse is input is realized.

  The configuration of the optical flip-flop according to the third embodiment of the present invention is shown in FIG. This optical D flip-flop includes a data optical input port 1 for data light D, a clock optical pulse input port 2 for clock optical pulse Clk, and an optical output port 3 for output light Q, and is a Mach-Zehnder optical switch. 102, an optical SR flip-flop 101, and a Mach-Zehnder optical switch 105.

  In the Mach-Zehnder type optical switch 102, the nonlinear waveguides 11 and 12 are arranged in both arms, and the data light D is input to the optical input port 13 and the optical input port 15 via the data optical input port 1, The clock light pulse Clk is input to the light input port 14 via the clock pulse light input port 2, and output light (S, R) is output from the light output port 16 and the light output port 17.

  The optical SR flip-flop 101 includes a set optical pulse input port 4 for the set optical pulse S and a reset optical pulse input port 5 for the reset optical pulse R, which are connected to the optical output port 16 and the optical output port 17, respectively. The output light (Q ′) is output from the optical output port 6.

  In the Mach-Zehnder optical switch 105, nonlinear waveguides 41 and 42 are arranged on both arms, and a clock light pulse Clk ′ for exciting the nonlinear waveguide is input to the optical input port 43 and the optical input port 45. Then, the output light (Q ′) from the optical SR flip-flop is input to the optical input port 44, and the output light Q is output from the optical output port 46 and guided to the optical output port 3.

  The optical D flip-flop according to the third embodiment has a configuration in which an optical gate (105) driven by a clock light pulse Clk ′ is provided at the subsequent stage of the optical D flip-flop according to the first embodiment. In the first embodiment, when the set light pulse S input to the optical SR flip-flop 101 is continuously input, the output light is held in one state, but is output every time the set light pulse S is input. The light intensity may fluctuate. In addition, even when the reset light pulse R is continuously input, the output light intensity is fluctuated every time the reset light pulse R is input although the output light is held in the zero state. In the third embodiment, by using the gate operation in the Mach-Zehnder optical switch 105, the output light Q from which the fluctuation of the light intensity is removed can be obtained.

  In the Mach-Zehnder optical switch 102, the data light D excites the nonlinear optical effect in the nonlinear waveguides 11 and 12. On the other hand, the clock light pulse Clk is inputted to the optical input port 14 and then once branched, passed through the nonlinear waveguides 11 and 12 and given a nonlinear phase shift, and then multiplexed. As a result, output light (S, R) having an intensity corresponding to the phase difference of the light when combined and interfered is generated and output from the optical output port 16 and the optical output port 17.

  When the data light D is 0, excitation to the nonlinear waveguides 11 and 12 is not performed, so that the phase difference when the clock light pulse Clk passes through the nonlinear waveguides 11 and 12 and interferes is π. The output light intensity at the light output port 16 is low, and the output light intensity at the light output port 17 is high. On the other hand, when the data light D is 1, excitation to the nonlinear waveguides 11 and 12 is performed, so that the phase difference when the clock light pulse Clk passes through the nonlinear waveguides 11 and 12 and interferes is 0. In this state, the output light intensity of the light output port 16 is 1 state high, and the output light intensity of the light output port 17 is 0 state low.

  In the optical SR flip-flop 101, when the set light pulse S is input to the set light pulse input port 4, the light intensity of the output light Q ′ from the optical output port 6 transitions to one state, and thereafter, the reset light pulse R Until the light is input to the reset light pulse input port 5. Conversely, when the reset light pulse R is input to the reset light pulse input port 5, the light intensity of the output light Q ′ transitions to a low 0 state, and thereafter, the set light pulse S is input to the set light pulse input port 4. Until the output light state is maintained.

  In the Mach-Zehnder optical switch 105, the clock light pulse Clk ′ excites the nonlinear optical effect in the nonlinear waveguides 41 and 42. On the other hand, after the output light Q ′ from the optical output port 6 of the optical SR flip-flop 101 is input to the optical input port 44, it is once branched, passed through the nonlinear waveguide 41 or 42, and given a nonlinear phase shift. After that, it is combined. Then, output light Q having an intensity corresponding to the phase difference of the light when combined and interfered is output from the optical output port 46.

  Next, the operation of the optical D flip-flop according to the third embodiment will be described with reference to the waveform diagrams shown in FIGS. 5A to 5G show the data light D, the clock light pulse Clk, the set light pulse S, the reset light pulse R, the output Q ′ of the optical SR flip-flop 101, the clock light pulse Clk ′, and the output light Q. Each waveform is shown. Here, the data light D is an RZ modulated optical signal having a bit rate of 40 Gb / s, and the clock light pulse Clk is an optical pulse train having a repetition frequency of 40 GHz and a pulse width of 6 ps.

  When the data light D is in the 0 state, the nonlinear waveguide is not excited in the Mach-Zehnder optical switch 102. At this time, the clock light pulse Clk passes through the nonlinear waveguide 11 or 12 and is multiplexed. The phase difference at that time is π. For this reason, an optical pulse is not output from the optical output port 16, but an optical pulse is output from the optical output port 17. That is, in the optical SR flip-flop 101, the reset light pulse R is continuously input to the reset light pulse input port 5. The output light Q ′ from the optical output port 6 is maintained in the 0 state.

  Next, when the data light D changes to the 1 state, the Mach-Zehnder optical switch 102 starts excitation of the nonlinear waveguide. At this time, the phase difference when the clock light pulse Clk is combined through the nonlinear waveguide 11 or 12 becomes zero. For this reason, after the first clock light pulse input after the data light D is in the 1 state, the light pulse is output from the light output port 16 and the light pulse is not output from the light output port 17. That is, in the optical SR flip-flop 101, the set light pulse S is input to the set light pulse input port 4, and the output light Q 'from the optical output port 6 changes to the 1 state, and is held thereafter.

  Subsequently, when the data light D changes to the 0 state, the Mach-Zehnder optical switch 102 does not excite the nonlinear waveguide. At this time, the phase difference between the branched light that the clock light pulse Clk passes through the nonlinear waveguide 11 and the branched light that passes through the nonlinear waveguide 12 is π. For this reason, after the first clock light pulse Clk input after the data light D is in the 0 state, the light pulse is not output from the light output port 16 and the light pulse is output from the light output port 17. become. That is, in the optical SR flip-flop 101, the reset light pulse R is input to the reset light pulse input port 5, and the output light Q 'from the light output port 6 transitions to the 0 state and is held thereafter.

  Here, in the waveform of the light Q ′ output from the optical output port 6 of the optical SR flip-flop 101, there is a fluctuation in the light intensity while holding the 1 state, or a fluctuation in the light intensity while holding the 0 state. Has occurred. These light intensity fluctuations are not preferable because they can cause bit errors. The light intensity fluctuation during the 1 state holding in the output light Q ′ occurs when the set light pulse S input to the optical SR flip-flop 101 is continuous, and the 0 light state during the 0 light state holding in the output light Q ′. Light intensity fluctuation occurs when the reset light pulse R input to the optical SR flip-flop 101 is continuous. Here, the output light Q ′ is further input to the Mach-Zehnder optical switch 105. In the Mach-Zehnder type optical switch 105, when the clock light pulse Clk ′ is input, it is set so as to open a switching window of 6 ps with respect to the input light Q ′, and thereby the intensity in the waveform of Q ′. The prominent part of the fluctuation is removed. Accordingly, with the output light Q, output light from which the light intensity fluctuation is removed can be obtained.

  The configuration of the optical flip-flop according to the fourth embodiment of the present invention is shown in FIG. This optical D flip-flop includes a data optical input port 1 for data light D, a clock optical pulse input port 2 for clock optical pulse Clk, and an optical output port 3 for output light Q. A Mach-Zehnder optical switch 106 is provided. , Mach-Zehnder type optical switch 102 and optical SR flip-flop 101.

  In the Mach-Zehnder type optical switch 106, the nonlinear waveguides 51 and 52 are arranged on both arms, and the data light D is input to the optical input port 53 and the optical input port 55 via the data optical input port 1, The clock light pulse Clk is input to the light input port 54 via the clock light pulse input port 2, and the output light P is output from the light output port 56.

  In the Mach-Zehnder type optical switch 102, the nonlinear waveguides 11 and 12 are arranged in both arms, and the data light D is input to the optical input port 13 and the optical input port 15 via the data optical input port 1, The output light P of the Mach-Zehnder optical switch 106 is input from the optical output port 56 to the optical input port 14, and output light (S, R) is output from the optical output port 16 and the optical output port 17.

  The optical SR flip-flop 101 includes a set optical pulse input port 4 for the set optical pulse S and a reset optical pulse input port 5 for the reset optical pulse R, which are connected to the optical output port 16 and the optical output port 17, respectively. The output light Q is output from the optical output port 6 and guided to the optical output port 3.

  In the optical D flip-flop of the fourth embodiment, in the input light to the optical SR flip-flop 101, a state in which the set light pulse S is continuously input or a state in which the reset light pulse R is continuously input does not occur. Can be. That is, the input of the set light pulse S and the reset light pulse R can always be performed alternately. For this reason, the light intensity fluctuation does not occur in the output light of the optical SR flip-flop 101.

  In the Mach-Zehnder optical switch 106, the data light D excites the nonlinear optical effect in the nonlinear waveguides 51 and 52. On the other hand, the clock light pulse Clk is inputted to the optical input port 54, and then branched once. After passing through the nonlinear waveguides 51 and 52, a nonlinear phase shift is given, and then multiplexed. Then, output light P having an intensity corresponding to the phase difference of the light when combined and interfered is generated and output from the light output port 56.

  Here, with respect to the data light D1 input to the nonlinear waveguide 51, the data light D2 input to the nonlinear waveguide 52 is provided with a delay of 1 bit. When the data light D1 is 0 and the data light D2 is 0, excitation to the nonlinear waveguides 51 and 52 is not performed. Therefore, the phase difference when the clock light pulse Clk interferes through the nonlinear waveguide is π. Yes, the output light P from the optical output port 56 becomes zero.

  On the other hand, when one of the data light D1 and the data light D2 is 0 and the other is 1, either one of the nonlinear waveguides 51 and 52 is excited, so that the clock light pulse Clk passes through the nonlinear waveguide and interferes therewith. The phase difference at that time is 0, and the output light P from the light output port 56 is 1.

  When the data light D1 is 1 and the data light D2 is 1, both the nonlinear waveguides 51 and 52 are excited, so that the phase difference when the clock light pulse Clk interferes through the nonlinear waveguide is π. The output light P from the light output port 56 is zero.

  As described above, the state of the output light P is determined by the exclusive OR (XOR) between the preceding and succeeding bits in the data light D.

  In the Mach-Zehnder optical switch 102, the data light D excites the nonlinear optical effect in the nonlinear waveguides 11 and 12. On the other hand, the output light P of the Mach-Zehnder optical switch 106 is input to the optical input port 14, and then branched once, passed through the nonlinear waveguide 11 or 12, and given a nonlinear phase shift, and then multiplexed. Is done. Then, output light (S, R) having an intensity corresponding to the phase difference of the light when combined and interfered is generated and output from the optical output port 16 and the optical output port 17.

  When the data light D is 0, excitation to the nonlinear waveguides 11 and 12 is not performed, so that the phase difference when the optical signal P passes through the nonlinear waveguides 11 and 12 and interferes is π, The light output port 16 has a low output light intensity 0 state, and the light output port 17 has a high light output intensity 1 state. On the contrary, when the data light D is 1, excitation to the nonlinear waveguides 11 and 12 is performed, so that the phase difference when the optical signal P passes through the nonlinear waveguides 11 and 12 and interferes is 0. Yes, the output light intensity of the light output port 16 is 1 state, and the output light intensity of the light output port 17 is 0 state.

  In the optical SR flip-flop 101, when the set light pulse S is input to the set light pulse input port 4, the light intensity of the output light Q from the optical output port 6 transitions to one state. The output light state is maintained until it is input to the reset light pulse input port 5. Conversely, when the reset light pulse R is input to the reset light pulse input port 5, the light intensity of the output light Q transitions to a low 0 state, and thereafter the set light pulse S is input to the set light pulse input port 4. Until the output light state is maintained.

  Next, the operation of the optical D flip-flop according to the fourth embodiment will be described with reference to the waveform diagrams shown in FIGS. 7A to 7H show waveforms of the data light D, the data light D1, the data light D2, the clock light pulse Clk, the output light P of the optical switch, the set light pulse S, the reset light pulse R, and the output light Q. Respectively. Here, the data light D is an NRZ modulated optical signal having a bit rate of 40 Gb / s, and the clock light pulse Clk is an optical pulse train having a repetition frequency of 40 GHz and a pulse width of 6 ps.

  In the Mach-Zehnder type optical switch 106, since the state of the output light is determined by XOR between the preceding and succeeding bits of the data light D, the data light D changes from 0 to 1, or changes from 1 to 0. Sometimes the output light P becomes 1. In the output light P, the optical pulse generated when the data light changes from 0 to 1 is input at the same time as 1 of the data light D is input in the Mach-Zehnder optical switch 102 in the next stage. That is, since the nonlinear waveguides 11 and 12 are excited, the input optical signal P is output from the optical output port 16. This becomes a set light pulse S for the optical SR flip-flop 101, which is input to the set light pulse input port 4, and the light Q output via the light output port 6 and the light output port 3 changes to 1.

  In the output light P, an optical pulse generated when the data light changes from 1 to 0 is input at the same time as 0 of the data light D is input in the Mach-Zehnder optical switch 102 in the next stage. That is, since the nonlinear waveguides 11 and 12 are not excited, the input optical signal P is output from the optical output port 17. This becomes the reset light pulse R for the optical SR flip-flop 101, which is input to the reset light pulse input port 5, and the light Q output via the light output port 6 and the light output port 3 changes to zero.

  As described above, the optical D flip-flop that controls the state of the output light corresponding to the state of the data light at the time when the clock light pulse is input is realized.

  In the configuration of the optical flip-flop of the fourth embodiment, an optical waveform shaping circuit that operates without preparing a clock light pulse is realized by inputting CW light instead of a clock light pulse to the input port 2. This is the fifth embodiment, and its configuration is shown in FIG.

  This optical waveform shaping circuit includes a data light input port 1 for data light D, a continuous light input port 2 for CW light, and an optical output port 3 for output light Q, and a Mach-Zehnder type optical switch 106, Mach A zender type optical switch 102 and an optical SR flip-flop 101 are included.

  In the Mach-Zehnder type optical switch 106, the nonlinear waveguides 51 and 52 are arranged on both arms, and the data light D is input to the optical input port 53 and the optical input port 55 via the data optical input port 1, The CW light is input to the light input port 54 via the continuous light input port 2, and the output light P is output from the light output port 56.

  In the Mach-Zehnder type optical switch 102, the nonlinear waveguides 11 and 12 are arranged in both arms, and the data light D is input to the optical input port 13 and the optical input port 15 via the port 1, and the Mach-Zehnder The output light P of the optical switch 106 is input from the optical output port 56 to the optical input port 14, and output light (S, R) is output from the optical output port 16 and the optical output port 17.

  The optical SR flip-flop 101 includes a set optical pulse input port 4 for the set optical pulse S and a reset optical pulse input port 5 for the reset optical pulse R, which are connected to the optical output port 16 and the optical output port 17, respectively. The output light Q is output from the optical output port 6 and guided to the optical output port 3.

  Similar to the optical D flip-flop of the fourth embodiment, also in this optical waveform shaping circuit, in the input light to the optical SR flip-flop 101, the set light pulse S is continuously input, or the reset light pulse R is It is possible to prevent the state of being continuously input. That is, the input of the set light pulse S and the reset light pulse R can always be performed alternately. For this reason, the light intensity fluctuation does not occur in the output light of the optical SR flip-flop 101.

  In the Mach-Zehnder optical switch 106, the data light D excites the nonlinear optical effect in the nonlinear waveguides 51 and 52. On the other hand, the CW light input to the optical input port 54 is once branched, passed through the nonlinear waveguides 51 and 52, and given a nonlinear phase shift, and then multiplexed. Then, output light P having an intensity corresponding to the phase difference of the light when combined and interfered is generated and output from the light output port 56.

  The data light D2 input to the nonlinear waveguide 52 is provided with a delay of 1 bit with respect to the data light D1 input to the nonlinear waveguide 51. When the data light D1 is 0 and the data light D2 is 0, excitation to the nonlinear waveguides 51 and 52 is not performed, so that the phase difference when the CW light passes through the nonlinear waveguide and interferes is π, The output light P from the optical output port 56 is zero.

  In addition, when one of the data light D1 and the data light D2 is 0 and the other is 1, either one of the nonlinear waveguides 51 and 52 is excited, so that when the CW light passes through the nonlinear waveguide and interferes therewith. The phase difference is 0, and the output light P from the light output port 56 is 1.

  Further, when the data light D1 is 1 and the data light D2 is 1, both the nonlinear waveguides 51 and 52 are excited, so the phase difference when the CW light passes through the nonlinear waveguide and interferes is π. The output light P from the light output port 56 becomes zero.

  As described above, the state of the output light P is determined by the exclusive OR (XOR) between the preceding and succeeding bits in the data light D.

  In the Mach-Zehnder optical switch 102, the data light D excites the nonlinear optical effect in the nonlinear waveguides 11 and 12. On the other hand, after the output light P of the Mach-Zehnder optical switch 106 is input to the optical input port 14, it is once branched, passed through the nonlinear waveguides 11 and 12, and given a nonlinear phase shift, and then multiplexed. Is done. Then, output light (S, R) having an intensity corresponding to the phase difference of the light when combined and interfered is generated and output from the optical output port 16 and the optical output port 17. When the data light D is 0, excitation to the nonlinear waveguides 11 and 12 is not performed, so that the phase difference when the optical signal P interferes through the nonlinear waveguide is π, and the optical output port 16 is a 0 state where the output light intensity is low, and a 1 state where the output light intensity of the light output port 17 is high. On the contrary, when the data light D is 1, excitation to the nonlinear waveguides 11 and 12 is performed, so that the phase difference when the optical signal P passes through the nonlinear waveguide and interferes is 0. The output port 16 has a high output light intensity 1 state, and the light output port 17 has a low output light intensity 0 state.

  In the optical SR flip-flop 101, when the set light pulse S is input to the set light pulse input port 4, the light intensity of the output light Q from the optical output port 6 transitions to one state. The output light state is maintained until it is input to the reset light pulse input port 5. Conversely, when the reset light pulse R is input to the reset light pulse input port 5, the light intensity of the output light Q transitions to a low 0 state, and thereafter the set light pulse S is input to the set light pulse input port 4. Until the output light state is maintained.

  Next, the operation of this optical waveform shaping circuit will be described with reference to the waveform diagrams shown in FIGS. 9A to 9H show waveforms of the data light D, the data light D1, the data light D2, the CW light, the output light P of the optical switch, the set light pulse S, the reset light pulse R, and the output light Q, respectively. Show. Here, the data light D is an NRZ modulated optical signal having a bit rate of 40 Gb / s.

  In the Mach-Zehnder type optical switch 106, since the state of the output light is determined by XOR between the preceding and succeeding bits of the data light D, the data light D changes from 0 to 1, or changes from 1 to 0. Sometimes the output light P becomes 1. In the output light P, the optical pulse generated when the data light changes from 0 to 1 is input at the same time as 1 of the data light D is input in the Mach-Zehnder optical switch 102 in the next stage. That is, since the nonlinear waveguides 11 and 12 are excited, the input optical signal P is output from the optical output port 16. This becomes a set light pulse S for the optical SR flip-flop 101, which is input to the set light pulse input port 4, and the light Q output via the light output port 6 and the light output port 3 changes to 1.

  In the output light P, an optical pulse generated when the data light changes from 1 to 0 is input at the same time as 0 of the data light D is input in the Mach-Zehnder optical switch 102 in the next stage. That is, since the nonlinear waveguides 11 and 12 are not excited, the input optical signal P is output from the optical output port 17. This becomes the reset light pulse R for the optical SR flip-flop 101, which is input to the reset light pulse input port 5, and the light Q output via the light output port 6 and the light output port 3 changes to zero. Thus, an optical waveform shaping circuit is realized.

  A semiconductor optical amplifier (SOA) can be used as the nonlinear waveguide of the optical circuit in each embodiment. In the first embodiment, an operation when using an SOA having a gain band in the wavelength 1550 nm band as the nonlinear waveguide will be described.

  The SOA is used as the nonlinear waveguides 11 and 12 in the Mach-Zehnder type optical switch 102 and the nonlinear waveguide in the optical SR flip-flop 101. The data light D is an NRZ modulated optical signal having a wavelength of 1555 nm and a bit rate of 40 Gb / s, and the clock light pulse Clk is an optical pulse train having a wavelength of 1545 nm, a repetition frequency of 40 GHz, and a pulse width of 6 ps.

  In the nonlinear waveguides 11 and 12, when the data light D is input, the carrier density decreases, and a nonlinear phase shift with respect to the clock light pulse Clk occurs. An optical pulse with a wavelength of 1545 nm output from the optical output port 16 of the Mach-Zehnder optical switch 102 is input as a set optical pulse S to the set optical pulse input port 4 of the optical SR flip-flop 101, and the Mach-Zehnder optical switch An optical pulse with a wavelength of 1545 nm output from the optical output port 17 of 102 is input to the reset optical pulse input port 5 of the optical SR flip-flop 101 as the reset optical pulse R. In the optical SR flip-flop 101, the state of the output light Q having a wavelength of 1555 nm is controlled by the set light pulse S and the reset light pulse R.

  Or the waveguide which comprised the active layer by the quantum dot (QD) can be used as a nonlinear waveguide of the optical circuit in each embodiment. In the first embodiment, an operation when a QD waveguide having an absorption band in the wavelength 1300 nm band is used as the nonlinear waveguide will be described.

  The QD waveguide is used as the nonlinear waveguides 11 and 12 in the Mach-Zehnder optical switch 102 and the nonlinear waveguide in the optical SR flip-flop 101. The data light D is an NRZ modulated optical signal having a wavelength of 1300 nm and a bit rate of 40 Gb / s, and the clock light pulse Clk is an optical pulse train having a wavelength of 1280 nm, a repetition frequency of 40 GHz, and a pulse width of 6 ps.

  In the nonlinear waveguides 11 and 12, when the data light D is input, it is absorbed by the QD, carriers are generated in the QD, and a nonlinear phase shift with respect to the clock light pulse Clk occurs. An optical pulse with a wavelength of 1280 nm output from the optical output port 16 of the Mach-Zehnder optical switch 102 is input as a set optical pulse S to the set optical pulse input port 4 of the optical SR flip-flop 101, and the Mach-Zehnder optical switch An optical pulse having a wavelength of 1280 nm output from the optical output port 17 of 102 is input to the reset optical pulse input port 5 of the optical SR flip-flop 101 as the reset optical pulse R. In the optical SR flip-flop 101, the state of the output light Q having a wavelength of 1300 nm is controlled by the set light pulse S and the reset light pulse R.

  In addition, by using a photonic crystal waveguide or a high refractive index difference waveguide as the optical waveguide constituting the optical circuit in each embodiment, the radius of curvature of the bent waveguide can be reduced, and the device can be miniaturized. It becomes. Here, the case where a photonic crystal waveguide is used will be described. First, the basic structure will be described.

A GaAs layer (thickness: 0.25 μm) is grown on a GaAs substrate via an Al _0.15 Ga _0.85 As layer (thickness: 2 μm) as a sacrificial layer, and periodic in the GaAs surface by electron beam lithography and dry etching. By forming a circular hole array structure, a two-dimensional photonic crystal (hereinafter referred to as 2DPC) is obtained. For confinement of light in the thickness direction, the sacrificial layer is selectively removed (wet etching) to leave only the core layer, thereby obtaining an air bridge type planar optical waveguide having a large refractive index contrast with the air layer. .

  In order to form a desired optical wiring on the optical waveguide substrate, an array in which the circular holes are missing, that is, a so-called defect waveguide is formed along the optical wiring in the periodic circular hole array. In designing the defect waveguide, the natural propagation mode of the waveguide is set in the band gap of the 2DPC. In this way, light is prohibited from leaking to the periphery of the surface, and propagates along the waveguide without loss to the surface.

The lattice constant (a) of 2DPC in the optical communication wavelength band is 0.3 to 0.4 μm, and most devices fit within the optical path length range of 10a to 1000a, so 2DPC exceeds several μ to several 100 μm. A compact optical device is provided. In addition, since the propagation loss due to the imperfection of the structure generated in the manufacturing process is sufficiently small as 0.8 dB / mm or less, a low propagation loss optical device of 4 dB or less can be constructed even with an element size of ˜5 mm ² , for example. is there.

  Such a basic structure of 2DPC can also be used for the portion of the nonlinear waveguide. In this case, quantum dots (hereinafter referred to as QD) made of InAs are embedded in the nonlinear waveguide portion in advance. QD shows a sharp absorption spectrum peak at an arbitrary wavelength around 1300 to 1550 nm. For this reason, the nonlinear waveguide made of QD induces a nonlinear refractive index change by absorbing the excitation light and generates a desired nonlinear phase shift. Therefore, in the optical flip-flop composed of 2DPC, the operation principle described in each embodiment is applicable.

  As described in the specification of Japanese Patent Application No. 2006-224165, in the optical SR flip-flop composed of 2DPC, the total optical path length of the feedback loop is 100 μm, and the interval between the set light pulse and the reset light pulse is reduced to several ps. It becomes possible to do. This means that the operation bit rate of the optical flip-flop can be increased to several hundred Gb / s. In other words, when 2DPC is used, an ultra-small all-optical switch is realized, so that an optical flip-flop operating at an ultra-high speed can be realized.

  In addition, in the nonlinear waveguide composing the optical flip-flop, only the case where the light used for pumping and the light subjected to the nonlinear phase shift propagate in the same direction has been described, but the same operation occurs when propagating in the opposite direction. Is obtained.

1 is a configuration diagram of an optical flip-flop according to a first embodiment of the present invention. FIG. 2 is a waveform diagram for explaining the operation of the optical flip-flop of FIG. 1, where (a) is a data light D, (b) is a clock light pulse Clk, (c) is a set light pulse S, and (d) is a reset. Optical pulses R and (e) are waveform diagrams of the output light Q. It is a block diagram of the optical flip-flop which concerns on 2nd embodiment of this invention. It is a block diagram of the optical flip-flop which concerns on 3rd embodiment of this invention. FIG. 5 is a waveform diagram for explaining the operation of the optical flip-flop of FIG. 4, where (a) is a data light D, (b) is a clock light pulse Clk, (c) is a set light pulse S, and (d) is a reset. The optical pulses R, (e) are the output Q ′ of the optical SR flip-flop, (f) is the clock optical pulse Clk ′, and (g) is the waveform diagram of the output light Q. It is a block diagram of the optical flip-flop which concerns on 4th embodiment of this invention. FIG. 7 is a waveform diagram for explaining the operation of the optical flip-flop of FIG. 6, where (a) is data light D, (b) is data light D1, (c) is data light D2, and (d) is a clock light pulse. Clk, (e) is the output light P of the optical switch, (f) is the set light pulse S, (g) is the reset light pulse R, and (h) is the waveform diagram of the output light Q. It is a block diagram of the optical flip-flop which concerns on 5th embodiment of this invention. FIG. 9 is a waveform diagram for explaining the operation of the optical flip-flop of FIG. 8, where (a) is data light D, (b) is data light D1, (c) is data light D2, (d) is CW light, (E) is the output light P of the optical switch, (f) is the set light pulse S, (g) is the waveform diagram of the reset light pulse R and (h) the output light Q. (A) is a circuit diagram of the SR flip-flop, and (b) is a characteristic table for explaining its operation. It is a schematic block diagram of the optical SR flip-flop proposed by the present applicant. FIG. 12 is a waveform diagram for explaining the operation of the optical SR flip-flop of FIG. 11, where (a) is a set light pulse S, (b) is a reset light pulse R, (c) is an output light A3 of the optical switch, ( d) is a waveform diagram of the output light Q of the optical SR flip-flop. (A) is a circuit diagram of a D flip-flop, and (b) is a characteristic table for explaining its operation.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Data optical input port 2 Clock optical pulse or continuous optical input port 3 Optical output port 4 Set optical pulse input port 5 Reset optical pulse input port 6 Optical output port 11 Nonlinear waveguide 12 Nonlinear waveguide 13 Optical input port 14 Optical input port DESCRIPTION OF SYMBOLS 15 Optical input port 16 Optical output port 17 Optical output port 21 Nonlinear waveguide 22 Nonlinear waveguide 23 Optical input port 24 Optical input port 25 Optical input port 26 Optical output port 31 Nonlinear waveguide 32 Nonlinear waveguide 33 Optical input port 34 Light Input port 35 Optical input port 36 Optical output port 41 Non-linear waveguide 42 Non-linear waveguide 43 Optical input port 44 Optical input port 45 Optical input port 46 Optical output port 51 Non-linear waveguide 52 Non-linear waveguide 53 Optical input port 54 Optical input port 55 Light input port G 56 Optical output port 61 Non-linear waveguide 62 Non-linear waveguide 63 Set optical pulse input port 64 Continuous optical input port 65 Reset optical pulse input port 66 Optical output port 67 Optical branching section 68 Optical output port 71 Non-linear waveguide 72 Non-linear waveguide 73 Optical input port 74 Optical input port 75 Optical input port 76 Optical output port 77 Optical branching unit 78 Optical multiplexing unit 101 Optical set reset flip-flop 102 Mach-Zehnder optical switch 103 Mach-Zehnder optical switch 104 Mach-Zender optical Switch 105 Mach-Zehnder optical switch 106 Mach-Zehnder optical switch 111 Mach-Zehnder optical switch 112 Mach-Zehnder optical switch

Claims (9)

An optical set / reset flip-flop having a set optical input port and a reset optical input port;
Switching that receives binary intensity modulated data light and clock light or CW light, and distributes part or all of the clock light or CW light to the set light input port and the reset light input port based on the data light Means and
A delayed data light substantially equal to the delayed data light is output to the optical output port of the optical set / reset flip-flop or the output port of the optical circuit connected to the optical output port. Features an optical flip-flop. The optical flip-flop according to claim 1,
The optical flip-flop characterized in that the switching means includes at least one Mach-Zehnder type optical switch. The optical flip-flop according to claim 1,
The switching means comprises a single Mach-Zehnder optical switch;
The Mach-Zehnder type switch divides the clock light, passes through a pair of arms and then multiplexes and interfers with a Mach-Zehnder type optical circuit, and first and second disposed on the pair of arms. A nonlinear waveguide; means for inputting the data light into the first and second nonlinear waveguides to excite a nonlinear optical effect; and nonlinear optical effects of the first and second nonlinear waveguides. An optical flip-flop comprising: means for guiding clock light combined and interfered by the Mach-Zehnder type optical circuit according to an excitation state to the set light input port or the reset light input port. The optical flip-flop according to claim 1,
The switching means includes a pair of Mach-Zehnder type optical switches,
The Mach-Zehnder type switch divides the clock light, passes through a pair of arms and then multiplexes and interfers with a Mach-Zehnder type optical circuit, and first and second disposed on the pair of arms. A nonlinear waveguide, and means for inputting the data light to the first and second nonlinear waveguides to excite a nonlinear optical effect,
The optical path length of the Mach-Zehnder optical circuit is determined so that the output lights of the pair of Mach-Zehnder optical switches are logically inverted from each other, and the clock light combined and interfered by one Mach-Zehnder optical switch is An optical flip-flop characterized in that the clock light combined and interfered by the other Mach-Zehnder type optical switch to the set light input port is guided to the reset light input port, respectively. The optical flip-flop according to claim 3 or 4,
As an optical circuit connected to the optical output port of the optical set reset flip-flop, including another optical gate,
The optical gate receives the output light of the optical set reset flip-flop and the second clock light, and transmits the output light of the optical set reset flip-flop based on the second clock light. And optical flip-flop. The optical flip-flop according to claim 3 or 4,
As an optical circuit connected to the optical output port of the optical set reset flip-flop, including another Mach-Zehnder type optical switch,
The Mach-Zehnder type switch is arranged on the pair of arms and the Mach-Zehnder type optical circuit for branching the output light of the optical set-reset flip-flop, passing through the pair of arms, and causing interference after multiplexing. First and second nonlinear waveguides, means for inputting other clock light into the first and second nonlinear waveguides to excite the nonlinear optical effect, and the output of the Mach-Zehnder optical circuit An optical flip-flop comprising means for guiding light to the outside. The optical flip-flop according to claim 1,
The switching means guides a single pulse of the clock light to the set light input port when 0 and 1 appear in a continuous bit string at the input of the data light, and 1 to a continuous bit string at the input of the data light. , And means for guiding a single pulse of the clock light to the reset light input port when 0 appears. The optical flip-flop according to claim 1,
The switching means includes first and second Mach-Zehnder optical switches;
The first Mach-Zehnder type switch is disposed on the pair of arms, and the first Mach-Zehnder type optical circuit for branching the clock light, passing the pair of arms, and causing interference after multiplexing. Means for inputting the data light into the first and second nonlinear waveguides with a difference of 1 bit from each other in order to excite a nonlinear optical effect; and With
The second Mach-Zehnder optical switch splits the output light of the first Mach-Zehnder optical circuit, passes through a pair of arms, and then interferes with the second Mach-Zehnder optical circuit. And third and fourth nonlinear waveguides disposed on the pair of arms, and means for inputting the data light to the third and fourth nonlinear waveguides to excite a nonlinear optical effect; Means for guiding the output light of the second Mach-Zehnder optical circuit to the set light input port or the reset light input port according to the excited state of the nonlinear optical effect of the third and fourth nonlinear waveguides; An optical flip-flop comprising: The optical flip-flop according to claim 8.
An optical flip-flop characterized in that CW light is input to the first Mach-Zehnder switch instead of the clock light.

Images