JPH012027A - light flip flop - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention is applied to optical logic circuit elements. In particular, the present invention relates to an optical flip-flop whose optical output state can be set and reset.

[Conventional technology]

Optical circuits can operate at higher speeds than electronic circuits, and have advantages such as excellent compatibility with transmission lines as signal processing circuits used for optical transmission. Optical flip-flops are a basic element in constructing optical logic circuits.

FIG. 6 shows the structure of a first conventional optical flip-flop, and FIG. 7 shows its characteristics.

This optical flip-flop has a structure in which the electrode of a normal semiconductor laser 41 is divided into two electrodes 42 and 43, and the hysteresis characteristics that occur between the optical input and the optical output and between the bias current and the optical output are suppressed. It was used.

Bias currents L and Iz are supplied to electrodes 42 and 43 of the semiconductor laser 41, respectively. Here, It +12 =
Let it be Ib. In this state, output light P0 cannot be obtained.

When input light P1 is supplied from the outside in this state, the optical output changes to the "on" state at a certain light intensity, and output light P0 is obtained. Even if the supply of input light P1 is stopped in this state, this optical flip-flop maintains the "on" state. Next, when the current injected into the semiconductor laser 1 is decreased, the optical output transitions to an "off" state at a certain current value, and remains "off" even when the bias current Ib is returned to the original state.

Using this relationship, set (output light P0
(changes from "off" to "on") and can be reset by a current pulse (output light P0 changes from "on" to "off"). A current pulse is obtained, for example, by stopping the supply of bias current to one of the electrodes 42 or 43.

Details of this first conventional example are described in Tomita et al., ``Experiments on Ultra-High Speed Optical Memory Operation in Bistable LD'', Lecture No. 904 at the 1985 Institute of Electrical Communication Engineers National Conference.

FIG. 8 shows a circuit diagram of a second conventional optical flip-flop.

Power source 61 is connected to the anode electrode of photodiode 63. A cathode electrode of the photodiode 63 is connected to one end of the resistor R2. Power source 62 is connected to the photodiode 640 cathode electrode. The anode electrode of the photodiode 63 is connected to one end of the resistor R3.

The other end of the resistor R2 and the other end of the resistor R3 are connected to each other, and this connection point is further grounded via the resistor R1 and connected to the cathode electrode of the polarized bistable semiconductor laser 67 via the amplifier 65. The cathode electrode of polarized bistable semiconductor laser 67 is also connected to power supply 66 . The anode electrode of the polarized bistable semiconductor laser 67 is grounded via a resistor R4. - An analyzer 68 is placed in the optical path of the output light P0 of the polarized bistable semiconductor laser 67.

FIG. 9 shows the characteristics of the optical output with respect to the injection current of the polarized bistable semiconductor laser 67. In this figure, the solid line indicates the output component of TM polarization, and the broken line indicates the output component of TE polarization. When the injected current is gradually increased, the output light is a current! T at 8
The E polarization component suddenly increases and the TM polarization component suddenly decreases.

When the injected current is gradually decreased, the current! , the TM polarization component suddenly increases. Therefore, when either the TE or TM polarized component is extracted by the analyzer 68, the optical output has hysteresis characteristics with respect to the injected current.

A conventional optical flip-flop using this characteristic is shown in FIG.

In this conventional example, when no light is injected into the photodiodes 63 and 64, a bias current Lb is injected into the polarized bistable semiconductor laser 67 by the power source 66. When the input light Pit is input to the photodiode 63, the polarization
When the current injected into the bistable semiconductor laser 67 increases and the input light P12 is input to the photodiode 64, the current injected into the polarized bistable semiconductor laser 67 decreases.

Therefore, for example, when the analyzer 68 transmits the output component of the TE polarized wave, when an optical pulse is input to the photodiode 63, the optical output of the polarized bistable semiconductor laser 67 is set to "on", and the optical output component is transmitted to the photodiode 64. Inputting a pulse resets it to "off". That is, it operates as an optical flip-flop.

For details of this second prior art example, see Jiming Liu, Digital Optical Signal Processing with Polarizing Bistable Semiconductor Lasers J, IεεEεε-Nal.
Off-Claontum Electronics Volume QB-21, Page 298 (JAI-MINGLlυ, rDigita
l 0ptical Signal Process
ssingwith Po1arization-B
istable Sem1conductorLas
er J, 18εεJ+Quantum Elect
ronlVOl, QB-21.

pp 298).

[Problem that the invention seeks to solve]

However, in the first conventional example, an electrical signal is used to reset the output, so it has a drawback that it is not suitable for constructing a logic circuit that operates entirely using light.

In addition, in the second conventional example, setting and resetting can be performed using optical pulses, but because the optical pulses are converted into electrical signals, the configuration is complicated and the speed of response is limited. there were.

In order to solve these problems, the applicant of the present application, in Japanese Patent Application No. 61-205316 and Japanese Patent Application No. 61-295457 which takes this application as the basis of priority, used a semiconductor laser to generate light without using electrical signals. We proposed a high-speed optical flip-flop that can be set and reset by pulses.

An object of the present invention is to provide a high-speed optical flip-flop that uses a Fabry-Perot resonator and can be set and reset by optical pulses without using electrical signals.

[Means for solving problems]

The optical flip-flop of the present invention includes a Fabry-Perot resonator having a structure in which a nonlinear medium whose refractive index changes depending on the light intensity is inserted between two opposing half mirrors; a first light injection means for steadily injecting light at a first frequency (f) that produces a hysteresis relationship between the input light intensity and the output light intensity of the resonator; a second optical injection means for injecting optical pulses having the same phase and frequency into the Fabry-Perot resonator at an input level such that the output optical intensity of the Fabry-Perot resonator is at a high level; and a third light injection means for injecting a light pulse having a frequency (f2) into the Fabry-Perot resonator at an input level such that the output light intensity of the Fabry-Perot resonator is at a low level.

A semiconductor optical amplification element to which a partial bias current slightly smaller than the oscillation threshold is supplied can also be used as the Fabry-Perot resonator.

[For production]

The present invention uses a Fabry-Perot resonator having a structure in which a nonlinear medium whose refractive index changes depending on the light intensity is inserted between two opposing half mirrors as an optical bistable element. In a Fabry-Perot resonator having such a structure, a hysteresis relationship occurs between the input light intensity and the output light intensity at a certain frequency. That is, as the input light intensity is increased, the output light intensity increases rapidly at a certain intensity. When the input light intensity is subsequently decreased, the output light intensity suddenly decreases at an intensity different from that at which the sudden increase occurred. When the input light intensity is increased again, the output light intensity increases again with the same intensity as before.

In the present invention, this phenomenon is utilized for transitioning an optical flip-flop from an off state to an on state.

Furthermore, when two lights of different frequencies are mixed and input into a Fabry-Perot resonator, the point at which the output light intensity changes rapidly with respect to a change in the input light intensity changes periodically. Therefore, by appropriately selecting the frequencies and intensities of the two lights, the output light intensity can be rapidly reduced.
It is possible to prevent the value from increasing while remaining decreased.

The present invention utilizes this phenomenon for transitioning an optical flip-flop from an on state to an off state.

The present invention uses a Fabry-Perot resonator as an optical bistable device, and this point is based on Japanese Patent Application No. 61-205 in which a semiconductor laser, that is, an oscillator, is used as an optical bistable device.
No. 316, which is different from the optical flip-flop disclosed in Japanese Patent Application No. 61-295457. In other words, in the case of a semiconductor laser, setting and resetting are performed using a frequency different from its free oscillation frequency, but in the case of a Fabry-Perot resonator,
The frequency of the output light becomes equal to the frequency of at least a portion of the input light.

A semiconductor optical amplification element can also be used as the Fabry-Perot resonator. A semiconductor optical amplification device is structurally equivalent to a semiconductor laser device, but differs from a semiconductor laser in that its bias current is set slightly smaller than the oscillation threshold and it oscillates in synchronization with input light.

〔Example〕

FIG. 1 is a block diagram of an optical flip-flop according to an embodiment of the present invention.

This optical flip-flop has a Fabry-Perot resonator 1 having a structure in which a nonlinear medium 13 whose refractive index changes depending on the light intensity is inserted between two opposing half mirrors 11 and 12.
and a first input light In which is steadily injected into this Fabry-Perot resonator 1 and has a first frequency f that causes a hysteresis relationship between the input light intensity and the output light intensity of this Fabry-Perot resonator 10. a light injection means, the light pulse Is having the same phase and frequency as the light of the first frequency;
a second light injection means for injecting into the Fabry-Perot resonator 1 at an input level such that the output light intensity of the Fabry-Perot resonator 1 is at a high level; and a light pulse I having a second frequency f2 different from the first frequency. , the above Fabry-Perot resonator l
and a third light injection means for injecting light into the Fabry-Perot resonator 1 at an input level such that the output light intensity of the light is at a low level.

The first to third light injection means are two half mirrors 2,
It is composed of 3. These half mirrors 2, 3
The half mirror 12 combines three optical paths into one.
The light enters the Fabry-Perot resonator 1 from the side.

In FIG. 1, half mirrors 11 and 12 and a nonlinear medium 13 are shown.
A space is shown between. Although such a space may be provided, the nonlinear medium 13 is
It is also possible to have a structure that fills the gap between 12 and 12.

The intensity of light output from the half mirror 12 side when light is incident on the Fabry-Perot resonator 1 from the half mirror 11 side will be explained. In the following description, a distance coordinate 2 is defined in the axial direction of the Fabry-Perot resonator 1, and the positions of the half mirrors 11 and 12 are assumed to be z=Q and z=L, respectively.

L is the resonator length.

Further, the direction from the half mirror 11 to the half mirror 12 is defined as a positive direction, and the opposite direction is defined as a negative direction.

Inside the Fabry-Perot resonator 1, light traveling in the positive direction and light traveling in the negative direction exist due to the opposing /%-fum mirrors 11 and 12. The optical electric field at each point in the Fabry-Perot cavity l is defined as light traveling in the positive direction at z=Q: E”(0)z=
Light traveling in the negative direction at Q: E-(0) Light traveling in the positive direction at Z=L: E''(L) Light traveling in the negative direction at z=L: E-(L) Incident light: E
ih ``eXp (: 1 (2yr f t + θ))] Output light * Eout is determined. Between each optical electric field, there is a relationship of (E, ut = t-E'' ([,).

Here, ′t: amplitude transmittance of verge mirror II, 12, r
: Amplitude reflectance of half mirror IL 12, φ :
This is the phase change that occurs while the light travels between the half mirrors IL 12 by L. Between the amplitude transmittance t and the amplitude reflectance r, t2
There is a relationship of +r2 =1. From these relational expressions, XEtn-eXI' f:i (2πf -φ10)
...---...-1 (1) is obtained. However, R is the intensity transmittance of the half mirrors 11 and 12, and there is a relationship of R=r''.

By the way, the phase change φ can be calculated using the refractive index n of the medium between the half mirrors 11 and 120 and the propagation constant k of light.
It is expressed as -L. A nonlinear medium 13 whose refractive index changes depending on the light intensity is inserted between the half mirrors 11 and 12, and since the light intensity is expressed as the square of the absolute value of the amplitude, n = nO + Kt ・I Eout I '. can be expressed. Here, no is a refractive index that does not depend on light intensity, and K1 is a proportionality constant.

Furthermore, the optical electric field was approximated as light E″(L) traveling in the positive direction at z=L, that is, E, .1/1.

This approximation is called the mean field approximation and is commonly used as a reasonable approximation. Therefore, the phase change φ is φ=(no +, ”l Eout 12)
・2 to k −L−δ+ ・l E, ut l 2−
- expressed as (2). Here, Kl and 2 are proportionality constants. By solving this equation and equation (1) simultaneously, the human output relationship of this Fabry-Perot resonator 1 can be determined.

FIG. 2 shows an example of a calculation result of the input/output relationship of the Fabry-Perot resonator 10 obtained by solving equations [1) and (2) simultaneously. In this figure, the horizontal axis shows the normalized input light intensity P, and the vertical axis shows the normalized output light intensity P. Indicates ut. Here, Pin = Kl・
lEt-l''P,ut=Kl·IEoutI''. Further, δ = 2.5 and R = 0.7. When the input light intensity is increased from zero, the output light intensity increases rapidly at a certain light intensity P up.

Further, when the input light intensity is decreased from this state, the output light intensity sharply decreases when the light intensity P down becomes weaker than the light intensity p up. When the input light intensity is at a level between the light intensity p up and the light intensity p do □, three solutions exist as the output light intensity for the input light intensity. Among these, a solution having an intermediate value is an unstable solution, and both the high output state and the low output state are stable states. By using these two states as an on state and an off state, respectively, the Fabry-Perot resonator 1 can be operated as an optical bistable element.

Semiconductors such as indium antimony InSb and gallium arsenide GaAs, and dielectrics such as carbon disulfide C82 and KTN can be used as the nonlinear medium exhibiting the human output characteristics shown in FIG. The nonlinearity of these semiconductors and dielectrics has also been confirmed experimentally.

Next, the operation of the optical flip-flop using this Fabry-Perot resonator 1 will be explained.

Here, the frequencies of the three lights input to the Fabry-Perot resonator 1, that is, the input light LH% set optical pulse I and the reset optical pulse ■, which are constantly input, are as follows. The following conditions shall be met. First, the frequency f1 of the input light IH and the optical pulse Is is set to a frequency at which the Fabry-Perot resonator 1 exhibits bistable characteristics with respect to the input light, for example, the characteristics shown in FIG. 2.

The frequency f2 of the optical pulse In is set so that the reciprocal of the frequency difference Δf=f, −f2 with respect to the frequency f1 is smaller than the response speed of the nonlinear medium 13. That is,
It is assumed that the refractive index of the nonlinear medium 13 sufficiently responds to frequency fluctuations of the frequency difference Δf.

Optical flip-flop operation can be obtained by inputting input light I, light pulse I, and light pulse IR satisfying such frequency conditions to the Fabry-Perot resonator 1 at appropriate input levels.

In order to obtain optical flip-flop operation, the Fabry-Perot resonator 1 is given an optical intensity P such that P dowg < P o < P up.
. Input light III is constantly input.

This input light I) is a kind of bias light for maintaining the system, and this input light 1. On the other hand, the Fabry-Perot resonator 1 takes either an on state or an off state.

First, it is assumed that the Fabry-Perot resonator 1 is in an off state. At this time, in the Fabry-Perot resonator 1, Pa +P,
> Pu.

When a light pulse I, with a light intensity P, is superimposed and inputted to the light input (2), the Fabry-Perot resonator 1 transitions to the on state, and remains on even if the light pulse Is is erased.

That is, the Fabry-Perot resonator 1 is set by the optical pulse Is of frequency fl and optical intensity P, and transitions from the off state to the on state.

Next, when an optical pulse It of frequency f2 is input to the Fabry-Perot resonator 1 which is in the on state, the Fabry-Perot resonator 1 is reset and switched from the on state to the off state. This operating principle can be obtained by considering the input/output characteristics of the Fabry-Perot resonator 1 when inputting three wavelengths. This principle will be explained below.

When inputting three wavelengths, the input light is converted into an input optical electric field E with a frequency f1.
, in+1 and the input optical electric field E i 11+ of frequency f2
Let's consider it in two parts. The input/output relationship for each light of frequency f, , f2 is as follows:
+θ2)]. Here, the notation of "IR" lφ, θ, and δ is the same as that used in the right side of equation (1) and equation (2), and the subscripts "1" and "2" indicate the values for frequencies f2 and f2, respectively. . Further, it is assumed that the refractive index of the nonlinear medium 13 sufficiently responds to the frequency difference Δf=f, -f between the frequencies f1 and f2. That is, it is assumed that the frequency difference Δf is smaller than the reciprocal of the response time of the nonlinear medium 13. Under this condition, it can be considered that φ1=φ2=φδ,=62=δ φ1NL=φ2ML=φゞ1. Specifically, the phase change term is φ1=n-L・2π・f1/C φ2=n-L・2π・f2/C (C is the speed of light), and the frequency difference Δf is Considering the response time of an actual nonlinear medium, it is several GHz to several tens of GHz or less, and the resonator length is about several micrometers to several tens of μm, so φ1-φ2
=n-2π-L(f+-L)/c×2π, and the phase change can be considered to be the same for frequencies f+ and fz.

Therefore, the input/output relationship of light intensity for each of frequencies f+ and fx is as follows: =δ+P Ou Le I + P Out* 2+
2 Pouto l” 1lut* 2XC
O5(2π・Δf−t+01□) ・・・・−・・・・(5) It becomes. Here, 3 is a proportionality constant, θ12=θl−
It is 02. In addition, regarding light intensity,
l ”P outo 2 = K3” l Eo
ut+ 2' 12P,,,, = K3・``E*
I l”.

It was standardized as Pt2=ni3・IEti, 212.

From equation [5], it can be seen that the phase change oscillates at a period of the frequency difference Δf. Accordingly, the ratio of the input light intensity to the output light intensity, that is, P. utwl vs. P1rl+1 and P, □, 2 vs. P i R12 oscillate.

FIG. 3 shows an example of calculation results of the input/output relationship when inputting three wavelengths. , in this figure, the horizontal axis is the input light intensity P lt+e
I and the vertical axis is the output light intensity P. It is set as ut*I. Further, in this figure, the input/output relationship when cos (2π·Δf−t+θ, 2)=±1 is shown by a solid line. Since the input/output relationship at any time is l cos(2π·Δf−t+01□)1≦1, it is located between the two solid lines.

In this calculation, δ=2.5 R=0.7 p th, 2 =0.02. In FIG. 3, the value when the optical pulse I is not input, that is, the value when Pin*2 is zero, is also shown by a broken line. This is the same curve as in FIG.

Using the input/output relationship shown in FIG. 3, it is possible to switch from an on state to an off state using an optical pulse. First, the light intensity P. Assume that the Fabry-Perot resonator 1 is in an on state with only the input light ■□ being input. There, a suitable light intensity, e.g. P l-, 2=
When inputting 0.02 light pulse ■8, Fabry-Perot
The human output relationship of the resonator 1 oscillates as shown in FIG.

During this oscillation, the light intensity P. There is a moment when only a low output state can be taken for the input light ■□. Referring to FIG. 3, the instant is when the value of CO8 (2π·Δf−t+θ12) is −1 or around it. Therefore, at this moment, the output state of the Fabry-Perot resonator 1 changes from the on state to the off state. When the output changes to the OFF state, there is no moment during this oscillation where only the high output state can be taken, so the output does not return to the ON state and remains in the OFF state. Even if the optical pulse I is erased and the input/output relationship is restored to its original state, the output state of the Fabry-Perot resonator 1 remains in the off state. That is,
Switching from the on state to the off state can be performed by the light pulse (1).

In this way, the Fabry-Perot resonator 1 is used as a bistable element, and is set from an off state to an on state by an optical pulse Is having the same phase and frequency as the input light IR, which is a bias light, and is set to an on state by an optical pulse Is having the same phase and frequency as the input light IR, which is a bias light. It is possible to reset from the on state to the off state by the weight 8, and a flip-drop operation can be performed.

FIG. 4 shows the structure of an optical flip-flop according to a second embodiment of the present invention.

This embodiment uses a semiconductor optical amplification element 1' as a Fabry-Perot resonator. The semiconductor optical amplification device 1' is structurally equivalent to a normal semiconductor laser device.

That is, the active layer itself is a nonlinear medium whose refractive index changes depending on the light intensity, and the opposing end surfaces each serve as a half mirror.

However, it differs from a semiconductor laser device in that the bias current supplied from the power supply 4 during operation is slightly smaller than the oscillation threshold. When this device is supplied with a partial bias current slightly smaller than the oscillation threshold, it operates as an optical bistable device.

The semiconductor optical amplifying element 1' has a gain with respect to light, and furthermore, the gain has a dependence on light intensity. Therefore, the output characteristics are slightly different from those of the Fabry-Perot resonator of the first embodiment. First, the fundamental relational expression between the optical electric fields is IEot = t-E"(L), (6). Here, E i Th, EoutSE"(0)
, E"(L), E-(0), E-(L), θ, r
, t, φ and δ are the same as those used in the explanation of the first embodiment. In addition, G is the gain received while the light travels the length of the resonator, g is the gain coefficient of the medium, α is the loss coefficient of the medium, go is the unsaturated gain coefficient, P□t is the saturated light intensity, b
is the ratio between the rate of increase in the real part and the rate of decrease in the imaginary part of the refractive index. Here, as in the first embodiment, mean field approximation was used. By solving these equations simultaneously, the human output characteristic of the semiconductor optical amplifier 1' corresponding to equation [1], Xfσ・Etn・exp [i(2πf t−φ+θ)
]−・(7) is obtained.

The material of the semiconductor optical amplifying element 1' is itself a nonlinear medium whose refractive index changes depending on the light intensity, and the same phase change φ as in the first embodiment can be obtained. Therefore, similarly to the first embodiment, by inputting an optical pulse having the same phase and frequency as the regularly inputted light, the semiconductor optical amplifying element 1' can be brought into the on state from the off state.

Next, a case will be described in which light having different frequencies is input. In that case, the input-output relationship when the frequency difference is sufficiently small and the nonlinear refractive index of the semiconductor optical amplification element 1' can sufficiently respond to the frequency difference is +2. ut*l・. ut+2 x cos (2π・Δf−t+θ12)・・−(8
) becomes. The notations of G, g, and φ are the same as in equation (6).

Moreover, P iR+ l s P Guts l is the input light intensity and output light intensity for the input light of frequency f, P
ih+ I, POuL+1 are the input light intensity and output light intensity for the input light of frequency f2, Δf=j, -f2, θ12
is the phase difference between the two input lights at the time of input. Here, for the same reason as in the first embodiment, it is assumed that the phase change for each frequency of light is the same.

Referring to equations (7) and (8), it can be seen that the phase change φ and gain coefficient g in equation (7) oscillate in accordance with the light intensity P expressed by the third equation in equation (8). Along with this oscillation, the input/output relationship expressed by the first and second equations (8) also oscillates. An example of calculation of this input/output relationship is shown in FIG.

In Figure 5, the vertical axis is P Out + I / P sut
The input/output relationship is shown with the horizontal axis as P 1h+l/Psut. In this figure, the solid line indicates the value when cos (2π·Δf−t+θ12)=±1. The input/output relationship at any given time is distributed between these two solid lines.

In this calculation, L=500jIJB R=0.35 δ=2.5 b=5 α=10 cm''''' go = 29.5 cm''''I P*-,z/P□, = 0.0005 did. Further, in FIG. 5, the input/output relationship when there is no input light of frequency f2, that is, when P 1h, t a =00, is also shown by a broken line.

The input/output relationship shown in FIG. 5 is similar to the input/output relationship shown in FIG. 3. Therefore, an input light having a frequency fls and an optical intensity Pa is constantly input to the semiconductor optical amplifying element 1', and an optical pulse having an optical intensity Ps having the same frequency and phase as this input light is used for setting. A flip-flop operation can be obtained in the same manner as in the first embodiment by using a light pulse having a different frequency from that of light for resetting.

In the above embodiments, spatially propagating light is assumed as the first to third light injection means, and an example is shown in which the optical paths are matched using a half mirror. The present invention can be similarly implemented using a directional coupler or an optical multiplexer used for wavelength multiplexing transmission.

In addition, although we have shown an example in which the input light that is steadily injected and the set light pulse pass through different optical paths, it is also possible to use a light source with variable output intensity and have them input from the same light source through the same optical path. The present invention can also be carried out in the same manner. Furthermore, the present invention can be similarly implemented by branching light from the same light source into two and switching one of the optical paths to turn on and off the set light pulse.

〔Effect of the invention〕

As explained above, the optical flip-flop of the present invention is
Switching between the on state and the off state can be performed using a light pulse without using an electrical signal. Therefore, it is possible to obtain an all-optical logic circuit with extremely high response time.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram of an optical flip-flop according to a first embodiment of the present invention. FIG. 2 is a diagram showing the relationship between the input light intensity and the output light intensity of a Fabry-Perot resonator. FIG. 3 is a diagram showing the relationship between input light intensity and output light intensity when three wavelengths are input. FIG. 4 is a configuration diagram of an optical flip-flop according to a second embodiment of the present invention. FIG. 5 is a diagram showing the relationship between input light intensity and output light intensity of a semiconductor optical amplification element. FIG. 6 is a structural diagram of a first conventional optical flip-flop. FIG. 7 is a diagram showing characteristics. FIG. 3 is a circuit diagram of a second conventional optical flip-flop. FIG. 9 is a diagram showing the characteristics of optical output with respect to injection current of a polarized bistable semiconductor laser. 1... Fabry-Perot resonator, 1'... semiconductor optical amplification element, 2.3.11.12... half mirror 113
...Nonlinear element, 4...Power supply. Is IR Tsuta 1 Kaijima 2 Diagram Pin, 2 Plan 3 Mouth Is IR Pant/Ps Shi- Kaoru 6 Diagram Injection Electric Grave 7 Diagram M 8 Diagram

Claims (2)

[Claims] (1) A Fabry-Perot resonator with a structure in which a nonlinear medium whose refractive index changes depending on the light intensity is inserted between two opposing half mirrors, and the input light of this Fabry-Perot resonator is a first light injection means that steadily injects light of a first frequency (f_1) that causes a hysteresis relationship between the intensity and the output light intensity, and a first light injection means that has the same phase and frequency as the light of the first frequency; a second light injection means for injecting a light pulse into the Fabry-Perot resonator at an input level such that the output light intensity of the Fabry-Perot resonator is at a high level;
third light injection means for injecting a light pulse having a second frequency (f_2) different from the first frequency into the Fabry-Perot resonator at an input level such that the output light intensity of the Fabry-Perot resonator is at a low level; Optical flip-flop with. (2) The optical flip-flop according to claim 1, wherein the Fabry-Perot resonator includes a semiconductor optical amplification element supplied with a bias current slightly smaller than its oscillation threshold.