RU2554286C1 - Isolation of signal with maximum intensity - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: proposed method comprises application of an integer number of pairs consisting of zero and first tunnel-coupled nonlinear-optical waveguides (TCNOW). Length of every TCNOW includes even or odd number of radiation power pumping at low input intensity when optical nonlinearity influence on power pumping may be neglected. Note here that low and high intensity signals affecting the pumping owing to nonlinearity is fed to zero TCNOW waveguide. Radiation from output of zero or first TCNOW waveguide is fed to zero waveguide of the next TCNOW pair. Parameters of all TCNOW and intensity range of input signal intensity at zero waveguide of every TCNOW are selected so that high intensity signal exits from zero waveguide at odd number of pumping for low input intensities in TCNOW wavelengths, or from first waveguide at even number of pumping for low intensities in wavelengths of these TCNOW.

EFFECT: isolation of signal portion with maximum intensity.

12 cl, 6 dwg

Description

The invention relates to the field of optical location and laser technology.

A known method of separating part of the signal with a maximum value of intensity, which consists in the following. A photodiode receives an optical signal and converts it into an electric signal, which is recorded by electronic equipment for small successive time intervals that follow one after another. The equipment integrates the signal for each time interval, and the duration of this interval is determined by the extremely short response time (resolution) of the equipment, i.e. high-speed electronic equipment. Comparing the elements of the array of signal power values for different intervals, find the interval corresponding to the maximum of the received signal, and the corresponding value of the maximum signal power.

The disadvantage of this method is its relative slowness, due to the fundamental limitation of the speed of electrical circuits, devices and elements.

Known used in laser systems, intracavity and non-resonant electro-optical methods for highlighting or cutting the maximum single pulse, or part of a single pulse [1]. In them, at a certain moment, a specially formed electrical pulse is supplied to the electro-optical modulator. The disadvantage of these methods is also their relative slowness: the optical pulses formed at the output have a duration of 0.5-10 ns [1].

The objectives of the claimed invention are the formation of ultrashort pulses and improving the accuracy of the optical location [2].

The technical result achieved in the claimed invention is that for the first time the task of isolating part of the signal with a maximum intensity value is solved by optical means.

Another technical result achieved in the claimed invention is a sharp decrease (by 4 orders of magnitude) of the duration of the allocated part of the signal with a maximum intensity value. Based on this invention, it is possible to quickly determine (find) the maximum value of the intensity of the received optical (electromagnetic) radiation and the corresponding time instant. This is important for location.

The indicated technical results are achieved in the method of isolating a part of the signal with a maximum intensity value, including the use of an integer (n) of pairs of tunnel-coupled nonlinear optical waveguides (TSNOV), “0” and “1”, an odd or even number is placed along the length of each TSNOV pumping radiation power at low input intensities, when the influence of optical nonlinearity on the process of pumping power can be neglected, while introducing a signal with both small and large values of the intensity, affecting the whole The nonlinearity is imposed on the pumping process to the TSNOV zero waveguide, and the radiation from the output of the TSNOV zero or first waveguide, respectively, is fed to the next waveguide of the next pair of TSNOVs, and the parameters of all TSNOVs and the input signal intensity range at the input of the zero waveguide of each TSNOV are selected so that the signal with a large intensity value, respectively, leaves the zero waveguide, with an odd number of pumpings for small input intensities along the length of these TSNOVs, or from the first waveguide of these TSNOVs, with even after pumping for low input intensities along the length of these TSNOV.

In the particular case, two pairs of TSNOV are used, while the signal is input into the zero waveguide of the first pair of TSNOV, and the radiation from the output of the zero or first waveguide of the first pair of TSNOV, respectively, is fed into the zero waveguide of the second pair of TSNOV, with the parameters of all TSNOV and the input signal intensity range the input of the zero waveguide of both pairs of TSNOVs are selected so that a signal with a large intensity value exits, respectively, from the zero waveguide, with an odd number of transfers for small input intensities along the length of these TSNOVs, or from the first waveguide of these TSNOV, with an even number of pumpings for small input intensities along the length of these TSNOV.

Usually TSNOV have cubic and / or quadratic nonlinearity.

As a rule, one or two pumping power of radiation is placed along the length of TSNOV at low input intensities.

In the presence of optical losses, the tunnel coupling coefficient decreases, and the length of the tunnel communication increases with the transition to subsequent TSNOV. Moreover, the coupling coefficient in subsequent TSNOV is proportional to the fraction of the signal power transmitted from previous TSNOV to the next.

Usually TSNOV are identical.

If the input signal is weak, then before applying the TSNOV to the indicated sequence, the signal under investigation is fed to the input of an optical transistor based on a nonlinear system with unidirectional distributed coupled waves [3–9].

As a rule, this optical transistor is made on the basis of a pair of TSNOVs, while the input of one (“zero”) or another (“first”) waveguide of this optical transistor is supplied with optical pump radiation from a laser, and the maximum value of the intensity of this pump radiation is higher than the threshold intensity at which the differential gain of the signal exceeds unity. Usually the value of the intensity of the input pump radiation is constant. Typically, the waveguides in these TSNOV are identical. In this case, as a rule, the pump intensity is close to the critical intensity of self-switching for this pair of TSNOV, near which the maximum differential signal amplification is achieved. Typically, the carrier frequency of the optical pump radiation is different from the carrier frequency of the signal.

In another particular case, the optical transistor is made on the basis of a nonlinear system with unidirectional distributed-coupled waves of different polarizations.

This method is based on the phenomenon of light self-switching that we discovered in tunnel-coupled nonlinear optical waveguides [3-9] and is illustrated by Figures 1-6. Figure 1 shows the self-switching curves [3-8]. The abscissa axis represents the normalized intensity at the input of the zero waveguide, and the ordinate axis represents the power transfer coefficient of the zero waveguide. The vertical dashed line in FIG. 1b and the left vertical one in FIG. 1a correspond to a threshold intensity above which the differential gain is greater than unity. Point M on the self-switching curve (Figure 1) corresponds to a critical intensity, near which the maximum differential signal gain is achieved. Figure 2 shows a diagram of devices for implementing the method. In figures 3A, 4A, 5A, 6A shows the temporary profiles of the input signals, the maximum of which must be found or highlighted. In other figures 3-6 shows the temporal profiles of the signals at the output of the first (b), second (c) and third (d) TSNOV.

Figures 3-6 correspond to certain maxima of the normalized intensity at the input and output of each pair TSNOV.

(Фиг.3б). Максимум нормированной интенсивности на выходе нулевого волновода второй пары $$J_{0l,MAX}^{(2)}=M2=0.88$$

(Фиг.3в). Максимум нормированной интенсивности на выходе нулевого волновода третьей пары $$J_{0l,MAX}^{(3)}=M3=0.806$$

(Фиг.3г). При малых входных мощностях на длине первой пары ТСНОВ укладывается одна перекачка энергии: L1=π, а на длине второй и третьей пары ТСНОВ укладывается 3 перекачки энергии: L2=L3=3π. Здесь и ниже для обозначения нормированных интенсивностей частично использованы обозначения [3-8].For Fig. 3, the maximum of the normalized intensity of the initial signal at the input of the zero waveguide of the first pair of TSNOVs is R _{0, MAX} = M0 = 1.1 (Fig. 3a). For the same Figure 3 at the output TSNOV maxima normalized intensity have the following values. The maximum of the normalized intensity at the output of the zero waveguide of the first pair $$J_{0l,MAX}^{(\mathrm{one})}=M\mathrm{one}=0.90$$

(Fig.3b). The maximum of the normalized intensity at the output of the zero waveguide of the second pair $$J_{0l,MAX}^{(2)}=M2=0.88$$

(Fig. 3c). The maximum of the normalized intensity at the output of the zero waveguide of the third pair $$J_{0l,MAX}^{(3)}=M3=0.806$$

(Fig. 3d). At low input powers, one transfer of energy is stacked along the length of the first pair of TSNOV: L1 = π, and three transfers of energy are stacked along the length of the second and third pair of TSNOV: L2 = L3 = 3π. Here and below, the notation [3-8] is partially used to denote normalized intensities.

(Фиг.4б). Максимум нормированной интенсивности на выходе нулевого волновода второй пары $$J_{0l,MAX}^{(2)}=M2=0.628$$

(Фиг.4в). Максимум нормированной интенсивности на выходе нулевого волновода третьей пары $$J_{0l,MAX}^{(3)}=M3=0.274$$

(Фиг.4г). При малых входных мощностях на длине первой пары ТСНОВ укладывается две перекачки энергии: L1=2π, на длине второй пары ТСНОВ укладываются три перекачки: L2=3π и на длине третьей пары ТСНОВ укладываются 7 перекачек энергии: L3=7π.For Fig. 4, the maximum of the normalized intensity of the initial signal at the input of the zero waveguide of the first pair of TSNOVs is R _{0, MAX} = M0 = 0.985 (Fig. 4a). For the same Figure 4 at the output TSNOV maxima normalized intensity have the following values. The maximum normalized intensity at the output of the first waveguide of the first pair $$J_{\mathrm{one}l,MAX}^{(\mathrm{one})}=M\mathrm{one}=0.985$$

(Fig.4b). The maximum of the normalized intensity at the output of the zero waveguide of the second pair $$J_{0l,MAX}^{(2)}=M2=0.628$$

(Figv). The maximum of the normalized intensity at the output of the zero waveguide of the third pair $$J_{0l,MAX}^{(3)}=M3=0.274$$

(Fig. 4d). At low input powers, two energy transfers are stacked along the length of the first pair of TSNOVs: L1 = 2π, three transfers are stacked along the length of the second pair of TSNOVs: L2 = 3π and 7 energy transfers are stacked along the length of the third pair of TSNOVs: L3 = 7π.

(Фиг.5б). Максимум нормированной интенсивности на выходе первого волновода второй пары $$J_{1l,MAX}^{(2)}=M2=0.985$$

(Фиг.5в). Максимум нормированной интенсивности на выходе первого волновода третьей пары $$J_{1l,MAX}^{(3)}=M3=0.985$$

(Фиг.5 г). При малых входных мощностях на длине первой, второй и третьей пары ТСНОВ укладывается две перекачки энергии: L1=L2=L3=2π.For Fig. 5, the maximum of the normalized intensity of the initial signal at the input of the zero waveguide of the first pair of TSNOV is R _{0, MAX} = M0 = 0.985 (Fig. 5a). For the same Figure 5 at the output TSNOV maxima normalized intensity have the following values. The maximum normalized intensity at the output of the first waveguide of the first pair $$J_{\mathrm{one}l,MAX}^{(\mathrm{one})}=M\mathrm{one}=0.985$$

(Fig.5b). The maximum normalized intensity at the output of the first waveguide of the second pair $$J_{\mathrm{one}l,MAX}^{(2)}=M2=0.985$$

(Fig. 5c). The maximum normalized intensity at the output of the first waveguide of the third pair $$J_{\mathrm{one}l,MAX}^{(3)}=M3=0.985$$

(Fig. 5 g). At low input powers, two energy transfers are stacked along the length of the first, second, and third pairs of TSNOV: L1 = L2 = L3 = 2π.

(Фиг.6б). Максимум нормированной интенсивности на выходе первого волновода второй пары $$J_{1l,MAX}^{(2)}=M2=0.492$$

(Фиг.6в). Максимум нормированной интенсивности на выходе первого волновода третьей пары $$J_{1l,MAX}^{(3)}=M3=0.246$$

(Фиг.6г). При малых входных мощностях на длине первой, второй и третьей пары ТСНОВ укладывается две перекачки энергии: L1=L2=L3=2π.For Fig.6, the maximum of the normalized intensity of the initial signal at the input of the zero waveguide of the first pair of TSNOV is equal to R _{0, MAX} = M0 = 0.985 (Figa). Fig.6 is built taking into account energy losses during its transmission from the first TSNOV to the second TSNOV and from the second TSNOV to the third TSNOV. The energy transfer coefficient is σ ₃₂ = σ ₂₁ = σ = 0.5. For Fig.6 at the output TSNOV maxima normalized intensity have the following values. The maximum normalized intensity at the output of the first waveguide of the first pair $$J_{\mathrm{one}l,MAX}^{(\mathrm{one})}=M\mathrm{one}=0.985$$

(Fig.6b). The maximum normalized intensity at the output of the first waveguide of the second pair $$J_{\mathrm{one}l,MAX}^{(2)}=M2=0.492$$

(Figv). The maximum normalized intensity at the output of the first waveguide of the third pair $$J_{\mathrm{one}l,MAX}^{(3)}=M3=0.246$$

(Fig.6g). At low input powers, two energy transfers are stacked along the length of the first, second, and third pairs of TSNOV: L1 = L2 = L3 = 2π.

For Fig.3-6, the "envelope" of the pulse train has the following ("bell-shaped") form:

I ₀₀ = I _max / ch ² [(tt _max ) / τ]

Figures 3-6 show discrete signals, but the proposed method is applicable to continuous signals.

As TSNOV in this method, it is sometimes convenient to use TSNOV with a variable distance between waveguides, in which the distance between the waveguides initially decreases to a certain minimum value, and then increases. Accordingly, the tunnel coupling coefficient in them initially increases from zero to a maximum value, and then decreases to zero. Those. the tunnel coupling coefficient varies in length according to a bell-shaped law. In such TSNOVs, as shown by us [8], effective and complete self-switching of optical radiation can occur.

To implement the proposed method TSNOV can have both cubic optical nonlinearity [3-8] and quadratic optical nonlinearity [9].

Examples of carrying out the invention

, θ - нелинейный коэффициент волновода, I₀₀ - интенсивность на входе нулевого волновода. Коэффициент туннельной связи подобран так, что нормированная длина ТСНОВ составляет L=π, т.е. на длине первых ТСНОВ укладывается одна линейная перекачка излучения. При этом на выходе нулевого волновода формируется сигнал, показанный на Фиг.3б. Этот сигнал значительно ужат во времени по сравнению с исходным: в нем остаются только импульсы, близкие к максимальному. Периферийная часть исходного сигнала (малая по интенсивности) перебрасывается на выход другого, первого, волновода. И эта часть сигнала как бы уводится из «игры» с помощью первого волновода. Далее, уже «поджатый во времени» сигнал с выходы нулевого волновода (первой пары ТСНОВ) подается на вход нулевого волновода следующей (второй) пары ТСНОВ (Фиг.2а). Параметры этой «второй» пары ТСНОВ подобраны так: L=3π, т.е. на длине вторых ТСНОВ укладывается три линейных перекачки излучения. На выходе нулевого волновода этих вторых ТСНОВ мы получаем еще более поджатый по времени сигнал (Фиг.3в). Остается максимальный импульс и два ближайших к нему импульса, причем интенсивности этих соседних импульсов сильно «зарезаны» по сравнению с этими импульсами в исходном сигнале. Далее, сигнал с выхода нулевого волновода второй пары ТСНОВ подается на вход нулевого волновода третьей пары ТСНОВ (Фиг.2а). Параметры этой «третьей» пары ТСНОВ подобраны так: L=3π, т.е. на длине третьих ТСНОВ укладывается три линейных перекачки излучения. И вот на выходе нулевого волновода этих третьих ТСНОВ выделяется только максимальный импульс (Фиг.3г), т.е. тот импульс, который соответствует максимальному значению исходного сигнала. Таким образом, поставленная задача решена: выделен максимум исходного сигнала и, значит, определен момент достижения максимума исходного сигнала.Example 1. Corresponds to the circuit shown in Fig.2A. The signal (having the form shown in Fig. 3a) is fed to the input of the first pair of TSNOVs and is inserted entirely into the zero waveguide of this pair, and the maximum input intensity I ₀₀ corresponds to the normalized intensity R ₀ = 1.1 for this pair. Hereinafter, R ₀ = I ₀₀ / I _0M , where I _0M is the critical intensity, and for identical TSNOV $$I_{0M}=\mathrm{four}K/\left|\theta \right|\left[\mathrm{four}-7\right]$$

, θ is the nonlinear waveguide coefficient, I ₀₀ is the intensity at the input of the zero waveguide. The tunnel coupling coefficient is chosen so that the normalized TSNOV length is L = π, i.e. along the length of the first TSNOV one linear pumping of radiation is laid. At the same time, at the output of the zero waveguide, the signal shown in Fig. 3b is formed. This signal is significantly shrunk in time in comparison with the initial one: only pulses close to the maximum remain in it. The peripheral part of the initial signal (low in intensity) is transferred to the output of another, first, waveguide. And this part of the signal, as it were, is removed from the “game” with the help of the first waveguide. Further, the signal "time-delayed" from the outputs of the zero waveguide (the first pair of TSNOV) is fed to the input of the zero waveguide of the next (second) pair of TSNOV (Fig. 2a). The parameters of this “second” pair of TSNOVs are chosen as follows: L = 3π, i.e. along the length of the second TSNOV three linear pumping of radiation fit. At the output of the zero waveguide of these second TSNOV, we get an even more time-compressed signal (Fig.3c). There remains the maximum pulse and the two pulses closest to it, and the intensities of these neighboring pulses are strongly “cut off” in comparison with these pulses in the original signal. Next, the signal from the output of the zero waveguide of the second pair of TSNOV is fed to the input of the zero waveguide of the third pair of TSNOV (Fig. 2a). The parameters of this “third” pair of TSNOVs are selected as follows: L = 3π, i.e. along the length of the third TSNOV, three linear pumping of radiation is stacked. And at the output of the zero waveguide of these third TSNOV, only the maximum pulse is allocated (Fig. 3d), i.e. the pulse that corresponds to the maximum value of the original signal. Thus, the problem has been solved: the maximum of the initial signal is highlighted and, therefore, the moment of reaching the maximum of the initial signal is determined.

Example 2. Corresponds to the circuit shown in Fig.2b. The radiation (having the form shown in Fig. 4a) is fed to the input of the first pair of TSNOVs and introduced entirely into the zero waveguide of this pair, and the maximum input intensity corresponds to the normalized intensity R ₀ = 0.985 for this pair. The tunnel coupling coefficient is chosen so that the normalized TSNOV length is L = 2π, i.e. along the length of the first TSNOV two linear pumping of radiation are stacked. As a result, part of the signal, near its maximum, is almost completely transferred to the output of the first waveguide. Next, the selected signal from this "first" waveguide is fed to the input of the next pair of TSNOV. Moreover, it is fed to the input of the zero waveguide of this pair. Those. the waveguide that was “first” for the first pair of TSNOVs becomes “zero” for the next pair of TSNOVs. It is easy to implement. To this end, a new waveguide is introduced at the continuation of the waveguide, which was “first” for the first pair, at a small distance from the exit from the first TSNOV, forming a new (second) pair of TSNOV with this waveguide. For the second TSNOV, this new waveguide plays the role of the “first” waveguide and selects and “removes from the game” the unnecessary part of the signal. For this, the length and coupling coefficient of the second TSNOV are selected so that L = π. At the output of the zero waveguide of these second TSNOV, we get an even more time-compressed signal (Fig.4c). There remains the maximum pulse and the two pulses closest to it, and the intensities of these neighboring pulses are greatly reduced ("slaughtered") compared to these pulses in the original signal. And finally, in order to finally isolate the maximum signal by separating it from two adjacent pulses, we feed the signal from the output of the zero waveguide of the second TSNOV to the input of the zero waveguide of the third TSNOV. The length and coupling coefficient of the second TSNOV are selected so that L = 7π. As a result, at the output of the zero waveguide of the third TSNOV, the desired signal is extracted (Fig. 4d), which corresponds to the maximum of the initial signal.

Example 3. Corresponds to the circuit shown in Figv. The signal, having the form shown in Fig. 5a, is fed to the input of the first pair of TSNOVs and is inserted entirely into the zero waveguide of this pair, and the maximum intensity of the input signal corresponds to the normalized intensity R ₀ = 0.985 for this pair. The tunnel coupling coefficient is chosen so that the normalized TSNOV length is L = 2π, i.e. along the length of the first TSNOV two linear pumping of radiation are stacked. As a result, part of the signal, near its maximum, is almost completely transferred to the output of the first waveguide. Next, the selected signal from this "first" waveguide is fed to the input of the next pair of TSNOV. Moreover, it is fed to the input of the zero waveguide of this pair. Those. the waveguide that was “first” for the first pair of TSNOVs becomes “zero” for the next pair of TSNOVs. It is easy to implement. To this end, a new waveguide is introduced at the continuation of the waveguide, which was “first” for the first pair, at a small distance from the exit from the first TSNOV, forming a new (second) pair of TSNOV with this waveguide. Next, the difference from Example 2 begins. For the second TSNOV, this new waveguide also plays the role of the “first” waveguide, but it does not select and “not remove from the game” the unnecessary part of the signal, but, on the contrary, selects the desired maximum signal (Fig.5b). For this, the length and coupling coefficient of the second TSNOV are selected so that L = 2π. At the output of the first waveguide of these second TSNOV, we get an even more time-compressed signal (Fig. 5c). There remains the maximum pulse and the two pulses closest to it, and the intensities of these neighboring pulses are greatly reduced in comparison with these pulses in the original signal. And, finally, in order to finally isolate the maximum signal by separating it from two adjacent pulses, we apply a signal from the output of the first waveguide of the second TSNOV to the input of the zero waveguide of the third TSNOV (Fig.2c). The length and coupling coefficient of the second TSNOV are again chosen so that L = 2π. As a result, at the output of the first waveguide of the third TSNOV, the desired signal is selected corresponding to the maximum of the initial signal (Fig. 5g). The advantage of this variant of the method is that the selected maximum signal (Fig. 5g) is almost equal in magnitude to the initial maximum signal received at the input of the first TSNOV (Fig. 5a). Those. there are almost no losses in the intensity of the maximum signal. But at the same time, it is necessary to establish quite accurately a certain ratio of the intensities of the input signals with the coupling coefficient TSNOV.

. Выберем коэффициент связи вторых ТСНОВ меньшим, чем коэффициент связи первых ТСНОВ, а именно K₂=σ₂₁K₁. Т.е. K₂ пропорционален доле передаваемой мощности. Длину вторых ТСНОВ l₂, выберем большей, чем длина первых ТСНОВ l₁, причем так: l₂=l₁/σ₂₁, чтобы нормированная длина связи вторых ТСНОВ по-прежнему была L=2π. Т.е. на длине связи вторых ТСНОВ также укладывается 2 линейных перекачки мощности (несмотря на меньший коэффициент туннельной связи). При выбранных параметрах все введенное во вторые ТСНОВ излучение, соответствующее его максимуму, выходит из первого волновода этих ТСНОВ. С выхода первого волновода вторых ТСНОВ излучение вводится в нулевой волновод третьих ТСНОВ. Допустим, передаваемая доля мощности составляет σ₃₂, т.е. $$I_{00}^{(3)}={\sigma }_{32}{I}_{1l}^{(2)}$$

Аналогично, выбирая параметры третьих ТСНОВ: K₃=σ₃₂K₂, l₃=l₂/σ₃₂, получаем, что искомый максимум излучения полностью выходит из первого волновода третьих ТСНОВ (Фиг.6г). С учетом потерь этот максимум составляет от максимума исходного излучения на входе первых ТСНОВ такую долю: $$I_{1l}^{(3)}={\sigma }_{32}{\sigma }_{21}{I}_{00}$$

. Обычно σ₃₂=σ₂₁=σ.Example 4. Corresponds to the circuit shown in Figv. At the junctions of the waveguides, and indeed in the waveguides themselves, the loss of radiation power is inevitable. Modern technologies can minimize losses, but still they cannot be completely eliminated. The given example shows that this method also works in conditions of losses, and even in conditions of large losses. Let, as in the previous example, the radiation (having the form shown in Fig. 6a) is fed to the input of the first pair of TSNOVs and introduced entirely into the zero waveguide of this pair (Fig. 2c), and the maximum input intensity corresponds to the normalized intensity R ₀ = 0.985 for this pair. The tunnel coupling coefficient and the TSNOV length are selected so that the normalized TSNOV length is L = 2π, i.e. along the length of the first TSNOV two linear pumping of radiation fit. As a result, part of the signal, near its maximum, is almost completely transferred to the output of the first waveguide (Fig.6b). Let, as in the previous example, the radiation from the output of the first waveguide of the first TSNOV be fed to the input of the zero waveguide of the second TSNOV. But now there are losses between the first TSNOV and the second TSNOV. Suppose the share of transmitted power is σ ₂₁ , i.e. $$I_{00}^{(2)}={\sigma }_{21}{I}_{\mathrm{one}l}$$

. We choose the coupling coefficient of the second TSNOV less than the coupling coefficient of the first TSNOV, namely K ₂ = σ ₂₁ K ₁ . Those. K _{2 is} proportional to the share of transmitted power. The length of the second TSNOV l ₂ , we choose longer than the length of the first TSNOV l ₁ , and so: l ₂ = l ₁ / σ ₂₁ so that the normalized bond length of the second TSNOV still L = 2π. Those. along the communication length of the second TSNOV, 2 linear pumping power is also laid (despite the lower coefficient of tunnel communication). With the chosen parameters, all the radiation introduced into the second TSNOV, corresponding to its maximum, leaves the first waveguide of these TSNOV. From the output of the first waveguide of the second TSNOV, the radiation is introduced into the zero waveguide of the third TSNOV. Suppose the transmitted fraction of the power of σ _32, i.e., $$I_{00}^{(3)}={\sigma }_{32}{I}_{\mathrm{one}l}^{(2)}$$

Similarly, choosing the parameters of the third TSNOV: K ₃ = σ ₃₂ K ₂ , l ₃ = l ₂ / σ ₃₂ , we find that the desired radiation maximum completely leaves the first waveguide of the third TSNOV (Fig.6d). Taking into account losses, this maximum makes up such a fraction of the maximum of initial radiation at the input of the first TSNOV: $$I_{\mathrm{one}l}^{(3)}={\sigma }_{32}{\sigma }_{21}{I}_{00}$$

. Typically, σ ₃₂ = σ ₂₁ = σ.

Similarly, losses in the circuits of Figs. 2a, b and other possible circuits are taken into account. Thus, it is possible to implement this method even in conditions of significant power loss. This is achieved by reducing the coupling coefficient (and increasing the bond length) of the subsequent TSNOVs relative to the previous ones, and by choosing a certain ratio between these coupling coefficients. The greater the loss, the lower the coupling coefficient of subsequent TSNOVs: the coupling coefficient of subsequent TSNOVs is proportional to the fraction of transmitted (passing) power and the coupling coefficient of previous TSNOVs. More precisely, the coupling coefficient of subsequent TSNOVs is equal to the product of the fraction of transmitted (passing) power and the coupling coefficient of previous TSNOVs.

LITERATURE

1. Measurement of spectral-frequency and correlation parameters and characteristics of laser radiation. Edited by A.F. Kotyuk, B.M. Stepanova. Radio and Communications Publishing House, Moscow, 1982

2. Laser measuring systems, ed. D.P. Lukyanova, M., 1981; V.V. Prayer, Optical location systems, M., 1981; M.S. Malashin, R.P. Kamensky, Yu.B. Borisov, Fundamentals of the design of laser location systems, M., 1983.

3. Mayer. "Optical transistors and bistable elements based on nonlinear light transmission by systems with unidirectional coupled waves." Quantum Electronics, 1982, Volume 9, No. 11, pp. 2296-2302.

4. A.A. Mayer A.A. "On the self-switching of light in a directional coupler." - Quantum Electronics, 1984, Volume 11, No. 1, p. 157-162.

5. D.D. Gusovsky, E.M. Dianov, A.A. Mayer et al. "Experimental observation of radiation self-switching in tunnel-coupled optical waveguides." - Preprint IOFAN No. 188, Moscow, 1986; Quantum Electronics, 1987, Volume 14, No. 6, p.1144-1147.

6. A.A. Mayer. "Optical self-switching of unidirectional distributed-coupled waves." UFN, 1995, volume 165, No. 9, pp. 1037-1075.

7. A.A. Mayer. "Experimental observation of the phenomenon of self-switching of unidirectional distributed-coupled waves" Physics – Uspekhi, 1996, Volume 166, No. 11, pp. 1171-1196.

8. A.A. Mayer. Self-switching of unidirectional distributed-coupled waves with a variable coupling coefficient. DAN ser. Fiz., 2009, Volume 428, No. 6, p. 752-755.

9. A.A. Mayer. The method of switching, control, amplification and modulation of optical radiation in quadratically nonlinear tunnel-coupled optical waveguides (options). RF patent No. 2153695. Priority June 10, 1998. Registered in the State Register on July 27, 2000.

Claims (12)

1. The method of extracting a part of the signal with a maximum intensity value, including the use of an integer number of pairs consisting of zero and first tunnel-coupled nonlinear optical waveguides (TSNOV), an odd or even number of transfers of radiation power at low input intensities is stacked along the length of each TSNOV, when the influence of optical nonlinearity on the power transfer process can be neglected, in this case, a signal with small and large intensity values that, due to nonlinearity, affect the process bumps into the TSNOV zero waveguide, and the radiation from the output of the TSNOV zero or first waveguide, respectively, is fed into the zero waveguide of the next pair of TSNOVs, and the parameters of all TSNOVs and the input signal intensity range at the input of the zero waveguide of each TSNOV are selected so that a signal with a large intensity respectively, from the zero waveguide, with an odd number of transfers for small input intensities along the length of these TSNOVs, or from the first waveguide of these TSNOVs, with an even number of transfers for small input intensities nsivnosti along the length of these TSNOV. 2. The method according to claim 1, characterized in that two pairs of TSNOV are used, the signal being introduced into the zero waveguide of the first pair of TSNOV, and radiation from the output of the zero or first waveguide of the first pair of TSNOV, respectively, is fed into the zero waveguide of the second pair of TSNOV, and the parameters of all TSNOVs and the range of the input signal intensity at the input of the zero waveguide of both pairs of TSNOVs are selected so that a signal with a large intensity value leaves the zero waveguide, respectively, with an odd number of transfers for low input intensities at ine TSNOV these, or these TSNOV first waveguide, with an even number of input perekachek for small intensities at a length of TSNOV. 3. The method according to claim 1, characterized in that TSNOV have cubic and / or quadratic nonlinearity. 4. The method according to claim 1, characterized in that in the presence of optical losses the tunnel coupling coefficient TSNOV decreases, and the length of the tunnel communication increases with the transition to subsequent TSNOV. 5. The method according to claim 1, characterized in that on the length of the TSNOV at low input intensities one or two pumpings of radiation power are stacked. 6. The method according to claim 1, characterized in that TSNOV are identical. 7. The method according to claim 1, characterized in that before applying the TSNOV to the indicated sequence, the test signal is fed to the input of an optical transistor based on a nonlinear system with unidirectional distributed coupled waves, which provides differential signal amplification. 8. The method according to claim 7, characterized in that before applying to the indicated TSNOV sequence, the signal is supplied to the input of the waveguide of the optical transistor based on the pair of TSNOV, and optical pump radiation from the laser is supplied to the input of this or another waveguide of this optical transistor, the maximum value the intensity of this pump radiation is higher than the threshold intensity at which the differential gain of the signal exceeds unity. 9. The method according to claim 8, characterized in that the intensity of the pump radiation is constant. 10. The method according to claim 8, characterized in that the pump intensity is close to the critical intensity of self-switching for this pair of TSNOV. 11. The method according to claim 8, characterized in that the carrier frequency of the optical pump radiation is different from the carrier frequency of the signal. 12. The method according to claim 7, characterized in that the optical transistor is made on the basis of a nonlinear system with unidirectional distributed-coupled waves of different polarizations.

Images