JP2001053684A - Optical pulse timing detecting circuit and optical time- division multiplexing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To actualized high-precision stability in a time domain by instantaneously observing and correcting variation in signal arrival timing due to physical expansion and contraction in a signal propagation direction caused by temperature variation of a long-size optical component, etc., without any expensive very-fast electric circuit. SOLUTION: An optical pulse cross-correlation reception part 32 detects an increase or decrease of overlap components between a reference optical pulse string and a signal light pulse string at a measurement position. The reception part 32 has an optical coupling part 34 which couples the reference light pulse string and signal light pulse string with each other, a couple of sum frequency light receivers 36 and 37 composed of nonlinear crystal which generates a sum frequency by inputting coupled pulses, a photodetecting element which generates two-photon absorption electric power, etc., a signal comparison part 38 which inputs the outputs of the receivers, etc. The signal comparison part 38 can specify the quantity and correction direction (increasing or decreasing) of delay to be corrected only by comparing two cross-correlation signal values. The correction signal of the signal comparison part 38 is fed back to an optical delay control circuit 31 to correct the delay quantity and correction direction.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical pulse timing detection circuit and an optical time division multiplexing apparatus, and more particularly to a broadband optical communication system in which a short optical pulse train is modulated to have an RZ (Return to Zero) format. The present invention relates to a circuit for detecting timing fluctuation at the time of optical pulse time multiplexing control and a stable operation of the optical time division multiplexing device. The present invention can also contribute to the generation, compression, switching, stabilization of the measurement system in general, and the reduction of jitter that require a long optical element (such as an optical fiber or a long-distance mirror optical path). is there.

[0002]

2. Description of the Related Art In signal processing, signal transmission, and measurement using short optical pulses, fluctuations in pulse intensity and timing are major factors that degrade the signal-to-noise ratio and resolution. FIG. 15 shows an example of a short optical pulse signal generation using a short pulse generating light source, a pulse compression fiber by dispersion, and an erbium-doped optical fiber amplifier, and a configuration example of a conventional optical time-division multiplexing apparatus using the signal and a cause of timing fluctuation. Show.
In FIG. 15, 1 is a short optical pulse generation circuit such as a laser diode, 2 is a pulse width compression circuit composed of a fiber with dispersion, 3 is an optical amplification circuit (EDFA) such as an erbium-doped optical fiber amplifier, 4 Is an optical branching unit that branches a short optical pulse into a plurality of M systems, 5 is a modulator that modulates the optical pulse with a transmission information signal, 6 is a delay adjuster that gives a desired delay to the modulated optical pulse, 7 is Reference numeral 8 denotes a modulator which modulates the optical pulse transmission information signal M of the M system and a delay adjuster which gives a desired delay to the modulated optical pulse. An M-system 9 is an optical multiplexing unit that outputs a time-division multiplexed signal of optical pulses by synthesizing optical pulses output from the M systems.

In the optical time-division multiplexing apparatus thus configured, short optical pulses a as shown in FIG. 1 are continuously generated from the short optical pulse generation circuit 1 at every predetermined repetition period f. , A short optical pulse a is input to the branching unit 4 and branched into a plurality (M) of optical waveguides. One of these branched short optical pulses is input to the first system 7, and the optical intensity is modulated in the modulator 5 by the transmission information signal synchronized with the repetition of the short optical pulse, and the RZ format as shown in FIG. The modulated short optical pulse b is obtained, and a predetermined time delay is given in the delay adjuster 6.

[0004] The other short optical pulse branched by the branching unit 4 is also input to each of the 2 to M systems, and similarly the modulated short optical pulse c. d. . . . It is said. And 1-M modulated short optical pulse b. c. d. . . . Is input to the optical multiplexing unit 9 and is synthesized on the time axis to be output as a time multiplexed signal z as shown in FIG. The cycle of this time multiplexed signal z is
It is 1 / M of the repetition period of the multiplexed pulse generated from the short light pulse generation circuit 1.

In order for the period of the time multiplexed signal z to be 1 / M of the repetition period of the pulse generated from the signal short pulse train generating circuit 1, the modulated short optical signal Pulse b. c. d. . . . Need to be positioned at equal intervals, the delay adjuster 6 is provided. As described above, an efficient optical communication system has been realized by time-multiplexing 1 / M-system short optical pulse signals.

[0006]

In the conventional optical time division multiplexing apparatus described above, the delay adjuster 6 is used at the time of initial startup.
Are fixed after adjusting the modulated short optical pulse so that the delay time difference between them becomes optimal. However, modulated short light pulses b. c. d. . . . The interval needed to generate
That is, the path from the output of the branching unit 4 to the input of the optical multiplexing unit 9 often has different arrangements and different lengths for reasons of device configuration and manufacturing.

Further, in order to obtain an ultrashort pulse signal, after a pulse is generated from a light source (signal short pulse train generation circuit) 1, a pulse width compression circuit (50-100m) using two fiber dispersion gradients or the like is used. 3 fiber type optical amplifier (1
It is necessary to use a long fiber-related component such as an increase in signal strength due to 0-50 m). Various components used for multiplexing and demultiplexing of signals are also included in the pigtail 1.
It is common to combine them with connecting fibers of about several meters called zero. When repetitive short optical pulse train f ₀ Hz is to respectively each wave modulation, when the bundling M wave, by passing through these elongate products, timing fluctuation (phase drift of Use point by expansion and contraction due to temperature change and stress ) Occurs.

For example, when a normal fiber component of 100 m is used, a change in the delay time of about 100 ps is caused by a temperature change of 10 ° C. of the medium (the thermal expansion and contraction rate of the fiber component is 2).
× 10 ^-5 / ° C). Therefore, for example, when the transmission rate f is 20 Gbit / s, even if the temperature fluctuates by only 3 degrees, a pulse timing fluctuation equivalent to 60% of one bit occurs at the time of reaching the modulator 5, and a modulation error of a signal is generated. Cause. This is the first problem to be solved for the timing fluctuation.

Next, even if individual signal modulation is successfully performed, if the pulse repetition period fluctuates,
Jitter occurs in the repetition frequency, and a short optical pulse b. c. d. . . . Cannot be accurately separated and good reception cannot be performed. However, if there is a difference of 2 m in the length of the optical circuit component fiber component of each system to be considered,
A temperature fluctuation of only 5 degrees causes a time fluctuation of 1 ps or more, and 120 Gbit / s (pulse width 3.5 ps, repetition interval 8 ps in the reported prototype: Kawanishi, etc .: International conference presentation
ECOC98 Post deadline paper 1, pp.41 120Gbit / s OTDM
In multiplexing for wideband transmission exceeding the system prototype), a signal error rate which cannot be ignored is deteriorated. Further, for example, when the optical circuits of a plurality of systems are set at different positions, even if the fiber length of each system can be unified to, for example, the same length of 4 m,
A timing fluctuation of 2 ps or more occurs when the environmental temperature difference is only about 2 degrees. This is the second problem to be solved for the timing fluctuation.

In order to suppress the influence of such a change in the length of the constituent fiber components, efforts have been made to suppress the timing shift and fluctuation by a technique called hybrid integration in which a long portion between elements is eliminated as much as possible (Takara) etc:
International conference presentation OFC99-FI2-1, pp.135 Integrated Optica
l Time-division multiplexer based on Planar Lightw
ave Circuit) is an advanced design and
It requires a production technology, and has not reached a low-cost, general-purpose technology.

Therefore, in an optical time-division multiplexing apparatus using short optical pulses, a simple circuit for observing (monitoring) a timing change of a signal during operation when an environment changes and correcting the signal has been strongly desired.

As a method for evaluating the timing and phase of such a pulse train, SAW (surface acoustic wave) is used at low speed.
Conventionally, a tank circuit using a wave (surface acoustic wave) filter and a PLL (phase-locked loop) circuit at a medium speed have been considered.

In Japanese Unexamined Patent Publication No. 8-237228, a PLL is provided in each branch stage 11 as shown in FIG.
A method has been proposed in which a phase detection circuit and a timing comparison circuit using the same are incorporated and frequency extraction and phase comparison are mutually performed. However, the provision of the phase comparison control circuit 15 including the PLL loop for each branch makes the time constant required for system stability redundant and complicates the control mechanism. In addition, the operating speed of the synchronous electric system of the normal PLL circuit is limited to about 10 GHz, and the baseband frequency f of the higher light source a (it is expected to use a light source of 40 GHz already in the future). Is difficult. Further, it is necessary to have a fixed repetition, the frequency of which is limited to the vicinity of a predetermined fixed value, and a band range that can follow. For this reason, Japanese Patent Application Laid-Open No. 8-237
The proposal of Japanese Patent Application Publication No. 228 has not been put to practical use because of the high cost due to the frequent use of high-speed electric circuits and the large-scale and complicated configuration.

As described above, the timing fluctuation (phase drift) of the above two types of pulses caused by the frequent use of long optical components is a problem that must be solved when handling a wideband optical signal. In addition, there is a demand for a simple, inexpensive, high-time-resolution, simple real-time delay evaluation unit and correction unit that does not depend on the repetition frequency used. Similar issues,
Obviously, it also occurs in optical measurement and monitoring systems that require high stability.

The present invention has been made in view of the above points, and has as its object to provide a long optical component which often becomes a problem in controlling a time lag of a pulse train which is essential in transmission and measurement of ultra-high speed and ultra-short pulses. (E.g., an optical fiber amplifier: EDFA), which compensates for fluctuations in signal arrival timing due to physical expansion and contraction in the signal propagation direction due to temperature changes and the like,
It is an object of the present invention to provide a technique for achieving high-precision stability in the time domain.

That is, an object of the present invention is to simply observe and correct a delay time difference between an optical pulse train serving as a time reference and a signal pulse train requiring phase synchronization without an expensive ultra-high-speed electric circuit. Another object of the present invention is to provide an optical pulse timing detection circuit and an optical time division multiplexer.

[0017]

In order to achieve the above object, the invention according to claim 1 comprises a reference light pulse train having a repetition frequency f ₀ / N serving as a time reference, and the same repetition frequency or N times as large as the reference light pulse train. Repeat the optical pulse timing detection circuit for inputting a signal light pulse train of frequency f _0, and branching means for branching said signal light pulse train inputs the output of one of said branching means, the signal optical pulse train and the reference optical pulse train Semi-fixed delay means for setting the time positional relationship so as to partially overlap, optical coupling means for multiplexing the output of the semi-fixed delay means and the reference optical pulse train, and receiving the output of the optical coupling means Optical pulse cross-correlation receiving means comprising: an optical receiver for observing an overlap component between the optical pulse trains; and an identification circuit for detecting an increase or decrease in the output of the optical receiver. It is characterized in.

Here, the optical receiver is a sum frequency light receiver including a generation unit of sum frequency light generated by overlapping of a plurality of pulses, and a receiver sensitive to the wavelength of the sum frequency light. The discrimination circuit may be characterized in that the change in the time positional relationship of the signal light pulse train is measured by using the increase or decrease of the signal output of the sum frequency light receiver.

The optical coupling means couples both the pulse trains of the reference light pulse train and the signal light pulse train in the same linear polarization direction. A sum type light generating material of type 1 to be generated and a receiver sensitive to a wavelength shorter than the wavelength of the input light, wherein the identification circuit generates sum frequency light by an overlapping component between two pulse trains to be compared. The method may be characterized in that a change in the time positional relationship of the signal light pulse train is measured using an increase or decrease in the output.

Further, the optical receiver uses a receiver having low sensitivity to the fundamental light wavelength and having high sensitivity to a shorter wavelength than the input light wavelength, and the discriminating circuit uses an overlap between two optical pulse trains to be compared. The method may be characterized in that a change in the temporal positional relationship of the signal light pulse train is measured by using an increase or a decrease in the two-photon absorption power due to the component.

The optical receiver generates an optical signal by an interaction between pulse trains having the same linear polarization direction.
A sum frequency generating material of the type and an avalanche photodiode receiver having a lower sensitivity to the fundamental light wavelength and a higher sensitivity to the shorter wavelength than the input light wavelength, wherein the identification circuit overlaps the two optical pulse trains to be compared. The variation of the time positional relationship of the signal light pulse train is measured using the sum of both the increase and decrease of the sum frequency light output and the increase and decrease of the two-photon absorption power due to the component.

Further, the optical coupling means uses a polarizing beam splitter having a plurality of input / output ports, and the reference light pulse train and the signal light pulse train enter from different surfaces of the polarizing beam splitter, and are linearly polarized orthogonal to each other. Type 2 sum frequency light that combines two pulse trains and generates a sum frequency light by phase matching when the polarization directions of the signal light pulse train and the reference light pulse train are orthogonal to each other. A generating material, a filter that transmits the sum frequency wavelength and blocks the fundamental wavelength, and has low sensitivity to the fundamental light wavelength,
A receiver having high sensitivity to a wavelength shorter than the wavelength of the input light, wherein the identification circuit detects an increase or decrease in the generation output of the sum frequency light due to an overlap component between the two optical pulse trains to be compared. Can be.

The branching means branches a part of the reference light pulse train and the signal light pulse train for monitoring.
A fixed output intensity optical amplifier for compensating for variations in the intensity of the signal light pulse train, and delay control means for outputting a delay control signal corresponding to the output signal value of the identification circuit can be provided. .

The signal light pulse train (optical frequency w
An optical pulse train having a repetition frequency f ₀ / N having an optical frequency (w2) different from that of 1) and synchronized in time is used as the reference optical pulse train, and the optical coupling means sets the reference optical pulse train and the signal optical pulse train on the same straight line. By combining in the polarization direction, both optical pulse trains are combined, and the optical receiver
A third-order nonlinear optical material that generates four-wave mixing signal light by interaction between two optical pulse trains, and a phase conjugate optical frequency ((2 × w1) of both without transmitting the reference optical frequency w1 and the signal optical frequency w2. -W2) or (2xw2-w1))
And a receiver having high sensitivity in the vicinity of a fundamental wavelength, wherein the identification circuit uses the increase or decrease of the output of the four-wave mixing signal light due to the overlapping component of the two optical pulse trains to be compared. A feature of the present invention is to measure a change in a time positional relationship of the signal light pulse train.

In order to achieve the above object, an optical time division multiplexing apparatus according to a ninth aspect of the present invention provides a branching means for branching a plurality of signal light pulse trains, and one of the repetitive light beams branched by the branching means. 9. A plurality of pulse signals according to any one of claims 1 to 8, wherein a pulse signal or a separately prepared pulse train is used as a reference light pulse train, and the timings of the signal light pulse trains are respectively detected using the remaining plural types of light pulse signals as the signal light pulse trains. An optical pulse timing detection circuit,
Delay control means for controlling the pulse timings of the remaining plurality of optical pulse trains, respectively, such that the optical pulse trains of the plurality of systems are arranged at equal intervals when the signal light pulse trains of the plurality of systems are time-multiplexed. An optical multiplexing unit that temporally combines the signal light pulse trains of the plurality of systems, the timings of which are controlled by the delay control unit, wherein the light pulse timing detection circuit includes the reference light pulse train and the signal light pulse train. And outputting a delay control signal for feedback-controlling a delay time to be given to each of the delay control means in accordance with a timing detection result of the optical pulse signals of the remaining plural systems detected by comparing the overlapping time positions between the delay control signals. It is characterized by doing.

Here, only one optical pulse timing detection circuit is provided in one system, and a selection timing control for generating a timing signal for sequentially selecting the plurality of optical pulse trains branched by the branching means. And a branch of the optical pulse trains of the plurality of systems.
An optical channel selection switch that selects one of the inputs, and sequentially selects the plurality of optical pulse trains branched by the branching unit in a time-division manner through the optical channel selection switch to the optical pulse timing detection circuit. Supply,
The output of the optical pulse timing detection circuit is distributed as a delay control signal to the delay control means of each target series by an electric signal selection switch interlocked with the optical channel selection switch, and is sequentially fed back to the delay control means. Can be characterized.

[0027] To achieve the above object, the present invention is, repetition frequency f ₀ / and the reference optical pulse train of N, repetition frequency f ₀ having the same repetition frequency or its N times and the reference optical pulse train as the time reference of claim 11 An optical pulse timing detection circuit for inputting the signal light pulse train, a branching means for branching the signal light pulse train, and a light for separating the reference light pulse train into two reference light pulse trains having orthogonal polarization directions and a certain delay time difference. The output of one of the splitting means and the branching means is input, and the time position of the signal light pulse train to be synchronized is set to be intermediate between the time positions of the two reference light pulse trains separated by the light separating means. A semi-fixed delay unit, wherein one of the two reference optical pulse trains separated by the optical separation unit and the delay process performed by the semi-fixed delay unit is performed. A part of the optical power of the signal light pulse train is multiplexed, and the other of the two reference light pulse trains separated by the light separating means and the rest of the light power of the signal light pulse train delayed by the semi-fixed delay means Optical cross-correlation receiving means, two optical pulse cross-correlation receiving means for receiving the output of the light synthesizing means and independently observing the overlap component between the respective optical pulse trains, and the two optical pulse cross-correlation receiving means And a signal comparing means for measuring a change in the time position of the signal light pulse train before and after by detecting an increase or decrease in the output of the signal light pulse train.

Here, the light separating means may be characterized in that a pulse is incident in a polarization direction of 45 degrees with respect to the main axis of the optical material having birefringence or circular polarization equivalent thereto.

Further, the two optical pulse cross-correlation receiving means each include a sum frequency light generating section generated by overlapping a plurality of pulses, and a sum frequency light including an optical receiver sensitive to the wavelength of the sum frequency light. The signal comparing means may be characterized in that the signal comparing means measures a change in a time positional relationship between the signal light pulse trains by increasing and decreasing signal outputs of the two optical receivers.

The light combining means comprises a polarization maintaining type optical coupling means having two input ports and one or two output ports, and a polarization separating means having one or two input ports and two output ports. And the two reference light pulse trains, whose polarization directions are orthogonal and have a certain delay time difference, are combined with one another so that the orthogonal polarization direction coincides with the main polarization axis direction (X-axis direction and Y-axis direction) of the light combining means. Input from the input port, the signal light pulse train is transmitted from the other input port so that the polarization direction is linearly polarized light of about 45 degrees with respect to the main polarization direction of the polarization maintaining type light combining means or circularly polarized light equivalent thereto. Incident, take out the combined light of the reference light pulse train and the signal light pulse train from the output port, and
The input light is input from the input port such that the polarization direction of the reference light pulse train coincides with the main polarization direction (the X′-axis direction and the Y′-axis direction) of the polarization splitting means. It can be characterized in that the directional components are separated and taken out respectively from two different output ports.

The optical pulse cross-correlation receiving means includes an avalanche photodiode receiver having low sensitivity to a fundamental light wavelength and high sensitivity to a shorter wavelength than an input light wavelength,
The signal comparing means may measure a change in a temporal positional relationship of the signal light pulse train by using an increase or a decrease in two-photon absorption power due to an overlapping component of two pulses to be compared.

Also, the optical pulse cross-correlation receiving means is a type 1 sum frequency light generating material that generates an optical signal by the interaction between pulses in the same linear polarization direction, has low sensitivity to the fundamental light wavelength, A receiver having a higher sensitivity to a shorter wavelength than the optical wavelength, wherein the signal comparing means increases or decreases the sum frequency optical output by the overlapping component of the two optical pulse trains to be compared.
The method may be characterized in that a change in the temporal positional relationship of the signal light pulse train is measured using the sum of both the increase and decrease in the photon absorption power.

The light combining means includes polarization combining means having two input ports and two output ports, and converts the two reference light pulse trains having the orthogonal polarization directions and a certain delay time difference into orthogonal polarization directions. Is incident from one of the input ports so as to coincide with the main polarization axis direction (the X-axis direction and the Y-axis direction) of the polarization synthesizing means, and the signal light pulse train is
The polarization direction is incident from the other input port such that the polarization direction becomes linearly polarized light of about 45 degrees with respect to the main polarization direction of the polarization synthesizing means, or circularly polarized light equivalent thereto. By taking out from the two output ports of the polarization combining means, the combining of the reference pulse train and the signal pulse train and the combining and separating between the light pulses of two different polarization directions can be performed simultaneously.

Further, the optical pulse cross-correlation receiving means includes:
A type 2 type sum frequency light generating material for generating a sum frequency light by phase matching when the polarization directions of the reference light pulse train and the signal pulse train are orthogonal to each other; A wavelength filter that blocks an optical wavelength, and an optical receiver that is less sensitive to the fundamental light wavelength and more sensitive to a shorter wavelength than the input light wavelength, and wherein the signal comparing means uses an overlapping component of two optical pulse trains to be compared. It can be characterized by detecting an increase or decrease of the sum frequency light.

The signal light pulse train (optical frequency w
An optical pulse light source having a repetition frequency f ₀ / N having an optical frequency (w2) different from that of 1) and synchronized in time is used as the reference optical pulse train, and the light synthesizing means combines the reference pulse train and the signal pulse train into two signals. The pulses are combined in the same linear polarization direction, and the two optical pulse cross-correlation receiving means each receive 4 pulses by the interaction between the two incident optical pulse trains.
A third-order nonlinear optical material that generates lightwave mixed signal light, and a phase conjugate lightwave frequency ((2 × w1−w2) or (2) of the two without transmitting the reference light frequency w1 and the signal light frequency w2.
× w2-w1)) and a light receiver having high sensitivity in the vicinity of the fundamental wavelength, and the signal comparing means includes a four-wave mixing signal light based on an overlapping component of two light pulse trains to be compared. The timing discrimination means compares the change of the overlapping component with each pulse train by measuring the change in the time positional relationship of the signal light pulse train by using the increase or decrease of the output of the signal light. A parallel optical receiver in which electrical outputs of two optical pulse cross-correlation receiving means are connected in parallel, and a difference between outputs of the two parallel optical receivers is used for discriminating a time position fluctuation direction.

The timing discriminating means is used for discriminating the direction of time position fluctuation by taking the difference or fraction of the electric output of the two optical pulse cross-correlation receiving means for comparing the fluctuation of the overlapping component with each pulse train. The average of the signal power is estimated by calculating the sum of the electric outputs, and both the time position direction and the fluctuation amount are detected by dividing both of the electric outputs.

To achieve the above object, the invention of claim 22 comprises a delay control means for controlling the timing of each of optical pulse trains of M systems (M is an integer of 2 or more) having a repetition frequency f ₀ , The optical time-division multiplexing apparatus for generating an optical pulse train of a repetition frequency f ₀ by multiplexing the optical pulse trains of the M systems at equal intervals on the time axis using an optical multiplexing means for multiplexing the optical pulse trains of A branching means for branching the optical pulse trains of the respective systems, and one of the optical pulse trains branched by the branching means or an optical pulse train prepared separately is used as a reference optical pulse train, and the timing of the optical pulse trains of the remaining systems is detected. 22. An optical pulse timing detection circuit according to claim 11, wherein the optical pulse timing detection circuit detects a change in a time position of each of the M-system optical pulse trains. The delay control means performs feedback control in accordance with the detection result so as to remove the time position fluctuation of each system.

Here, the optical pulse timing detection circuit is provided in only one system, and a selection timing controller for generating a timing signal for sequentially selecting the branched plural optical pulse trains; An optical channel selection switch that accommodates a plurality of optical pulse trains, sequentially selects the plurality of optical pulse trains in a time-division manner using the timing signal of the selection timing controller, and supplies the optical pulse train to the optical pulse timing detection circuit; The output of the optical pulse cross-correlation receiving means is distributed as a delay control signal to each target stream by an electric signal selection switch interlocked with the optical channel selection switch, and is sequentially fed back to the corresponding delay control means. Can be characterized.

(Operation) In the first embodiment of the present invention, the amount of displacement from the reference position and its direction can be specified by increasing or decreasing the overlapping component of the reference light pulse train and the signal light pulse train at the measurement position. The delay amount to be corrected and the correction direction (whether to increase or decrease the delay) can be specified with high accuracy by evaluating only one signal value caused by the optical nonlinear effect. According to the present invention, the time position between pulse trains can be detected at 1/10 or less of the pulse width for a high-speed repetitive pulse. However, since it is sufficient to read the average power of the nonlinear signal obtained from each repetitive signal, high-speed optical reception is possible. It does not need to be composed of a container or an electric circuit. Therefore, the timing detection accuracy does not depend on the response speed of the optical receiver and the electric circuit to be configured. The accuracy is very high because it depends on the optical response of a non-linear material that generates a sum frequency and a light receiving element that generates two-photon absorption power. An optical pulse train used for wider band optical transmission has a shorter pulse width, and has expandability such that nonlinear sensitivity and time resolution are further improved.

Further, in the second embodiment of the present invention, the amount of displacement from the reference position and the direction thereof are specified by increasing / decreasing the mutually overlapping components of the reference light pulse train and the signal light pulse train separated into two at the measurement position. it can. The delay amount to be corrected and the correction direction (whether to increase or decrease the delay) can be specified with high accuracy only by comparing the two cross-correlation signal values. Even if the signal strength of the pulse train to be evaluated fluctuates due to a change in the operation mode and each received photovoltaic power fluctuates, no error in timing detection occurs because the relative power of the two receivers is compared. Time position between pulse trains is 1/10 of light pulse width
Although it has a high time resolution that can be separated in the following, it is only necessary to read the average power of the nonlinear signal obtained from each of the repetitive signals. Therefore, the timing detection accuracy does not depend on the response speed of the optical receiver and the electric circuit to be configured. The accuracy is very high because it depends on the optical response of a non-linear material that generates a sum frequency and a light receiving element that generates two-photon absorption power. An optical pulse train used for wider band optical transmission has a shorter pulse width, and nonlinear sensitivity and time resolution are further improved.

[0041]

Embodiments of the present invention will be described below in detail with reference to the drawings.

(First Embodiment) FIG. 1 shows a principle configuration of an optical pulse timing detection circuit according to a first embodiment of the present invention. This optical pulse timing detection circuit passes a signal short pulse train generating circuit 21 for generating a short pulse train (a signal light pulse train having the same repetition frequency as the reference light pulse train or a repetition frequency f ₀ times N times), and the generated signal light pulse train. Long optical path & parts 22 subject to temperature fluctuation
A reference light pulse train generation circuit 23 for outputting a reference light pulse train having a repetition frequency f ₀ / N as a time reference from the generated signal light pulse train, and a signal light having a frequency f ₀ passing through the long optical path & component 22 A signal pulse branching unit (optical branching unit) 24 for branching a pulse train, and a semi-fixed delay which receives one output of the signal pulse branching unit 24 and sets the reference light pulse train and the signal light pulse train so as to partially overlap in time position. Part 2
5, an optical coupling unit 26 that combines the output of the semi-fixed delay unit 25 and the reference optical pulse train from the reference optical pulse train generation circuit 23.
A sum frequency light receiver 27 that receives an output of the optical coupling unit 26 and observes an overlap component between the optical pulse trains, and a signal identification circuit 28 that detects an increase or decrease in the output of the sum frequency light receiver 27, According to the detection result of the signal identification circuit 28, the frequency f ₀
When it is necessary to adjust the delay of the signal light pulse train, the optical delay control circuit 31 disposed between the long optical path & component 22 and the signal pulse branching unit 24 controls the delay. As a result, a timing stabilizing pulse is obtained from the other output of the signal pulse branching unit 24. The optical coupling section 26, the sum frequency optical receiver 27, and the signal identification circuit 28 constitute an optical pulse cross-correlation receiving section 29.

In the present timing detection circuit, an optical nonlinear material that generates more optical output (power in the case of two-photon absorption) when a plurality of pulses are superimposed is added to the sum frequency optical receiver 27.
Have inside. If the signal pulses do not overlap in time with the reference pulse, the sum frequency light is not enhanced and the output of each receiver is proportional to the sum of the squares of each pulse signal strength.

Since a pulse signal in which a plurality of pulses are superimposed has an extremely high peak intensity, conversion into sum frequency light (SHG (second harmonic generation) light when both pulses have the same wavelength) is performed with high efficiency. Is The sum frequency light generated by the optical nonlinear material is absorbed by the short wavelength receiver and converted into an electric signal. Further, when the incident light peak intensity is high in the semiconductor, light having a wavelength without light receiving sensitivity (in the case of basic optical communication, a basic wavelength of 1.5 μm) is incident due to the two-photon absorption phenomenon and up to twice the photon energy. The electron-hole pairs are excited to generate photovoltaic. Although this phenomenon does not necessarily generate double harmonic light, the current at the photocathode of the receiver increases in proportion to the square of the incident peak power, so that the identification performance of the present receiver can be improved. Since the receiver has no sensitivity at the fundamental wavelength, it does not become a noise signal. for that reason,
A high signal-to-noise (S / N) ratio is obtained, and the receiver can accurately evaluate the timing overlap state between the ultrashort pulses.

In the first embodiment of the present invention, the repetition frequency cannot be specified unlike a high-speed photoelectric conversion receiver or a PLL circuit using an electric circuit. When synchronizing at a frequency f or an integer multiple fM, an integer fraction f / M,
The optical pulse cross-correlation receiver 29 optically superimposes the pulses (cross-correlation),
It becomes a very accurate time position detection unit.

Further, since the present optical pulse timing detection circuit does not use an electric resonance circuit, any bit rate can be used as long as pulses can be synchronized and compared. Conversely, in a closed optical pulse multiplexing transmission device (optical time division multiplexing device), the work of re-detecting the known information of frequency is omitted, and only the information of the time position (phase within a bit) of the repetitive pulse train is stored. Specializes in optical detection in relation to reference light. Thus, a highly accurate and flexible optical detection unit can be obtained.

Therefore, if the present optical pulse timing detection circuit is used, an ultra-high repetition optical time multiplex system, etc.
Simple and accurate time position detection is possible even in a broadband application field which is difficult with an electric circuit, and both of the two types of time delay fluctuation problems described in the section of the related art can be solved.

(Second Embodiment) FIG. 2 shows the principle configuration of an optical pulse timing detection circuit according to a second embodiment of the present invention. In this optical pulse timing detection circuit,
Signal short pulse train generation circuit 21, long optical path & parts 2
2. The reference pulse train generating circuit 23, the optical branching unit 24, the semi-fixed delay giving unit 25, and the optical delay control circuit 31 have the same configuration and function as those of the first embodiment of the present invention shown in FIG. Therefore, detailed description thereof is omitted. The configuration of the optical pulse cross-correlation receiving unit 32 is different from that of the optical pulse cross-correlation receiving unit 29 in FIG.

The optical pulse cross-correlation receiving section 32 of this embodiment
Has an optical pulse separating unit 33, an optical pulse combining unit 34, a separating unit 35, sum frequency optical receivers 36 and 37, and a signal comparing unit 38 for timing identification. Optical pulse separation unit 33
Separates a reference pulse train having a repetition frequency serving as a time reference into two light pulses (AB in the figure) having orthogonal polarization directions and a certain delay time difference. The optical pulse combining unit 34 determines that the time position of the signal light pulse (or reference light pulse) sequence to be synchronized input from the semi-fixed delay imparting unit 25 is such that the separated polarization directions are orthogonal to each other when the optical receiver is incident and the fixed delay time difference , One of the separated reference light pulse trains and a part of the optical power of the signal light pulse train are multiplexed, and the other of the separated reference light pulse trains is set. And the rest of the optical power of the signal pulse train. The separating unit 35 separates the optical pulse train into two in order to independently observe the overlap component between the optical pulse trains.

The two sum frequency optical receivers 36 and 37 independently observe the overlap components between the separated optical pulse trains. The signal comparing unit 38 detects an increase or decrease in the output of the two sum frequency optical receivers 36 and 37 by comparing the signals, and identifies the timing by observing the fluctuation before and after the time position of the signal light pulse train. As a result of this timing identification, the timing of the signal light pulse train is adjusted via the optical delay control circuit 31 if necessary.

In the optical pulse timing detection circuit of the present embodiment, the optical pulse cross-correlation receiving unit 32 generates two optical outputs (electric power in the case of two-photon absorption) when a plurality of pulses overlap. A non-linear material is included in the sum frequency optical receivers 36,37. If the signal pulses do not overlap in time with the reference pulse, the sum frequency light is not enhanced and the output of each receiver is proportional to the sum of the squares of each pulse signal strength.

Since a pulse signal in which a plurality of pulses are superimposed has a very high peak intensity, conversion to sum frequency light (SHG when both pulses have the same wavelength) is performed with high efficiency. The generated sum frequency light is transmitted to the short wavelength receivers 36, 3
7 absorbs light and is converted into an electric signal. Furthermore, when the incident light peak intensity is high in the semiconductor, even if light of a fundamental wavelength (1.5 nm in the case of basic optical communication) having no light receiving sensitivity due to the two-photon absorption phenomenon is incident on the photocathode, The electron-hole pairs are excited to twice the photon energy of the wave, producing photovoltaics. This phenomenon does not generate sum frequency light, but the current on the photocathode of the receivers 36 and 37 increases in proportion to the square of the incident power.
The identification performance of the receivers 36 and 37 can be improved. Since the receivers 36 and 37 have no sensitivity at the fundamental wavelength, they do not become noise signals, so that a high S / N ratio is obtained,
The receiver can accurately evaluate the timing overlap state between the ultrashort pulses.

In this embodiment, as in the first embodiment, a PL using a high-speed photoelectric conversion receiver or an electric circuit is used.
As in the case of the L circuit, the repetition frequency cannot be specified. However, when the reference pulse and the signal pulse are synchronized at the same repetition frequency f or an integer multiple of fM, or a fraction of an integer, the pulses are synchronized. By detecting the superposition (cross-correlation) of optically, a very accurate time position detecting unit can be obtained.

Further, the optical pulse timing detection circuit of the present embodiment does not use an electric resonance circuit, as in the first embodiment, and therefore has the same configuration as long as the pulses can be compared in synchronization. Can support any bit rate. Conversely, in the closed apparatus, the work of re-detecting the information of the known frequency, which is known, is omitted, and only the information of the timing (phase within a bit) of the repetition pulse is optically related to the reference light obtained by dividing the information into two. Specializing in detecting. Thus, a highly accurate and flexible optical timing detection unit can be realized.

Therefore, if the optical pulse timing detection circuit of the present embodiment is applied, as in the first embodiment, a simple and highly accurate time can be obtained even in a wide band which is difficult in an electric circuit such as an optical time multiplexing system with an extremely high repetition. Position detection is possible. Any of the above two types of time delay fluctuation can be solved.

By the way, if the degree of coincidence between the reference pulse and the signal pulse is simply checked, it is sufficient to monitor the increase / decrease of the overlap component between the single pulses as in the first embodiment. If there is any deviation from the timing due to output fluctuation due to pulse power fluctuation, it is necessary to instantly understand the amount and direction of the deviation and send an instruction to correct it to the control circuit in order to correct the timing. . For this purpose, a signal indicating that the delay has increased (timing is delayed) and a signal indicating that the delay has decreased (timing has advanced) are simultaneously provided, and the signals are compared to determine an optimum value. Information to give is desired. If such a combination and such information can be provided, information on the amount of delay and information on the direction can be simultaneously determined, and a circuit that always controls the stable point becomes easy.

In order to easily obtain an overlap correlation signal of each pulse before and after such a desired timing, as shown in FIG. 3, the reference signal has an optically constant delay difference (polarization 2) The second embodiment of the present invention, in which a pulse train is divided into two separable pulse trains and a signal pulse train is interposed therebetween, is effective.

An example in which this embodiment is effective will be described. FIG.
As shown in (A), in a configuration using only one second-order optical nonlinear phenomenon, a double change in the input light intensity becomes a four-fold change in the received signal. Even if the pulse peak intensity does not change, the same effect cannot be avoided even if the repetition (duty ratio) changes. Regarding the optical nonlinear signal, the signal output of the sum frequency optical receiver fluctuates greatly when the incident average optical power fluctuates, even if the amount of pulse overlap (that is, the temporal positional relationship between the two) does not change. Even if the timing relationship between pulses and the average power are constant, if the pulse quality (time width, tailing, peak intensity) fluctuates, the cross-correlation component will increase or decrease.
The output value of each receiver fluctuates.

On the other hand, as in the second embodiment of the present invention shown in FIG. 3B, for example, when a pulse polarization is obliquely incident on the main axis of an optical crystal having birefringence, the polarization direction Therefore, the pulse incident on each polarized light is separated and output into two pulses having a birefringence difference in the crystal and a time delay difference proportional to the length. If the two separated pulses are used as reference light pulses and the time position relationship with the signal pulse is compared, when the cross-correlation component with the previous pulse A increases, the cross-correlation component with the rear pulse B decreases at the same time. The two cross-correlation signals can be easily separated by polarization. If the difference of the sum frequency optical signal due to the two cross-correlations showing the contradictory changes or the division value is taken, the influence of the output fluctuation of the individual receiver irrelevant to such timing fluctuation can be canceled.

[0060]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, specific embodiments of the present invention will be described in detail with reference to the drawings. First to fourth examples will be described with reference to FIGS. 4 to 9 as specific examples of the first embodiment of the present invention in FIG. 1, and then the first to fourth examples of the present invention in FIGS. 2 and 3 will be described. As specific examples of the second embodiment, the fifth to seventh examples will be described with reference to FIGS.

(First Embodiment) FIG. 4 shows the configuration of a first embodiment of the time multiplex control device of the present invention. Here, an example in which the former (first problem) of the two pulse time position fluctuation problems described above in the column of the problem to be solved by the invention will be described. The first problem is a large fluctuation of the pulse timing before the signal modulation unit due to the insertion of the pulse width and intensity adjustment unit due to the long fiber component of about 100 m.

In this example, a semiconductor mode-locked laser having a repetition frequency f ₀ electrically synchronized with the signal light is used as the input signal for the reference light pulse train generating circuit 23. As the optical coupling unit 26 that introduces the signal light pulse train 42 and the reference pulse train 43 to the sum-of-round optical receiver 27 of the optical pulse cross-correlation receiving unit, a polarization maintaining fiber coupler or the like is used.

As the sum frequency optical receiver 27 for observing the overlapping component (sum frequency), a dielectric material such as a dielectric material KTP (KTiOPO ₄ ) or Li
A semiconductor thin-film avalanche photodiode (APD) or a photomultiplier tube (PM) optical receiver 27-3 is used as a receiver using NbO ₃ , AANP (2-adamantylamino-5-nitropyridine) crystal 27-1 or the like. I am preparing. An extremely compact sum frequency optical receiver 27 is realized by sandwiching a condenser lens 27-2 between the dielectric crystal 27-1 and the semiconductor receiver 27-3 and bonding these components together.

An increase or decrease in the reception signal of the sum frequency optical receiver 27 is detected by a signal identification circuit 28, and the detection result is fed back to an optical delay control circuit (optical path length controller) 31 as a delay control signal. The optical pulse timing is adjusted and controlled to the optimum delay position.

In FIG. 4, another electrically synchronized light source (reference light pulse train generating circuit) 23 is used as a reference pulse light source. Of course, the pulse train divided by (not shown) can be used as it is as a reference pulse train.

Most of the pulse signals incident on the sum frequency optical receiver 27 according to the present invention are converted into sum frequency signals in the optical crystal 27-1. The optical system 2 converts the portion of the optical crystal 27-1 that could not be converted into a sum frequency signal into a focus on a light receiving surface of a receiver 27-3 transparent to the fundamental wavelength.
Set 7-2.

In a semiconductor optical receiver 27-3 such as an APD or a pin receiver, there is essentially no sensitivity to an optical signal having an energy (longer wavelength) lower than the band gap energy. However, when the signal peak intensity is extremely high, a photocurrent is generated near the receiver active layer due to a two-photon absorption phenomenon. This light receiving current also increases by the square of the peak intensity of the fundamental wavelength signal as in the reception of the sum frequency signal, and thus contributes to the correlation receiving sensitivity due to the overlap signal between pulses. First
In the embodiment of the present invention, since the reference pulse train and the signal pulse train are multiplexed by the polarization directions that match each other, the type 1
KTP, LiNbO, which are sum frequency generating optical crystals of
_The signal sensitivity can be obtained by adding the two-photon absorption power and the sum frequency signal without distinction by using a _three crystal or the like and re-focusing the light on the photoelectric surface of the APD receiver.

FIG. 5 schematically shows the fluctuation of the output of the sum frequency optical receiver 27 when the time position of the signal pulse train moves forward and backward with the case where the overlap between the reference pulse trains is half as a reference point. The optimal time position of the delay reference signal is set at the intersection (the JUST point in the figure) of the broken line in FIG.
Set a threshold value of 8. When the output (receiver power) of the receiver 27 decreases from the set value of the threshold value, the signal identification circuit 28 controls the optical delay control circuit (variable delay circuit) to increase the overlapping component of the reference pulse train and the signal pulse train. When the instruction is issued to the optical delay control circuit 31 and the threshold value is increased, the instruction is issued to the optical delay control circuit 31 in a direction to reduce the overlapping component of the reference pulse train and the signal pulse train. in this way
The signal discrimination circuit 28 issues a command for an operation which is the reverse of the case of the increase when the increase is made. Thus, maintaining the optimum output value leads to maintaining the stable delay position.

The pulse time resolution of the measuring apparatus using the sum frequency is generally about half a half width of several hundred femtoseconds (however, it slightly varies depending on the thickness of the crystal used).
However, when the signal discrimination circuit 28 discriminates the increase or decrease of the overlap between the pulses, it is only necessary to evaluate the variation instead of the absolute value, so that a high resolution of 100 femtoseconds or less is possible. FIG.
If the time position between the pulses is far away or the overlap position is reversed due to a position shift exceeding the detection range (high-precision detection range) shown in the graph, instantaneous detection of the correct position becomes difficult. However, the present invention is intended to perform a correction within a range where signal quality can be maintained before deterioration becomes noticeable, assuming a slow drift from the set position from the beginning, so that high accuracy is required. For example, the detection range is sufficient in a time region about several times the pulse width.

(Second Embodiment) The above-described method of detecting and stabilizing the optical pulse timing according to the present invention also solves the second problem of time position fluctuation described above in the section of the problem to be solved by the invention. it can. A specific countermeasure in this case will be described below as a second embodiment of the present invention. The second problem is that the optical pulse output from the signal short pulse train generating circuit 1 in FIG. Are modulated by the modulator 5 to obtain optical pulses b, c, d... Of a plurality of systems. Here, due to differences in the optical path length, arrangement, temperature environment, etc. of the components of each system of about several meters, the time positional relationship of the multiplexed pulse signals that should be arranged at equal intervals in the optical multiplexing unit 9 is mutually different. The problem is that it fluctuates.

FIG. 6 shows the configuration of the time multiplexing control apparatus according to the second embodiment of the present invention. The time multiplexing control device of this example includes a signal short pulse train generator 21, an optical amplifier (EDFA) 22-
2. Optical branching unit 48, modulator 49, delay adjuster 50, optical branching unit 51, pigtail 52, optical coupler 53, sum frequency optical receiver 54, semi-fixed delay unit 55, variable delay circuit 56, optical multiplexing unit 57, and an optical pulse timing detection circuit 58.

The optical pulse output from the signal short pulse train generation circuit 21 passes through an optical amplifier (EDFA) 22-2,
The light is branched into a plurality of light beams by an optical branching unit 48, and each of the branched light pulses is modulated by a modulator 49 according to transmission information, so that light pulses b, c, d,... As described above, the optical branching unit 48 branches into a plurality of optical pulses, and the optical pulse timing detection circuit 58 detects the timing of each optical pulse. The group of the optical branching units 51 and the semi-fixed delay units 55 for the number of branches (M) uses one optical pulse train among a plurality of systems guided from the optical branching unit 48 at the preceding stage as a reference pulse train, and the remaining M-1 systems An optical pulse train branched from is used as a signal pulse train. The optical coupler 53 superimposes each of the reference pulse train and the optical pulse train and sends the resultant to the sum frequency optical receiver 54. Sum frequency optical receiver 5
4 is a sum frequency generation crystal 54-1 and a lens 54-
2. An optical receiver 54-3. The optical pulse timing detection circuit 58 is a sum frequency optical receiver 54 for the number of branches (M).
Based on the cross-correlation signal output from, the timing of each optical pulse of the plurality of systems branched by the optical branching unit 48 is detected. Each of the variable delay circuits 56 is inserted into each optical pulse path so as to be arranged at equal intervals when the optical pulses of a plurality of systems are time-multiplexed, and is added to the position detection signal output from the optical pulse timing detector 58. By setting the delay time of the optical pulse accordingly, feedback control is performed so that the optical pulse trains of the plurality of systems are time-multiplexed at equal intervals.

The optical pulse timing detection circuit 58, as shown in FIG. 5 described above, sets a proper position (voltage value) of the cross-correlation signal (error voltage) characteristically received by the time position.
Is set, and the operation direction of the variable delay circuit 56 is reversed when the threshold value is equal to or higher than the threshold value. Since the system of the present example can operate with such a simple instruction system, a discriminating IC for identifying each signal level without using a computer is required.
(Integrated circuit) is sufficient, and the control system can be configured very simply and inexpensively.

The sum frequency light receiver 54 of the present embodiment for comparing the overlap components between the reference and signal pulses is composed of a KTP crystal 54-1 for generating sum frequency light and an APD receiver for amplifying two-photon absorption power. Since a high photon density can be obtained with both the photocathode 54-3, the lens 54-2 is focused twice, and high receiving sensitivity can be achieved at low cost.

The relative positional relationship between the reference and signal pulses after passing through the optical multiplexer 53 is uniquely determined. Thereafter, even if the arrival timing at the sum frequency optical receiver 54 via the redundant optical circuit drifts, the relative positional relationship between the reference and the signal pulse does not change. Therefore, as means for further increasing the receiving sensitivity, an optical amplifier (not shown) is provided with an optical coupler (optical coupler) 53 and a sum frequency optical receiver (SHG element) 5.
4 can be inserted.

(Third Embodiment) Further, as the configuration of the pulse timing detection circuit, a configuration that is more efficient than the first embodiment of the present invention can be considered. As a specific example, FIG. 7 shows a configuration of an optical time division multiplexing apparatus according to a third embodiment of the present invention and an example of signal output. In the present embodiment, the signal short pulse train generation circuit 21 is provided with an original pulse timing drift control system having substantially the same configuration as that of FIG. 4 of the first embodiment, and a multiplex having substantially the same configuration as that of FIG. 6 of the second embodiment. An inter-pulse timing control system is connected in series. The original pulse timing drift control system includes an optical branching unit 60, a long component 61 such as an EDFA, a variable delay circuit 62, an optical branching unit 63, a semi-fixed delay unit 64, a delay control circuit 65, a sum frequency optical receiver 66, and It has an optical multiplexer (PBS) 67. The sum frequency optical receiver 66 includes a sum frequency generating crystal 66-1, an optical filter 66-2, and an optical receiver 66-3. The multi-pulse timing control system includes an optical splitter 68, a modulator 69, a delay adjuster 70, an optical splitter 71, a pigtail 72, and an optical multiplexer (PBS) 7.
3, a sum frequency optical receiver 74, a semi-fixed delay unit 75, a variable delay circuit 76, an optical multiplexing unit 77, and an optical pulse timing detection circuit 78. The sum frequency optical receiver 74 includes a sum frequency generating crystal 74-1, an optical filter 74-2, and an optical receiver 74-.
3 is provided.

In the third embodiment, the optical branching unit 60,
The reference light pulse train branched by 71 and the signal light pulse train branched by 63 and the like are combined into an optical multiplexer (PBS) 6.
In 7 and 73, since light is multiplexed in polarization directions orthogonal to each other, as a means for generating sum frequency light, a type 2 type crystal (for example, AANP) that generates a sum frequency with two waves having polarization directions different from each other is used as a sum. High frequency optical receiver 66, 74
Can be used. In that case, the optical multiplexers 67 and 73
, The monitor signal having a higher signal-to-noise ratio can be obtained at the same time than the cutoff of the two-photon absorption signal.

In the first embodiment shown in FIG. 4, if the peak power of the light that matches one of the polarization directions is sufficiently large, the second harmonic light is generated even if the two pulses do not overlap. Therefore, as shown in FIG. 5, even when there is no cross-correlation component, the signal of the sum frequency optical receiver is observed, and this fluctuation becomes the background noise that degrades the accuracy of the comparison signal and the timing control signal.

However, in the configuration of the third embodiment,
A sum frequency is generated for the first time as an interaction between two orthogonal pulses, and a single-polarized signal alone does not generate a sum frequency optical signal. Therefore, when there is no overlapping component between pulses,
Noise due to the second harmonic light generated for each single pulse, which cannot be ignored in Type 1, is remarkably suppressed, and high S
/ N ratio can be detected.

FIG. 8 shows the relationship between the overlapping time position relationship between the reference / signal pulse trains and the output power of the receiver in the configuration of the third embodiment. In this configuration, since the fundamental light is cut by the filters 66-2 and 74-2, the optical receiver 66-
No. 3, 74-3 does not need to consider the effect of two-photon absorption. Therefore, a light receiving element such as a photomultiplier tube having a smaller quantum efficiency than the APD but having an extremely large electron multiplication factor can be used as the optical receivers 66-3 and 74-3.
Further, in this configuration, when combining the reference pulse train and the signal pulse train with the PBSs 67 and 73, different polarizations orthogonal to each other are used, so that the coupling loss can be reduced to 0 in principle. The optical power loss is smaller than that of the embodiment, and a high S / N ratio can be obtained. With such a clever structure of components, a receiver having a high signal-to-noise ratio as shown in the third embodiment is realized for the first time.

In each of the first and third embodiments, the signal indicating the time delay of the received signal is photovoltaic power generated by an ultra-high-speed optical pulse, but the positional relationship between the repeated pulse trains is bit by bit. Does not fluctuate, the optical receiver 6
The response bands of 6-3 and 74-3 may be low. That is,
The response band of the optical receivers 66-3 and 74-3 is sufficient to be 100 Hz or less such that the variation of the long-term average value of the signal is observed and the signal is output to the optical path length control circuits (variable delay circuits) 62 and 76. It is. Accordingly, the optical receiver 6 is a low-cost but inexpensive component such as a photomultiplier tube or an APD receiver having high reception sensitivity.
6-3 and 74-3 can be easily configured.

Optical Nonlinear Material 6 Used for Sum Frequency Light Generation
In 6-1 and 74-1, in order to obtain high conversion efficiency, it is necessary to propagate in the crystal for a long time. Generally, in the D33 direction, which is the most efficient plane orientation, a rapid phase mismatch occurs due to mismatch of the propagation constant, and it is difficult to make the interaction long. Therefore, a long interaction length can be maintained while maintaining high efficiency by periodically correcting the group velocity dispersion by a process such as ion implantation. By using such a quasi-phase matching (QPM) to form an optical waveguide having good confinement for incident light, higher interaction efficiency and light efficiency can be obtained than in the case of using the plate crystal described in the first embodiment. Density is obtained. By using such a non-linear element, it is possible to provide sorted optical receivers (sum frequency optical receivers) 66 and 74 having good mode coupling efficiency with optical fibers used in a network and good receiving sensitivity.

(Fourth Embodiment) FIG. 9 shows the configuration of a fourth embodiment of the present invention. Also in the present embodiment, the method of monitoring the time fluctuation of the signal pulse train by using the reference pulse train is the same as in the first to third embodiments.

The optical time division multiplexing apparatus of this embodiment comprises a signal short pulse train generation circuit 21, an optical amplification circuit (EDFA) 80, a branch unit 81, a modulator 82, a delay adjuster 83, and an optical branch unit 8.
4, pigtail 85, semi-fixed delay unit 86, optical channel selection circuit 87, fixed output optical amplifier 88, optical coupling unit 8
9, a sum frequency optical receiver 90, a signal comparing section 91, an electric selection circuit 92, a variable delay circuit 93, a delay control circuit 94, a selection timing controller 95, and an optical multiplexing section 96.

The optical pulse cross-correlation receiving section 9 includes an optical coupling section 89, a sum frequency optical receiver 90, a signal comparing section 91 for comparing an input value with a threshold value, and a selection timing controller 95.
7 is configured. The optical pulse cross-correlation receiving unit 9
The optical pulse timing detection circuit 98 is composed of the optical branching section 84, the pigtail 85, the semi-fixed delay section 86, the optical channel selection circuit 87, the fixed output optical amplifier 88, the electric selection circuit 92, and the delay control circuit 94.

In the case where individual pulse timing detection circuits are provided for each of a plurality of systems as in the second embodiment, when the number of branches and the configuration scale are increased, the optical sensitivity and fiber length between the detection devices are increased. With respect to errors, variations in electrical thresholds, etc., individual differences and an increase in the number of parts become problems. Specific means for solving this point will be described below.

Signal pulse 1.2.3. . . A set of pulse timing detection circuits is used to detect fluctuations in the M pulse timing. That is, one of the optical pulse signals branched by the optical demultiplexer 84 is used as a reference pulse train. The pulse train of the system to be compared is selected by the optical channel selection circuit (many-to-one optical switch) 87, combined with the reference pulse train of one system by the optical coupler 89, and reaches the optical receiver 90. As a result, the fluctuation of the timing of the system selected by the electric selection circuit 92 is observed, and the fluctuation of the timing is corrected by the linked delay control circuit 94.

3.4. Sequentially through the optical channel selection circuit 87 and the electric selection circuit 92 . . The delay fluctuation of the M system is detected, and the correction in the delay control circuit 94 is repeated. Since the expansion and contraction of the optical path and the fluctuation of the timing due to the temperature change and the like are generally sufficiently slower than the operation speed of the optical channel selection circuit 87, the fluctuation of the delay fluctuation of the non-selected system is controlled by the sequentially turning control. It can be corrected sufficiently quickly. Thereby, the configuration and the number of components of the optical time division multiplexing device can be simplified.

As the optical channel selection circuit 87,
A switch array (not shown) using a semiconductor optical amplifier array as a gate is used.
(Planar Lightwave Circuit
t: Quartz-based lightwave circuit) Various devices such as a selective optical switch based on a change in the coupling ratio of a waveguide circuit can be considered. In this configuration, not only the optical pulse coupler (optical coupling unit) 89 but also a single set of peripheral devices such as the gain stabilizing optical amplifier 80 is sufficient. In this respect, the effect of power saving is great. When a semiconductor optical amplifier array is used as the optical channel selection circuit 87, it can also serve as a function of a fixed output optical amplifier 88 described below.

The detection of the time position of the signal is performed by the signal comparing section 91 and the appropriate threshold value set in the middle of the correlation output having the peak and the sum frequency optical receiver 9 as shown in FIG.
The comparison is made with the output of 0. Since the change in the relationship with the set threshold value is fed back to the adjustment amount of the optical path length, a change in the average light intensity due to switching of the input system or the like may be recognized as a change in the delay and cause a malfunction. In particular, in the present configuration in which the second-order optical nonlinear phenomenon is used in the sum frequency optical receiver 90, twice the fluctuation of the input light intensity becomes four times the fluctuation of the received signal. Similar effects are unavoidable even with variations in repetition (bit rate). Therefore, in the present embodiment, a gain-adjusting optical amplifier (AGC-
By arranging Amp) as the fixed output optical amplifier 88, a mechanism is provided for suppressing the fluctuation of the incident light intensity due to the change of the operating condition of the signal pulse train and the fluctuation of the reception intensity.

(Fifth Embodiment) FIG. 10 shows the configuration of a fifth embodiment of the present invention corresponding to the second embodiment of the present invention shown in FIG. Here, an example will be described in which the first problem of the two pulse time position fluctuation problems described above is addressed. The first problem is a large fluctuation of the pulse timing before the signal modulation unit due to the insertion of the pulse width and intensity adjustment unit due to the long fiber component of about 100 m.

Here, a semiconductor mode-locked laser which is electrically synchronized with the signal light is introduced into the reference light pulse train generating circuit 23, and the optical coupling section 34 which introduces the signal light pulse train and the reference pulse train into the receiving portion is a polarized light. A holding type fiber coupler (coupler) is used, a beam splitter (PBS) is used as a polarization separation unit 35 between pulses, and a polarization separation unit (PBS) is used as an optical pulse cross-correlation receiving unit 32 for observing an overlapping component. 35, a pair of sum frequency light receivers 36 and 37 each composed of a dielectric thin film avalanche photodiode (APD) and a dielectric KTP or LiNbO 3 crystal which is a sum frequency light generating optical crystal. The dielectric crystals 36-1 and 37-1 and the semiconductor receivers 36-3 and 3-3
An extremely compact pulse overlap component receiver is realized by sandwiching the condenser lenses 36-2 and 37-2 between 7-3 and attaching the two components.

An increase / decrease in the signals received by the sum frequency optical receivers 36 and 37 is detected by a signal comparing section 38, and a pulse motor type optical delay control circuit 31 which uses the detected signal as a signal for delay control is provided. ing. In FIG. 10, another light source electrically synchronized is used as the reference light pulse train. However, it is of course possible to use the pulse branched near the emission port of the signal light source (signal short pulse train generation circuit) 21 as it is as the reference pulse train. It is.

Optical pulse cross-correlation receiver 32 according to the present invention
Most of the pulse signals incident on the sum frequency optical receivers 36 and 3
7 is converted into a sum frequency optical signal in the dielectric crystals 36-1 and 37-1, but the portion that could not be converted here is the light receiving surface of the semiconductor receivers 36-3 and 37-3 transparent to the fundamental wavelength. Optical system (condensing lens) 36-
2, 37-2 are set.

In the semiconductor optical receivers 36-3 and 37-3 such as the APD and the pin receiver, the sensitivity is reduced for an optical signal having a wavelength lower than the wavelength originally defined by the band gap energy. do not have. However, when the signal peak intensity is extremely high, photoelectromotive force is generated near the photocathode of the receiver due to a two-photon absorption phenomenon. This current also increases by the square of the signal strength similarly to the reception of the sum frequency optical signal, and thus contributes to selective signal reception sensitivity. In the fifth embodiment, since the reference pulse train and the signal pulse train are multiplexed in the same polarization direction, a dielectric material such as a dielectric KTP or LiNbO3 crystal which is a sum frequency light generating optical crystal is used. By re-focusing on the upper side, the signal sensitivity can be obtained by adding the two-photon absorption power and the sum frequency signal without distinction.

The change in the outputs of the receivers 36 and 37 when the time position of the signal pulse train moves forward and backward with reference to the center time position of the reference light pulse train separated into two by the optical pulse separation unit 33 is shown. As shown in FIG. When the reference point of the comparison signal is set to the zero point (Just) in FIG. 11, when the delay of the signal pulse increases +, the output of the sum frequency light of the receiver 36 (A) shown in FIG. Receiver 3 at the same time
7 (B) is reduced. If the signal comparing section 38 compares the difference between the outputs of the receivers A and B, it outputs a positive output (control signal). On the other hand, when the delay of the signal pulse decreases, the output of the sum frequency light of the receiver A decreases, and at the same time, the sum frequency high output of the receiver B increases, and the signal comparison unit 38 outputs a negative output (control signal). .

When the comparison output of the signal comparing section 38 is negative,
The optical delay control circuit 31 issues a command to the pulse motor in the direction of increasing the overlapping component, and when the comparison output of the signal comparing section 38 is positive, the optical delay control circuit 31 decreases the overlapping component.
Issues a command to the pulse motor. Accordingly, the optical delay control circuit 31 maintains a stable delay position only by determining whether the output signal of the signal comparison unit 38 is positive or negative.

The pulse time resolution of the measuring apparatus using the sum frequency is generally about half a half-width of several hundred femtoseconds (however, slightly different depending on the thickness of the crystal used).
However, if the signal comparing unit 38 evaluates the variation, not the absolute value, of the variation of the overlap between the pulses, that is, the difference, etc., it is sufficient to evaluate the variation, so that a high resolution of 100 femtoseconds or less is possible. Since each of the reference pulse train and the signal light pulse train has a high peak value, even when there is no overlap between pulses, a sum frequency light signal depending on each pulse itself is generated to some extent, and the signal-to-noise ratio is slightly reduced.

If the time position between the pulses is far away or the overlap position is reversed due to the position shift exceeding the detection range (high-precision detection range) shown in FIG. 5, the instantaneous detection of the correct position is performed. Becomes difficult. However, in the present invention,
Originally, assuming a slow drift from the set position, the purpose is to make corrections within a range that can maintain signal quality stability before deterioration becomes noticeable, so if the accuracy is high, the detection range is A time region of about several times the pulse width is sufficient.

By the way, even if the light intensity of the reference pulse train or the signal pulse train slightly changes, the sum frequency light output which is proportional to the square of the signal peak intensity greatly changes. A change in the average light intensity due to switching of the input system or the like is recognized as a change in the delay, and may cause a malfunction. In particular, in the configuration of the present invention using the second-order optical nonlinear phenomenon, the input light intensity of 2
The double variation is a four-fold variation of the received signal. Similar effects are unavoidable even with variations in repetition (bit rate). However, in this embodiment, the two sum frequency optical receivers 3
Since the variation of the two received signals is monitored by 6, 37, the delay control signal output from the signal comparing section 38 does not malfunction even in such a case.

In this case, the optical delay control circuit 31
As can be seen, a suitable comparison formula (eg, a difference) of two cross-correlation signals characteristically received by time position
Can be operated with a simple instruction system in which the operation direction of the variable delay circuit is reversed in the direction above and below. Therefore, a computer is not required, and the discrimination I for identifying the signal level for each is possible.
It is sufficient to provide C, and the control system can be configured very simply and inexpensively.

The receivers 36 and 37 for comparing the overlapped components between the reference pulse and the signal pulse are provided with the K signal for generating the sum frequency light.
To obtain a high photon density on both the TP crystals 36-1 and 37-1 and the photocathode of the APD receivers 36-3 and 37-3 for amplifying the two-photon absorption power, the lenses 36-2 and 37-2 are used. It is focused twice and achieves high receiving sensitivity at low cost.

The relative positional relationship between the reference pulse and the signal pulse after passing through the optical coupling unit 34 is uniquely determined.
Thereafter, even if the arrival timing drifts via a redundant optical circuit, the relative positional relationship does not change. Therefore, an EDFA (optical amplifier) is used as a means for further increasing the receiving sensitivity.
Can be inserted between the optical coupling unit 34 and the sum frequency optical receivers 36 and 37.

(Sixth Embodiment) The method of detecting and stabilizing the optical pulse timing according to the present invention can also solve the second problem of the time position fluctuation described above. A specific countermeasure in this case will be described below as a sixth embodiment of the present invention. The second problem is that in the conventional optical pulse multiplexing control of FIG. 16, an optical pulse output from the signal short pulse train generating circuit 1 is branched into a plurality of parts by a branching unit 4, and each of the branched optical pulses is Modulation is performed in the modulator 5 by the transmission information, and a plurality of systems of optical pulses b. c. d. . . It is said. Here, the optical path length, arrangement, temperature environment, etc. of several meters of the components of each system are different, and the
Therefore, there is a problem that the time positional relationship of the multiplexed pulse signals which should be arranged at equal intervals fluctuates with each other.

FIG. 12 shows a configuration of an optical pulse timing detection circuit and an optical time division multiplexing apparatus according to a sixth embodiment of the present invention, and an example of output signals. 12, the same components as those in FIG. 7 are denoted by the same reference numerals. Here, 100 is an optical pulse cross-correlation receiver for the number of branches (M), 101 is a signal comparison unit for the number of branches (M), and 102 is a pair of sum frequency optical receivers.

In the apparatus of this embodiment, a plurality of systems of optical pulses are respectively branched, and a branching unit 71 for detecting the optical pulse timing is provided at each of them and arranged at equal intervals when the plurality of systems of optical pulses are time-multiplexed. And a semi-fixed delay unit 70 capable of making the delay time inserted in each optical pulse path variable, and one optical pulse train among a plurality of systems guided by the branch unit 71 as a reference pulse train, and the remaining M −
An optical pulse train split from a plurality of systems is used as a signal pulse train, and a polarization beam splitter (PB) for the number of branches (M) in which the reference pulse train and the signal pulse train overlap each other.
S) and a sum frequency light receiver for the number of branches (M) including a sum frequency light generating crystal, a lens, and an optical receiver for inputting the output light of each optical multiplexer 73. Pulse cross-correlation receiver 100 and optical pulse cross-correlation receiver 100
Signal comparing sections (optical pulse timing detecting circuits) 101 for the number of branches (M) for detecting the pulse timing based on the cross-correlation signal output from the CPU, and variable delay for performing feedback control based on the position detection signals output from the signal comparing section 101 With the provision of the circuit 76, a plurality of optical pulse trains are time-multiplexed at equal intervals.

In this embodiment, since the reference pulse train and the signal pulse train are multiplexed by the respective optical multiplexers 73 composed of polarization beam splitters in the polarization directions orthogonal to each other, the main axis is used as a means for generating the second harmonic. Uses a type 2 type crystal (for example, AANP) that generates a second harmonic at a sum frequency of different lights as the sum frequency light generation crystal. In that case, the output of the SHG (second harmonic) signal when there is no overlapping component, which is a problem in the fifth embodiment, can be set to 0 in principle. Furthermore, when the signal of the basic pulse train is cut off by the filter in the optical multiplexer 73, only the cross-correlation component between the orthogonal pulses is received, so that a monitor signal with a high signal-to-noise ratio can be obtained.

FIG. 13 shows the relationship between the overlapping time position between the reference and signal pulse trains and the output intensity of the two receivers A and B in this configuration. In this configuration, when the reference pulse train and the signal pulse train are combined, different polarizations orthogonal to each other are used. Therefore, when combining using a polarization beam splitter, the coupling loss can be reduced to zero in principle.

In this case, a higher signal
A monitor signal having a noise ratio can be obtained. This is because in the fifth embodiment, the type I phase matching condition (the sum frequency light or the second harmonic light (SHG) is generated efficiently when the polarization of the two incident basic light pulses is parallel to each other) Is used, the second harmonic light of the reference light pulse (or the signal light pulse) is generated even if there is no overlap between the reference light pulse and the signal light pulse. The second harmonic light becomes background noise light, and deteriorates the detection accuracy of the comparison signal obtained from the signal comparison unit 101 and the accuracy of the timing control. However, in the configuration of the sixth embodiment, the type II phase matching condition
A nonlinear optical material of efficiently generating sum frequency light or second harmonic light (SHG) when the polarizations of two basic light pulses are orthogonal to each other is used for the light pulse cross-correlation unit 100. Therefore, when both the reference light pulse and the signal light pulse whose polarization directions are orthogonal to the light pulse cross-correlation unit 100 are incident, the sum frequency light is generated. However, when only one of the reference light pulse or the signal light pulse is incident. Generates neither sum frequency light nor SHG. Therefore, in the configuration of the sixth embodiment, when there is no inter-pulse overlap component, noise due to the SHG output signal generated for each individual pulse which cannot be ignored in type 1 is remarkably suppressed, and a high S / N ratio is obtained. Detection becomes possible.

Since the wavelength filter cuts off the reference light pulse train and the signal light pulse train, it is not necessary to consider the effect of two-photon absorption. Therefore, the photomultiplier tube has a smaller quantum efficiency than the APD but has a very large electron multiplication factor. A high-sensitivity light receiving element such as a photomultiplier (PM) can be used for the optical pulse cross-correlation receiving unit 100. Also, in this configuration, when combining the reference pulse train and the signal pulse train, different orthogonal polarizations are used, so that the coupling loss can be reduced to zero in principle. Has low optical power loss and high S /
An N ratio is obtained. The receiver having the high signal-to-noise ratio as shown in the present embodiment is realized for the first time by such a clever configuration of the parts.

In any of the fifth and sixth embodiments,
The signal indicating the time delay output from the optical pulse cross-correlation receiving unit 100 is an ultra-high-speed photovoltaic power, but the positional relationship between repeated pulse trains does not change for each bit, so that the bandwidth of the receiving unit 100 is equal to the length of the signal. The time average value is observed and the signal is output to the signal comparing unit 101 of the optical path length control circuit.
An output of 00 Hz or less is sufficient. Therefore, the circuit and the device of the present embodiment can be easily constituted by inexpensive components such as a photomultiplier tube and an APD receiver having a low speed but high receiving sensitivity.

The optical nonlinear material used for sum frequency generation needs to propagate long in the crystal in order to obtain high conversion efficiency. Generally, the most efficient plane orientation is D33.
In the direction, the propagation constant mismatch causes a rapid phase mismatch and is difficult to interact long. Therefore, a long interaction length can be maintained while maintaining high efficiency by periodically correcting the group velocity dispersion by a process such as ion implantation. Such quasi-phase matching (QP
M) If an optical waveguide with good confinement to incident light is formed by using the treated crystal, higher interaction efficiency and light density can be obtained than in the case of using the flat plate described in the fifth embodiment and the like. .

When the QPM crystal is used in an autocorrelator or the like for performing time-resolved measurement, pulses are crossed from spatially different directions, and the output of the overlapping portion is detected. In this case, since the interaction time is long in the QPM crystal, the waveform is deteriorated due to the ambiguity of the overlapping position timing of the measurement pulse, and the time resolution is significantly reduced. However, as in this embodiment, the time position between the pulses is determined by the optical coupler 7.
After fixing at 3, if a pulse is injected on the same optical axis to compare the increase / decrease of the sum frequency output intensity depending on the relative positional relationship, the time resolution is not significantly reduced. On the other hand, the photosensitivity is significantly enhanced as the crystal lengthens. By using such a nonlinear element, it becomes possible to realize a selective optical receiver having good mode coupling efficiency and good receiving sensitivity even for an optical fiber used in a network.

(Seventh Embodiment) Further, when an individual pulse timing detection circuit is provided for each of a plurality of systems,
The redundancy when adding a similar system is excellent. But,
When the number of branches and the configuration scale increase, individual differences such as optical sensitivity between the detection devices, fiber length errors, and variations in the electrical threshold value, and an increase in the number of components become problems. A seventh embodiment of the present invention as a specific example for solving this point will be described with reference to FIG.

In FIG. 14, the same components as those in FIG. 9 are denoted by the same reference numerals. Here, reference numeral 104 denotes an optical pulse separation unit, and reference numeral 105 denotes an optical pulse cross-correlation receiving unit.

In this embodiment, the signal pulses 1.2.
3. . . In order to detect each of the M pulse timing fluctuations, a set of optical pulse cross-correlation receiving units 105 is used as optical pulse timing detection means. One of the optical pulse signals split by the optical splitter 84 is used as a reference pulse train. Two pulse trains to be compared are selected by an optical channel selection circuit (many-to-one optical switch) 87, the reference pulse train and the selected one pulse are combined by an optical coupling unit 87, and the combined pulse light is output. Arrives at the optical pulse cross-correlation receiving unit 105 having a plurality of sum frequency optical receivers.

The signal comparing section 91 uses the output signal from the optical pulse cross-correlation receiving section 105 to output the signal to the optical channel selecting circuit 8.
The fluctuation of the pulse timing of the system selected in 7 is observed. The detection result of the signal comparing section 91 is output to the electric selection circuit 92 by the timing control of the selection timing controller 95.
Is fed back to the variable delay circuit 93 as a delay control signal for the channel. In accordance with the timing control of the selection timing controller 95, the order of 3.
4. . . The detection and correction of the delay fluctuation of the M system are repeated.

The expansion and contraction of the optical path and the fluctuation of the timing due to the temperature change and the like are generally sufficiently slower than the operation speed of the optical channel selection circuit 87. Therefore, the fluctuation of the delay fluctuation of the non-selected system goes around sequentially. It can be corrected sufficiently quickly by the coming control. Thereby, the configuration and the number of components of the optical time division multiplexing device can be simplified. As the optical channel selection circuit 87, a switch array using a semiconductor optical amplifier array as a gate is used as an example, but various other devices such as a selection optical switch based on a change in the coupling ratio of a PLC waveguide circuit are also conceivable. In this configuration, not only the optical coupling unit 89 but also a single set of peripheral devices such as a signal comparing unit 91 and an optical amplifier (not shown) is sufficient. In this respect, the effect of labor saving is great. When a semiconductor optical amplifier array is used as the optical channel selection circuit 87, it can also serve as an optical amplifier.

(Other Embodiments) The principle of using the cross-correlation component of an optical pulse for time position measurement is the same,
The means for receiving the increase / decrease of the overlap component between the pulses is not limited to the generation of the sum frequency light or the generation of the two-photon absorption power as exemplified in the above-described embodiments of the present invention. For example, in the case where a reference pulse train is provided separately from a signal pulse train, if the wavelength of the reference pulse train is different from the wavelength of the signal pulse train, after performing pulse-to-pulse synthesis in an optical coupling unit, an optical non-linear medium such as a semiconductor optical amplifier ( By injecting both a signal pulse train and a reference pulse train into a not-shown pulse train, a fixed wavelength (for example, a reference pulse wavelength l (w
1) In the case of the signal pulse wavelength l (w2), the four-wave mixing output wavelength l (2w1-w2) or l (2w2-w)
In 1)), a correlation signal is obtained. If only a four-wave mixing output is received by a wavelength selection filter (not shown), it has the same signal increase / decrease characteristics as the sum frequency light.
The same effect as in the case of generating photon absorption power can be expected.

Finally, as a means for giving a gain to increase the sensitivity of the receiving circuit, an optical amplifier may be incorporated at a stage before reception, and the invention is not limited to amplification at an electric signal stage. Here, the incorporation of a fiber-type optical amplifier or a long electrical wiring or the like generally causes delay fluctuation itself in normal optical pulse reception and causes an error in timing detection, but in the present invention, The time timing between a plurality of pulses is fixed when passing through an optical coupling unit (optical coupler or polarization beam splitter), and thereafter, even if the time is guided to a receiver via an amplifier having an optical path length expansion / contraction, the time position of the two is determined. The relative relationship does not change. Therefore, according to the present invention, the incorporation of a fiber type optical amplifier or the like does not cause an error factor in the timing detection accuracy.

[0121]

As described above, according to the present invention, the amount of displacement from the reference position and its direction are specified by increasing or decreasing the mutual overlapping components of the reference light pulse train and the signal light pulse train separated into two at the measurement position. Since it is possible to evaluate only one signal value caused by the second-order optical nonlinear effect or to compare two cross-correlation signal values, the delay amount to be corrected and the correction direction (whether to increase the delay or not) Can be specified with high accuracy.

According to the first embodiment of the present invention, for a high-speed repetitive pulse, the time position between pulse trains can be detected at 1/10 or less of the pulse width, but the average of nonlinear signals obtained from each repetitive signal can be detected. Since it suffices to read the power, there is no need to configure a high-speed optical receiver or electric circuit. Therefore,
The timing detection accuracy does not depend on the response speed of the optical receiver and the electric circuit to be configured. Further, according to the second embodiment of the present invention, even when the light intensity of the pulse train to be evaluated fluctuates due to a change in the operation mode and each received photovoltaic power fluctuates, a relative comparison between the two receiver powers is performed. Therefore, no error in timing detection occurs. In addition, for a high-speed repetitive pulse, it has a high time resolution that can separate the time position between pulse trains by 1/10 or less of the optical pulse width, but it is sufficient to read the average power of the nonlinear signal obtained from each repetitive signal. It does not need to be constituted by a high-speed optical receiver or an electric circuit. Therefore, the timing detection accuracy does not depend on the response speed of the optical receiver and the electric circuit to be configured.

Further, the present invention is extremely accurate because it depends on the optical response of a light-receiving element that generates a two-photon absorption power and a nonlinear material that generates a sum frequency. An optical pulse train used for wider band optical transmission has a shorter pulse width, and has expandability such that nonlinear sensitivity and time resolution are further improved.

As described above, according to the present invention, a simple, low-speed and inexpensive optical / electrical circuit and a small number of passive optical components are used, and a highly accurate time position measurement for a highly repetitive optical pulse can be performed. Fluctuation can be compensated. Further, according to the present invention, different bit rates can be easily coped with the same configuration, so that feedback control is performed so that the relative time position of the optical pulse does not fluctuate in the optical time division multiplexing device which is sequentially expanded. Therefore, it is possible to easily maintain the equidistant multiplex arrangement of the optical pulse trains of a plurality of systems.

Therefore, according to the present invention, a good optical communication system having a low error rate can be stably operated for a long period of time, and a change in bit rate, multiplicity and optical power can be handled. Even when expansion such as an increase in the number of systems in the apparatus and an increase in transmission rate in the future, effects such as stable and easy operation continuation can be obtained.

The concept of the present invention becomes more effective as the pulse becomes shorter, and can be applied regardless of the level of repetition. Therefore, it can be easily applied not only to optical communication but also to generation and control of a short pulse for measurement.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a first embodiment of a principle configuration of an optical pulse timing detection circuit of the present invention.

FIG. 2 is a block diagram showing a second embodiment of the principle configuration of the optical pulse timing detection circuit of the present invention.

FIG. 3 is a conceptual diagram illustrating a response to signal fluctuation of the correlation between a reference optical pulse train and a signal pulse train separated into two temporally.

FIG. 4 is a block diagram illustrating a configuration of an optical pulse timing detection circuit according to a first example of the present invention.

FIG. 5 is a timing chart showing a relationship between an overlapping time position relationship between a reference pulse train and a signal pulse train and a receiver output intensity in the first embodiment of the present invention.

FIG. 6 is a block diagram illustrating a configuration of an optical time division multiplexing device according to a second embodiment of the present invention.

FIG. 7 is a block diagram showing a configuration of an optical time division multiplexing device according to a third embodiment of the present invention using a type 2 sum frequency crystal.

FIG. 8 is a timing chart showing a relationship between an overlapping time position relationship between a reference pulse train and a signal pulse train and a receiver output intensity in the third embodiment of the present invention.

FIG. 9 is a block diagram showing a configuration of an optical time division multiplexing device according to a fourth embodiment of the present invention using a QPM waveguide, a two-photon absorption and fixed output optical amplifier (optical power compensation amplifier).

FIG. 10 is a block diagram illustrating a configuration of an optical pulse timing detection circuit and a control circuit according to a fifth embodiment of the present invention.

FIG. 11 is a timing chart showing a relationship between an overlapping time position relationship between a reference pulse train and a signal pulse train and a receiver output intensity in the fifth embodiment of the present invention.

FIG. 12 is a block diagram illustrating a configuration of an optical time division multiplexing device according to a sixth embodiment of the present invention.

FIG. 13 is a timing chart showing a relationship between an overlapping time position relationship between a reference pulse train and a signal pulse train and a receiver output intensity in the sixth embodiment of the present invention.

FIG. 14 is a block diagram showing a configuration of an optical time division multiplexing device according to a seventh embodiment of the present invention using a QPM waveguide, two-photon absorption, and a selector.

FIG. 15 is a block diagram showing a configuration example of a conventional optical time division multiplexing apparatus and a factor of timing fluctuation thereof.

FIG. 16 is a block diagram showing a configuration of a conventional optical time-division multiplexing device incorporating a number of electrical phase detection circuits (PLLs and the like).

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Signal short pulse train generation circuit 2 Pulse width compression circuit 3 Optical amplification circuit (EDFA) 4 Branch unit 5 Modulation unit 6 Delay adjuster 7 1 system 8 M system 9 Optical multiplexing unit 10 Pigtail 11 Optical branching unit 12 Variable delay circuit 13 Phase Control circuit 15 Phase comparison control circuit 16 Phase difference comparator 17 Phase detector 18 Frequency filter 19 Electric amplifier 20 Optical receiver 21 Signal short pulse train generation circuit 22 Long optical path & parts (temperature fluctuation part) 22-1 Pulse width compression Circuit 22-2 Optical amplification circuit (EDFA) 23 Reference optical pulse train generation circuit 24 Optical branching unit 25 Semi-fixed delay unit 26 Optical coupling unit 27 Sum frequency optical receiver 27-1 Optical crystal 27-2 Condensing lens 27-3 Semiconductor Receiver 28 Signal identification circuit 29 Optical pulse cross-correlation receiving unit 31 Optical delay control circuit 32 Optical pulse cross-correlation receiving unit 33 Demultiplexing unit 34 Optical coupling unit 35 Demultiplexing unit 36, 37 Sum frequency optical receiver 38 Signal comparing unit 40 Channel branching / modulation / mutual delay control / multiplexing unit 41 Timing drift 42 Signal light pulse train 43 Reference light pulse train 45 Light pulse timing detection Circuit 48 Optical branching unit 49 Modulator 50 Delay adjuster 51 Optical branching unit 52 Pigtail 53 Optical coupler 54 Sum frequency optical receiver 54-1 Dielectric crystal 54-2 Condensing lens 54-3 Semiconductor receiver 55 Semi-fixed delay Unit 56 variable delay circuit 57 optical multiplexing unit 58 optical pulse timing detection circuit 60 optical branch unit 61 long component 62 variable delay circuit 63 optical branch unit 64 semi-fixed delay unit 65 delay control circuit 66 sum frequency optical receiver 66-1 dielectric Body crystal 66-2 Optical filter 66-3 Semiconductor receiver 67 Optical multiplexer (PBS) 68 Optical branching unit 6 Modulator 70 Delay adjuster (semi-fixed delay unit) 71 Optical branch unit 72 Pigtail 73 Optical multiplexer (PBS) 74 Sum frequency optical receiver 74-1 Dielectric crystal 74-2 Optical filter 74-3 Semiconductor receiver 75 Half Fixed delay section 76 Variable delay circuit 77 Optical multiplexing section 78 Optical pulse timing detection circuit 80 Optical amplifier circuit (EDFA) 81 Branch section 82 Modulator 83 Delay adjuster (semi-fixed delay section) 84 Optical branch section 85 Pigtail 86 Semi-fixed delay Unit 87 optical channel selection circuit 88 fixed output optical amplifier 89 optical coupling unit 90 sum frequency optical receiver 90-1 dielectric crystal 90-2 condenser lens 90-3 semiconductor receiver 91 signal comparison unit 92 electric selection circuit 93 variable delay Circuit 94 Delay control circuit 95 Selection timing controller 96 Optical multiplexing unit 97 Optical pulse cross-correlation receiving unit 98 Optical pulse Timing detection circuit 100 optical pulse correlation receiver 101 signal comparator 102 sum frequency light receiving unit 104 the light pulse separator 105 optical pulse correlation receiver

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) // G01B 11/16 (72) Inventor Hidehiko Takara 2-3-1 Otemachi, Chiyoda-ku, Tokyo Japan F term (reference) in Telegraph and Telephone Corporation 2F065 DD04 EE00 FF32 FF49 GG08 HH09 JJ01 JJ18 LL00 LL02 LL26 LL37 QQ00 QQ11 QQ25 QQ27 5K002 BA02 CA11 CA12 CA14 DA03 FA01

Claims (23)

[Claims] 1. A repetition frequency f ₀ / N serving as a time reference
A reference light pulse train, and an optical pulse timing detection circuit for inputting a signal light pulse train having the same repetition frequency or N times the repetition frequency f ₀ as the reference light pulse train, a branching means for branching the signal light pulse train, And a semi-fixed delay unit that sets the reference light pulse train and the signal light pulse train to partially overlap in time position. The output of the semi-fixed delay means and the reference light pulse train An optical receiver for receiving the output of the optical coupler and observing an overlapping component between the optical pulse trains, and an identification circuit for detecting an increase or decrease in the output of the optical receiver. An optical pulse timing detection circuit comprising: an optical pulse cross-correlation receiving means. 2. The optical pulse timing detection circuit according to claim 1, wherein the optical receiver is a generating section of a sum frequency light generated by overlapping a plurality of pulses, and a receiver having high sensitivity to the wavelength of the sum frequency light. And a discrimination circuit that measures a change in the time positional relationship of the signal light pulse train using an increase or decrease in the signal output of the sum frequency light receiver. Detection circuit. 3. The optical pulse timing detection circuit according to claim 1, wherein the optical coupling means couples both the pulse trains of the reference optical pulse train and the signal light pulse train in the same linear polarization direction. A type 1 sum frequency light generating material that generates an optical signal by interaction between pulse trains in a linear polarization direction, and a receiver that is sensitive to a wavelength shorter than the input light wavelength; An optical pulse timing detection circuit for measuring a change in a time position relationship of the signal light pulse train by using an increase or decrease in a generation output of a sum frequency light due to an overlap component between two light pulse trains. 4. The optical pulse timing detection circuit according to claim 3, wherein the optical receiver uses a receiver having low sensitivity to a fundamental light wavelength and having high sensitivity to a shorter wavelength than an input light wavelength; An optical pulse timing detection circuit, wherein the discrimination circuit measures a change in a time positional relationship of the signal light pulse train by using an increase / decrease of two-photon absorption power due to an overlap component between two light pulse trains to be compared. 5. The pulse timing detection circuit according to claim 4, wherein the optical receiver is a type 1 sum frequency generating material that generates an optical signal by an interaction between pulse trains having the same linear polarization direction, and a basic light. An avalanche photodiode receiver having a low sensitivity to wavelengths and a high sensitivity to wavelengths shorter than the input light wavelength, wherein the discriminating circuit increases and decreases the sum frequency light output by an overlap component between two light pulse trains to be compared. An optical pulse timing detection circuit for measuring a change in a time positional relationship of the signal light pulse train using a sum of both increases and decreases in photon absorption power. 6. The optical pulse timing detection circuit according to claim 1, wherein the optical coupling means uses a polarization beam splitter having a plurality of input / output ports, and the reference optical pulse train and the reference light pulse are output from different planes of the polarization beam splitter. The signal light pulse train is incident and the two pulse trains are multiplexed with linearly polarized light orthogonal to each other, and the optical receiver performs phase matching when the polarization directions of the signal light pulse train and the reference light pulse train are orthogonal to each other. A type 2 sum frequency light generating material that generates sum frequency light, a filter that transmits the sum frequency wavelength and blocks the fundamental wavelength, and has a low sensitivity to the fundamental light wavelength and a wavelength shorter than the input light wavelength. A receiver having high sensitivity, wherein the identification circuit detects an increase or decrease in the generation output of the sum frequency light due to an overlap component between the two optical pulse trains to be compared. Timing detection circuit. 7. The optical pulse timing detection circuit according to claim 6, wherein the branching unit branches a part of the reference light pulse train and the signal light pulse train for monitoring, and monitors a variation in the intensity of the signal light pulse train. An optical pulse timing detection circuit comprising: a fixed output intensity optical amplifier for compensating; and delay control means for outputting a delay control signal corresponding to an output signal value of the identification circuit. 8. The optical pulse timing detection circuit according to claim 1, wherein an optical frequency (w2) different from the signal light pulse train (optical frequency w1) and a repetition frequency f ₀ / temporally synchronized.
N light pulse trains are used as the reference light pulse trains, and the optical coupling means combines the two light pulse trains by combining the reference light pulse trains and the signal light pulse trains in the same linear polarization direction. A third-order nonlinear optical material that generates four-wave mixing signal light by the interaction between the two optical pulse trains, and a phase conjugate optical frequency ((2) of the two without transmitting the reference optical frequency w1 and the signal optical frequency w2. × w1-w2) or (2 × w2-w1));
A receiver having high sensitivity in the vicinity of a fundamental wavelength, wherein the discriminating circuit uses the increase or decrease of the output of the four-wave mixing signal light due to the overlapping component of the two light pulse trains to be compared, and the time position relationship of the signal light pulse train. An optical pulse timing detection circuit for measuring fluctuations in the light pulse timing. 9. A branching means for branching a plurality of signal light pulse trains, and one of the repeated light pulse signals or a pulse train separately prepared by the branching means is used as a reference light pulse train. The plurality of optical pulse timing detection circuits according to any one of claims 1 to 8, wherein each of the plurality of optical pulse timing detection circuits detects the timing of the signal light pulse train as the signal light pulse train. Delay control means for controlling the pulse timings of the remaining plurality of optical pulse trains, respectively, such that the optical pulse trains of the plurality of systems are arranged at equal intervals when multiplexed, and the timing is controlled by the delay control means. Optical multiplexing means for temporally combining the signal light pulse trains of the plurality of systems, the optical pulse timing detection circuit A delay time to be given to each of the delay control means according to a timing detection result of the remaining plurality of optical pulse signals detected by comparing the overlap time positions between the reference optical pulse train and the signal optical pulse train. An optical time division multiplexing device for outputting a delay control signal for feedback control. 10. The optical time-division multiplexing apparatus according to claim 9, wherein the optical pulse timing detection circuit is provided only in one system, and the optical pulse trains of the plurality of systems branched by the branching unit are sequentially arranged. A selection timing controller for generating a timing signal for selection; an optical channel for accommodating the branches of the plurality of optical pulse trains and for selecting one input temporally using the timing signal of the selection timing controller A selection switch, wherein the plurality of optical pulse trains branched by the branching means are sequentially selected in a time division manner through the optical channel selection switch and supplied to the optical pulse timing detection circuit, The output of the detection circuit is used as a delay control signal by an electric signal selection switch linked to the optical channel selection switch. It is distributed to the delay control unit elephant series, when the light characterized in that it is fed back sequentially said delay control means division multiplexing apparatus. 11. Repetition frequency f ₀ / time reference
An optical pulse timing detection circuit for inputting N reference light pulse trains and a signal light pulse train having the same repetition frequency as the reference light pulse train or a repetition frequency f ₀ which is N times as large as the reference light pulse train; A light separating means for separating the reference light pulse train into two reference light pulse trains having orthogonal polarization directions and having a certain delay time difference; and inputting one output of the branching means, and the time position of the signal light pulse train to be synchronized is A semi-fixed delay unit that is set to be in the middle of the time position of the two reference light pulse trains separated by the light separation unit, and one of the two reference light pulse trains separated by the light separation unit and the The two reference light pulses which are combined with a part of the optical power of the signal light pulse train delayed by the semi-fixed delay means and separated by the light separating means. An optical synthesizing unit that multiplexes the other of the optical pulse trains with the rest of the optical power of the signal light pulse train delayed by the semi-fixed delay unit. Two optical pulse cross-correlation receiving means for observing the signal light; and a signal comparing means for measuring a change in the time position of the signal light pulse train before and after by detecting an increase or decrease in the output of the two light pulse cross-correlation receiving means. 1. An optical pulse timing detection circuit comprising: a timing identification unit configured by: 12. The optical pulse timing detection circuit according to claim 11, wherein the light separating means converts the light pulse into a polarization direction of 45 degrees with respect to a main axis of the birefringent optical material or a circular polarization equivalent thereto. An optical pulse timing detection circuit, which is incident. 13. The optical pulse timing detection circuit according to claim 11, wherein the two optical pulse cross-correlation receiving units each generate a sum frequency light generated by overlapping a plurality of pulses, and a wavelength of the sum frequency light. And a high-sensitivity optical receiver, wherein the signal comparing means measures a change in the time positional relationship of the signal light pulse train by increasing or decreasing the signal outputs of the two optical receivers. Characteristic light pulse timing detection circuit. 14. The optical pulse timing detection circuit according to claim 13, wherein the light synthesizing unit is a polarization maintaining type optical coupling unit having two input ports and one or two output ports, and one or two or more polarization maintaining type optical coupling units. A polarization separation unit having an input port and two output ports, wherein the two reference light pulse trains whose polarization directions are orthogonal to each other and have a fixed delay time difference are converted into a polarization main axis direction (X axis Direction and the Y-axis direction) from one of the input ports so that the signal light pulse train is
The polarization direction is incident from the other input port such that the polarization direction becomes linearly polarized light of about 45 degrees with respect to the principal axis direction of polarization of the polarization maintaining type light combining means or circularly polarized light equivalent thereto, and the reference light pulse train and the signal light pulse train are input. Taking out the combined light from the output port, and entering the combined light from the input port such that the polarization direction of the reference light pulse train matches the main polarization direction (X′-axis direction and Y′-axis direction) of the polarization separation means; An optical pulse timing detection circuit, wherein an X'-axis direction component and a Y'-axis direction component of the combined light are separated and taken out from two different output ports. 15. The avalanche photodiode receiver according to claim 14, wherein said optical pulse cross-correlation receiving means has low sensitivity to a fundamental light wavelength and has high sensitivity to a shorter wavelength than an input light wavelength. An optical pulse timing detection circuit, wherein the signal comparison means measures a change in a time position relationship of the signal light pulse train by using an increase or a decrease in two-photon absorption power due to an overlapping component of two pulses to be compared. 16. The type 1 sum frequency light generating material according to claim 15, wherein said optical pulse cross-correlation receiving means generates an optical signal by an interaction between pulses having the same linear polarization direction. And a receiver having low sensitivity to the fundamental light wavelength and having high sensitivity to shorter wavelengths than the input light wavelength, wherein the signal comparing means increases or decreases the sum frequency light output due to the overlapping component of the two optical pulse trains to be compared. An optical pulse timing detection circuit for measuring a change in a time positional relationship of the signal light pulse train by using a sum of the two and the increase and decrease of the two-photon absorption power. 17. The optical pulse timing detection circuit according to claim 16, wherein said light combining means includes a polarization combining means having two input ports and two output ports, wherein said polarization directions are orthogonal and a fixed delay time difference is provided. The two reference light pulse trains are input from one of the input ports such that the orthogonal polarization directions coincide with the main polarization direction (X-axis direction and Y-axis direction) of the polarization combining means. The polarization direction is incident from the other input port such that the polarization direction becomes linear polarization of about 45 degrees with respect to the main polarization direction of the polarization combining means or circular polarization equivalent thereto, and the combined light of the reference light pulse train and the signal light pulse train is formed. By taking out from two output ports of the polarization combining means, the combination of the reference pulse train and the signal pulse train and the combination and separation between the light pulses of two different polarization directions can be performed. Light pulse timing detection circuit, which comprises carrying out at. 18. The optical pulse timing detection circuit according to claim 17, wherein the optical pulse cross-correlation receiving means performs phase matching when the polarization directions of the reference optical pulse train and the signal pulse train are orthogonal to each other. A type 2 sum frequency light generating material that generates sum frequency light, a wavelength filter that transmits the sum frequency light wavelength and blocks the basic light wavelength, and has low sensitivity to the basic light wavelength and is shorter than the input light wavelength. An optical pulse timing detection circuit comprising: an optical receiver having high sensitivity to a wavelength; wherein the signal comparison means detects an increase or decrease in sum frequency light due to an overlapping component of two optical pulse trains to be compared. 19. The optical pulse timing detection circuit according to claim 11, wherein an optical frequency (w2) different from the signal light pulse train (optical frequency w1) and a temporally synchronized repetition frequency f ₀ /
N light pulse light sources are used as the reference light pulse train. The light synthesizing means combines the reference pulse train and the signal pulse train in the same linear polarization direction, and the two light pulse cross-correlation receiving means A third-order nonlinear optical material that generates four-wave mixing signal light by the interaction between two incident light pulse trains, and a phase conjugate wave light frequency of both without transmitting the reference light frequency w1 and the signal light frequency w2 ( (2 × w1-w2) or (2 × w2-w)
1)) including a wavelength filter that transmits light and an optical receiver having high sensitivity in the vicinity of a fundamental wavelength, wherein the signal comparing unit generates and outputs the four-wave mixing signal light by an overlapping component of two optical pulse trains to be compared. An optical pulse timing detection circuit for measuring a change in a time positional relationship of the signal light pulse train by using an increase or decrease of the signal light pulse train. 20. The optical pulse timing detection circuit according to claim 19, wherein said timing discriminating means parallelly outputs electric outputs of said two optical pulse cross-correlation receiving means for comparing a variation of an overlapping component with each pulse train. An optical pulse timing detection circuit including a connected parallel optical receiver, wherein a difference between outputs of the two parallel optical receivers is used for identifying a time position fluctuation direction. 21. The optical pulse timing detection circuit according to claim 20, wherein the timing discriminating means compares a difference between electric outputs of the two optical pulse cross-correlation receiving means for comparing a variation of an overlapping component with each pulse train. The fractional fraction is used to identify the temporal position variation direction, the signal output power is estimated by summing the electric outputs, and the temporal position direction and the variation amount are calculated by dividing both of the electric outputs. An optical pulse timing detection circuit for detecting both. 22. An M system having a repetition frequency f ₀ (M is 2
Using the delay control means for controlling the respective timings of the optical pulse trains of the above (integer) and the optical multiplexing means for multiplexing the optical pulse trains of the M system, the optical pulse trains of the M system are equally spaced on the time axis. In an optical time-division multiplexing apparatus that multiplexes and generates an optical pulse train having a repetition frequency f ₀ , a branching unit that branches each of the M optical pulse trains, one of the optical pulse trains branched by the branching unit, 22. An optical pulse train of the M system, comprising: an optical pulse train separately prepared as a reference optical pulse train, and detecting a timing of an optical pulse train of the remaining system. Of the respective time positions are detected by the optical pulse timing detection circuit, and the delay control means detects the time position fluctuations of the respective systems in accordance with the detection results. Optical time division multiplexing apparatus characterized by a feedback control so that removed by. 23. The optical time division multiplexing device according to claim 22, wherein only one optical pulse timing detection circuit is provided in one system, and a timing signal for sequentially selecting the branched optical pulse trains. A selection timing controller that generates the plurality of optical pulse trains, and sequentially selects the plurality of optical pulse trains in a time-division manner using the timing signal of the selection timing controller; An optical channel selection switch to be supplied to a detection circuit, wherein the output of the optical pulse cross-correlation receiving means is distributed to each target series by an electric signal selection switch interlocked with the optical channel selection switch as a delay control signal,
An optical time division multiplexing device which is sequentially fed back to the corresponding delay control means.