JP2826325B2 - Optical information processing apparatus and optical information processing method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical information processing apparatus, and in particular, reads information spatially arranged in parallel using an ultrafast optical pulse and The present invention relates to a device for converting a pulse train into a series.

[Conventional technology]

Since light can simultaneously access a large number of two-dimensionally arranged points using its spatial coherency, optical information processing with a unique arithmetic function that is not available in ordinary computers using semiconductors The device can be formed.

FIG. 2 shows a conceptual diagram of this type of conventional apparatus. The light emitted from the laser light source 1 is expanded in diameter by a beam expander 2 and then incident on several stages of spatial light modulators (3, 4, etc.). This spatial light modulator refers to a device that can change the light transmittance at a point corresponding to an nxm matrix in general by an electro-optic effect or the like. By passing through at least two or more spatial light modulators,
Basic operations such as OR (OR) and more advanced operations (differentiation etc.) corresponding to image processing can be executed at the speed of light. The result is incident on the matrix of the photodetector 5, and the intensity of light incident on each point is converted into an electric quantity. If a high-speed laser pulse is used, an electric pulse will be output.

However, when an output from this type of device is input to a normal computer or transmitted by an optical fiber, a large amount of parallel information is converted into a time series. Is required.

FIG. 3 shows a conceptual diagram thereof. Light (image) 11 is applied to the photoelectric conversion surface 12 of the image pickup tube, and the charges generated at each point are scanned zigzag with the electron beam 13 and read out as time-series signals. Moreover, if the output shown in FIG. 2 is read out by the apparatus shown in FIG. 3, the information calculated at the speed of light will be significantly slowed down because the parallel-serial conversion unit becomes a net.
In addition, related to this kind of prior art include Applied Optics, Vol. 27, p. 1661 (1988) or Vol. 27, p. 1674 (1988), and the like.

[Problems to be solved by the invention]

The present invention provides a method for rapidly converting a large amount of spatially parallel information from an optical information device into a time-series pulse while keeping the light as described above. Also, the present invention
An ultra-high-speed modulator for transmitting information with ultra-high-speed pulses of picosecond to femtosecond is also provided.

[Means for solving the problem]

 The configuration of the present invention will be described based on the embodiment shown in FIG.

The ultrahigh-speed laser light pulse from the laser light source 1 is spatially expanded in diameter by the beam expander 2, and then enters the n × m matrix spatial light modulator 3 as a plane wave.
The light having passed through is incident on the spatial light modulator 4 as an n × m light pulse bundle having a small diameter. This is because each phase of the matrix has its phase adjusted so that a delay time of τ seconds occurs, and when it is incident on the optical fiber 61 by the lens 51, the nm pulses are transmitted in the time interval It will enter one after another at τ.

In order for this configuration to be actually possible, the phase modulation plate 4
In this case, the delay of nm × τ seconds that should be generated must be realized, for example, as a change in the film thickness of the medium. This is relatively easy if the laser pulse is a picosecond to femtosecond ultrafast pulse, and in this case it can be an ultrafast optical computer different from a conventional electrical computer.

In the case of laser light pulse on the order of 10 picoseconds, the fourth
As shown in the figure, a practical method is to make the optical pulse after passing through the spatial light modulator 3 enter the optical fiber matrix 7 at the position of the screen 52.

The length of this optical fiber is set so that each of the n ² fibers has a delay of time τ seconds (10 pico seconds), and the fiber output is output by a lens 53 to an ultra-high-speed photodetector.
It is incident on 62 and converted into an electric pulse train.

In each of FIGS. 1 and 4, the spatial light modulator 3
If the light transmittance of each point of the matrix is set to either 1 or 0, the input light pulse to the optical fiber of FIG. 1 and the input light pulse to the photodetector 62 of FIG. A pulse train with a value of 0 is obtained.

The clock operation in the configuration of the present invention will be described with reference to FIG. As shown in FIG. 2A, a pico (10 ^-2 sec) second pulse is repeatedly emitted from a laser light source (not shown) with a clock time t _{0 of} about nano (10 ^-9 sec) second. This pulse is incident on the spatial light modulator 3 of an n × m matrix as a plane wave, and then is incident on the optical fiber 7 with a delay of τ seconds at each point. FIG. 5 shows an example of n-3 and m-4. That More No Yes 12 pulses as shown in FIG (b), the 12 bit optical pulse train to be 1 mass repeatedly occurs every t _0. Therefore, if it is possible to update the light transmittance of the spatial light modulator during the time t _0, which operates as an optical modulator of the transmission rate B ₀ 1 / tau (pits / sec).

〔Example〕

Hereinafter, the contents of the present invention will be described in more detail based on examples.

1 in FIG. 1 is 1p for each repetition time t ₀ (1 to 10 msec).
It is a laser light source that can emit an ultrashort light pulse with a width of sec to 10 psec. Since this output light has sufficient spatial coherency, at least 10
It can be expanded to a beam with a diameter of mmφ. This light is vertically incident on the light modulator 3 in the space, and each light transmission window is irradiated with a plane wave pulse of the same intensity. The light pulse that has passed through the spatial light modulator 3 is converted into a light pulse bundle having a small diameter corresponding to the diameter of the window by an optical gate function of modulating the transmittance of the window by 1.0. To avoid diffraction effects, the window opening should be 50 compared to the wavelength λ.
What is necessary is just to make it as large as about λ. The information of the spatial light modulator 3 is
There is a need to be rewritten to erase during the time t ₀ of until the next laser pulse. If a high-speed GaAs or InP-based electric circuit is used, the above rewriting can be sufficiently performed within t ₀ to ₁₀ nsec.

The light pulse bundle whose amplitude has been modulated to 1,0 in this way enters the spatial light modulator 4. Here, there is a window whose thickness is changed so that a delay time τ (second) is generated for each phase of the light pulse. This light is converged on the optical fiber 61 by the lens 51. At this time, an optical pulse corresponding to 1 or 0 enters the optical fiber at every time interval τ seconds, and this can be transmitted as a time series signal as it is.

If the spatial modulator is an n × m matrix, there will be n × m × τ pulses during the clock time t ₀ . For this to be sufficiently possible, the full width at half maximum t ₁ (second) of the original laser beam must satisfy the relationship of t ₁ <τ≪t ₀ .

If the delay τ is on the order of 1 picosecond, it corresponds to a thickness difference of about 100 μm when a substance having a refractive index of 3 is used. When τ is large, the spatial phase modulation plate 2
As shown in FIG. 4, it is easier to temporarily introduce each pulse into the fiber matrix 7 so that the length of each fiber causes a delay time τ. If the delay τ is 10 picoseconds, a fiber with a refractive index of 1.5 corresponds to a length difference of about 2 mm. After the output of the optical fiber delay line is aligned on a certain plane, it is narrowed down to a high-speed photodetector 62 (shown simply by one lens 53 in FIG. 4) with an appropriate lens system and converted into an electric pulse train. . Alternatively, input is made to the optical fiber as in FIG.

〔The invention's effect〕

As described above, the apparatus of the present invention can easily convert parallel information of a 10 × 10 matrix into 100 time-series pulses with a clock time of 1 to 10 nanoseconds. This is effective as clock light pulse width t ₁ is narrowed less than 10 picoseconds.

[Brief description of the drawings]

FIG. 1 is a conceptual diagram of an optical information processing apparatus according to one embodiment of the present invention,
FIG. 2 is a conceptual diagram of a conventional optical information processing device, FIG. 3 is a conceptual diagram of a parallel-serial conversion device corresponding to a conventional image pickup tube, and FIG. 4 is a device of a second embodiment of the present invention. FIG. 5 is a conceptual diagram illustrating the clock operation of the optical information processing apparatus according to the present invention. DESCRIPTION OF SYMBOLS 1 ... Light source, 2 ... Beam expander, 3 ... Spatial light modulator, 4 ... Spatial phase modulation plate, 5 ... Photodetector row, 7
... Optical delay line.

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Tomohiro Ohno 1-280 Higashi Koigakubo, Kokubunji-shi, Tokyo Inside the Central Research Laboratory of Hitachi, Ltd. (56) References JP-A-63-129326 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G02F 1/01 G02F 3/00

Claims (4)

(57) [Claims] An optical pulse having a full width at half maximum of t ₁ is repeated in a time interval.
The light pulse from the light source for generating at t _0, after being incident on the plurality of light modulating windows that are spatially arranged, each of said plurality of light modulator windows this irradiated light pulse is subjected to desired modulation And an optical information processing apparatus having an optical system that generates a time-series optical pulse train after each of these optical pulses propagates with a predetermined delay time, wherein the optical modulation rate of the optical modulation window is updated. The time required for the optical information processing apparatus is larger than the full width at half maximum t ₁ of the light pulse and smaller than the repetition time t ₀ . 2. The method according to claim 1, wherein the light pulses are applied to a plurality of spatially arranged light modulation windows in an n × m matrix, and then a plurality of light pulses from each of the light modulation windows are determined. The phase of the propagation medium is adjusted so as to reach a certain point, which is a delay time, and a plurality of light pulses from the plurality of light modulation windows are respectively n × m time series having a delay of time τ seconds. When converted to an optical pulse train, the incident optical pulse is converted to t ₀ > n ×
An optical information processing apparatus, wherein light is incident at a clock time of m × τ (seconds), and each of the optical modulation windows is light-modulated as desired in synchronization therewith. 3. A repeated time interval of light pulses of the full width at half maximum t ₁
The light pulse from the light source for generating at t _0, after being incident on the plurality of light modulating windows that are spatially arranged, each of said plurality of light modulator windows this irradiated light pulse is subjected to desired modulation And these light pulses are propagated with a predetermined delay time, respectively, and then an optical information processing method for generating a time-series light pulse train, wherein the time required for updating the light modulation rate of the light modulation window is , greater than the full width at half maximum t ₁ of the optical pulse, the optical information processing method characterized by repeated less than the time t _0. 4. A light pulse is applied to a plurality of spatially arranged light modulation windows, and the irradiated light pulse passes through each of the plurality of light modulation windows after being subjected to a desired modulation. An optical information processing method in which these light pulses transmitted through each light modulation window are subjected to a predetermined delay to generate a light pulse train based on these light pulses, wherein the light pulse is transmitted to the plurality of light modulation windows. An optical information processing method, wherein the optical modulation of the optical modulator is updated during a period of repeated irradiation.