JPH0433019A - Optical information processor - Google Patents

Abstract

PURPOSE:To obtain an optical memory device capable of handling a time sequence signal, by providing plural optical memory cells to store information according to the stored amount of carriers and a means to transfer, exchange or calculate the stored carriers among these optical memory cells. CONSTITUTION:Plural optical memory cells M1-M9 are provided to store the information according to the stored amounts of the carriers, and the means is provided to transfer, exchange or calculate the stored carriers among these optical memory cells. Namely, since the carrier showing the information is moved among the optical memory cells M1-M9, the information can be transferred, delayed and stored among the optical memory cells. On the other hand, since the time sequence signals can be hourly inverted when transferring the temporarily stored time sequence information backward, the computing element of collation and convolution can be realized not only as an optical memory or a delay line. Thus, the information can be transferred among the optical memory cells M1-M9, the time sequence signals can be stored and delayed and further, the optical information processor suitable for integration can be obtained.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an optical information processing device, and in particular, an optical information processing device capable of storing, delaying, or time-reversing time-series optical input information. It is also suitable for optical information processing devices suitable for integration.

[Prior art] Fig. 4 (a> is, for example, The 5ociety of
Photo-Optical Instrument
P of tation Engineers (SPIE)
roceedings vol, 963, pp, 25
5-259 is a plan view showing a conventional optical information processing device, and FIG. 4(b) is a cross-sectional view of each optical memory element. (3) and (4) are the anode and cathode electrodes respectively connected to the optical memory element (2), (5) is the n-Ga layer provided on the substrate (1).
As layer, (6) is n-AlGaAs layer provided in a part of this n-GaAs layer (5), (7) is this n-AlG
f)-AlGaAs layer provided on the aAs layer, (8>
is the n-GaA provided on this p-AlGa'As layer.
The s layer is a p-G layer provided on this nGaAs layer.
The aAs layer, <10) is a p-GaAs layer provided on this p-GaAs layer, (11) is input light, and (12) is output light. The anode electrode (3) is a p-GaAs layer (10
), and the cathode electrode (4) is connected to the n-GaAs layer.

Next, the operation of the conventional optical information processing apparatus shown in FIG. 4 will be explained. First, the operation of the pnpn optical memory element shown in FIG. 4(b) will be explained. As the voltage between the anode electrode (3) and the cathode electrode (4) is increased, pn
Initially, the pn optical switch is in an off state where no current flows, but
When the voltage becomes larger than a certain value (switching voltage, for example, 4V), the switch is turned on, a large current flows, and light is emitted.

Therefore, if a voltage slightly smaller than the switching voltage (for example, 3.5 V) is applied, the switch will turn on if there is input light (11) from above. Even after this voltage is set to zero, a positive refresh pulse (1, OV) is applied at a constant period (for example, 10 μs).

ions) to the anode voltage! If applied between fl(3) and the cathode electrode (4), the on state is maintained.

Then, when a large voltage (3,5 V) is applied again, if the element is in the on state, it emits light.6 Resetting from the on state to the off state is executed by applying a negative pulse voltage. Memory operation is achieved in this way.

FIG. 4(a) shows the pnpn optical memory elements arranged in a two-dimensional array. Information can be stored in any optical memory element by moving the input light (11). Also, i
When the anode electrode (3) in the row is set to +1.8V and the cathode electrode (4) in the jth column is set to -1°7■, the coordinates (i, j)
The information of the corresponding optical memory element is read out.

[Problems to be Solved by the Invention] Since the conventional optical information processing device is configured as described above, in order to store time-series optical information, a device for branching or deflecting the input light must be installed in the input section. In addition, there was a problem in that information in one optical memory element could not be transferred to an adjacent optical memory element.

This invention was made to solve the above-mentioned problems, and is an optical information processing method that can transfer information between optical memory devices, store and delay time-series signals, and is suitable for integration. The purpose is to obtain equipment.

[Means for Solving the Problems] An optical information processing device according to the present invention includes a plurality of optical memory elements that store information based on the amount of accumulated carriers, and transfers, exchanges, or calculations of accumulated carriers between these optical memory elements. It is equipped with means.

[Operation] In this invention, since carriers representing information are moved between optical memory elements, information transfer between optical memory elements,
Therefore, delay and storage become possible.

Furthermore, by transferring the stored time-series information in the opposite direction, the time-series signal can be time-reversed, so that it can be used not only as an optical memory or a delay line, but also as a correlation and convolution arithmetic unit.

Embodiment] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. 1st
The figure is a sectional view showing one embodiment of the present invention, and (6) to
(9), (11), and (12> are the same as those described above. In the figure, (13>) is an electrode, (14) is a GaAs substrate to which this electrode (13) is attached, and (15)
is an n-AlGaAs layer (
6) SiO□ layer provided through the layer, (16) is this S
A plurality of polysilicon electrodes are provided at predetermined intervals on the iO2 layer (15), (17) is a light output part, and Mi (1≦i≦9
) is an optical memory element, φi (1≦i≦9) is a control electrode of each optical memory element Mi, and φ1o is a drive electrode of the optical output section (17). Note that the boundary portion between the optical memory elements constitutes a means for transferring, exchanging, or calculating stored carriers.

Next, the operation of the embodiment of the present invention shown in FIG. 1 will be explained. Information is assumed to be a digital value of "1" and O. There are three modes of operation: write, transfer, and read. First, we will discuss the write mode. A negative voltage (for example, φ + -2V)
Apply.

Then, a depletion layer is formed at the bottom, and when there is input light (11) in this state, carriers are generated in this depletion layer,
Holes are accumulated near the SiO2 layer (15) of the n-AlGaAs layer (6). When there is no input light (11), the storage hole is zero.

In this way, information is written as the amount of stored carriers. If φ is left at -2V, the depletion layer is thicker under the control electrode φ than in the periphery, so a potential well is created for the holes. As a result, the information is retained.

Next, the transfer mode will be explained with reference to FIG. Assume that holes are accumulated in the optical memory element M1 as shown in FIG. 2<a). At this time, the control electrode φ+=2V, and the control electrode φ,=Ov.

To transfer this hole to the optical memory element M2, see Figure 2 (
As shown in b) and (c), the voltage of the control electrode φ1 is set to -2V.
to 0■, and the voltage of the control electrode φ2 is shifted from OV to 12■. Then, the potential well moves from the optical memory element M1 to M2, and at the same time, the accumulated carriers also move. Information is transferred in this way.

Finally, we will discuss the read mode. Light output section (17
) has a pnpn thyristor structure.

First, a voltage slightly smaller than the switching voltage is applied to the drive electrode φ1°. At this time, the p-GaAS layer (7)
There is a well for the hole.

Next, the information of the optical memory element M next to the optical output section (17) is transferred to the optical output section (17). If the optical memory element M is "1", holes flow into the p-GaAs layer (7) and the thyristor is switched on. And it emits light. Conversely, if the optical memory element M is "0", the thyristor remains off and does not emit light. Information is read out in this way.

As described above, in this embodiment, since information can be transferred between each optical memory element, time-series signals can also be stored. Furthermore, since threshold processing is performed by the thyristor of the light output section (17), it is possible to solve the problem of reduction in holes that occurs during information transfer, that is, hole transfer. .

Incidentally, in the above embodiment, the circuit was operated as an optical memory, but it can also be operated as an optical delay circuit.

For example, Figure 5(a) is from the Fall 1989 Proceedings of the Japanese Society of Applied Physics, Vol. 3, pP, 789. This is a conventional optical delay circuit shown in lecture number 30aZD-1. When light enters from one end (21) of the optical fiber (20), the light enters the optical fiber (
20) and exits from the other end (22) with a delay of just enough time to pass through the terminal (20). As a result, as shown in FIG. 5(b), the predetermined time t
The delay is obtained. However, in the case of this conventional optical delay circuit, the device is large, and to obtain a delay of 5 μs, for example, I
There was a problem that Km of optical fiber was required.

Therefore, the operation of this embodiment when an optical memory is applied to an optical delay circuit will be described. The configuration of the device is the same as that shown in FIG. When input light (11) is input, the information is sequentially transferred to the adjacent optical memory element. Once all the information has been input, the shift is stopped and the desired time information is stored. Thereafter, the information is sequentially sent to the optical output section (17) and read out as output light (12). In this way, optical delay is obtained. Note that the delay time can be set arbitrarily.

Next, the time reversal mode will be explained with reference to FIG. First, optical information X l +
2+..., Xn are input sequentially, and the input optical information is transferred to the adjacent optical memory element (Fig.
). Transfer is stopped when all optical information has been input. Each optical memory element M8. ...1Mo each has optical information X
. , . The optical information sent to the optical output section (17) is output as an optical signal in the order of X. and Xnxl (Fig. 3(c)).
In this way, a time-reversed signal of the input signal is obtained as an output signal.

Note that if it is used as a simple delay circuit, for example, convolution or time reversal mode can be used to perform correlation calculations. Furthermore, it is obvious that the present invention can be used as a normal memory, such as a dynamic memory.

[Effects of the Invention] As described above, the present invention includes a plurality of optical memory elements that store information based on the amount of accumulated carriers, and a means for transferring, exchanging, or calculating the accumulated carriers between these optical memory elements. As a result, an optical memory device that can handle time-series signals can be realized, and since it has a chip structure, it can be integrated, making it possible to obtain a small, stable, and inexpensive device, and it is also possible to obtain a delay circuit for optical signals at the same time. This has the effect of providing an optical information processing device.

[Brief explanation of drawings]

FIG. 1 is a sectional view showing an embodiment of the present invention, FIGS. 2 and 3 are sectional views for explaining the operation of the invention, and FIG. 4 is a plan view showing a conventional optical information processing device. cross section,
FIG. 5 is an explanatory diagram of a conventional optical delay circuit. In the figure, (6> is an n-AlGaAs layer, (7) is a p-AlGaAs layer, and (7) is a p-AlGaAs layer.
-GaAs layer, (8) is n-GaAs layer, (9) is p-
AlGaAs layer, (13) is electrode, (14) is GaAs
The substrate, (15) is a 3402 layer, (16) is a polysilicon electrode, (17) is a light output part, and Ml is an optical memory element. In the drawings, the same reference numerals indicate the same or corresponding parts.

Claims (1)

[Claims] Optical information processing comprising: a plurality of optical memory devices that store information based on the amount of accumulated carriers; and means for transferring, exchanging, or calculating the accumulated carriers between these optical memory devices. Device.