JPH03100526A - optical switch array - Google Patents

Abstract

PURPOSE:To reduce the insertion loss of a semiconductor waveguide type optical switch array by constituting the optical switch array by combining a circuit brought to crossbar type connection with a branching circuit. CONSTITUTION:Each input terminal is allowed to branch into two pieces by a branching circuit 7, respectively, passes through an NXN switching unit 6, and also, uses a branching circuit 8 as an optical multiplexing circuit and reaches an output terminal. Consequently, the inevitable loss of 3dB and an excessive loss (r) in the branching circuits are generated two times in branching and optical multiplexing, but the terminal passes through the NXN optical switch unit 6 only once. In such a manner, as long as the insertion loss of the NXN optical switch unit 6 is larger than the insertion loss of all the branching circuits, the insertion loss of a 2NX2N optical switch array can be reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a semiconductor waveguide type optical switch array for realizing an optical switch.

[Conventional technology]

Semiconductor waveguide optical switch arrays for realizing optical switching equipment have been previously described in IEEE, Journal of Selected Areas Communication, and Nisneecy 6 (IEEE J, 5ele).
ct, Areas Commun, SAC6)19
1988, pages 1262 to 1266. In the above technique, semiconductor waveguide optical switch units are connected in a crossbar configuration to form a 4×4 optical switch array that can be operated by 4 people and has 4 outputs.

[Problem to be solved by the invention]

The above-mentioned conventional technology has a large insertion loss of light waves and does not give any consideration to an optical switch array having a large number of input/output terminals. In order to clarify the problem, the second
This will be explained using the diagram and FIG. FIG. 2 is a schematic diagram of a conventional semiconductor waveguide type 4×4 optical switch array 4. As shown in FIG.

16 optical switch units 1 are connected in a crossbar type, and 4
It is configured to have a human power output of 2.4 and 3. The insertion loss of the above optical switch array is as follows:
It can be classified into a coupling loss δ at the bending portion 5, a bending loss E at the bending portion 5, a propagation loss of the light wave, and a scattering loss at the branching portion of each optical switch. As shown in Figure 2, the maximum insertion loss value occurs in the above configuration when the light wave passes through the path between the input 24 and the output 34, and at this time, the path length and the number of passing branches are maximum. become.

If the number of input and output terminals of the optical switch array is increased using the same configuration as the conventional example, the path length and the number of branches will monotonically increase, and therefore the insertion loss will also monotonically increase due to the increase in propagation loss and scattering loss. FIG. 3 shows the above state, in which the horizontal axis represents the number of channels M and the vertical axis represents the insertion loss.

The loss in the figure is an illustration of the insertion loss in a path showing the worst insertion loss of an optical switch array with NXN crossbar type connection. The four lines shown in the figure are.

It shows the characteristics when the value of the scattering loss β at the branch part is changed, and here, the coupling loss δ and the bending loss ε are calculated as 7 dB and 2 dB, respectively.

As shown in Figure 3, as the number of channels (number of inputs and outputs) of the optical switch array with crossbar connections increases, the insertion loss increases dramatically, and when the scattering loss β = 1.5 dB in the conventional example, N =
At 32, it reaches 130dB. N at β=0.5dB
=32, the insertion loss is about 70 dB, which is not practical at all.

An object of the present invention is to obtain a multi-input multi-output optical switch array with low insertion loss.

[Means to solve the problem]

The above object is to provide an optical switch array having multiple inputs and multiple outputs in which waveguide type optical switch units having a function of switching the output end of an incident light wave according to an electrical control signal are coupled in multiple stages. This is achieved by constructing the array by combining crossbar-connected circuits and branch circuits.

[For production]

The operation of the present invention will be explained with reference to FIG. Figure 1 is NXN
2 using optical switch array unit 6 and branch circuit 7°8
This diagram schematically shows the configuration method when configuring an NX2N optical switch array. 6 Each input terminal is connected to a branch circuit 7.
The signal is branched into two, passed through an NXN optical switch unit, and further reaches the output end using the branch circuit 8 as a multiplexing circuit.

As a result, the unavoidable 3 dB loss in one branch circuit and the excess loss γ will occur twice in the branch and gold wave, but NXN
Since the optical switch unit 6 is passed only once, the insertion loss of the 2NX2N optical switch array is reduced as long as the insertion loss of the NXN optical switch unit is greater than the insertion loss of all the branch circuits. It is obvious from Fig. 1 that in order to construct a 4N It is.

Figure 4 shows the insertion loss when configuring a 32 x 32 optical switch array with multi-stage connections of various NXN optical switch array units and branch circuits using the above configuration method. It shows the results calculated using the same parameters. The horizontal axis shows the number of channels N, and the vertical axis shows insertion loss.

In addition, in the figure, the line connecting the X points shows the inevitable change in loss that occurs in the branch circuit. From Fig. 4, when β is 1.5 dB, the insertion loss of the 32x32 optical switch array is 41.5 dB using a 2x2 optical switch array unit and a branch circuit, and when β = 0.5 dB, the insertion loss is 41.5 dB.
It is shown that the insertion loss can be reduced to 35.7 dB by using the x4 optical switch array unit and the branch circuit.

In other words, compared to simply configuring a 32x32 optical switch array connected with a crossbar, the
dB insertion loss is reduced. With this level of insertion loss, it is sufficient to make a practical optical switch array by slightly improving the coupling loss, bending loss, and propagation loss, or by compensating the loss by about 10 to 20 dB using an optical amplifier, etc. can do. Although the configuration of a 32 x 32 optical inch array has been described here, it goes without saying that similar effects can be obtained for optical switch arrays of larger or smaller scale. However, the input/output fraction M
If the size of the switch is less than 16, it is difficult to expect any industrial advantage even with the optical switch array of the present invention. Because it consists of an array of electronic switches, another advantage is that it is compatible with electronic control circuits when the purpose is to replace the current electronic exchange with an optical exchange and dramatically expand its functions. . That is, the fact that the optical switch array constructed by this construction method produces a 32×32 optical switch array with practical performance is advantageous in a new sense.

As is clear from Figure 4, when an MXM optical switch array is configured using this configuration method of multi-stage connection of NXN optical switch array units and branch circuits, there is an optimal value of N that minimizes insertion loss. As a result of various studies, the inventors found that the optimal value of N is an integer value closest to the value of n given by the following equation.

22, α is the propagation loss per optical switch, β is the branching loss, γ is the excess loss of the branching circuit, and the value of n does not depend on M. In addition, although the values of N and M are integers that are powers of 2 to facilitate the above configuration, this is not necessary.

〔Example〕

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing the basic concept of the present invention and the configuration of an embodiment, and FIG. 4 is a diagram showing the relationship between the insertion loss and the value of N when a 32×32 optical switch array according to the present invention is configured in NXN optical switch array units. FIG.

First Example Each optical switch array unit 6 of 2×2.4×4.8×8 was manufactured using the same technique as in the conventional example.

Optical fiber type 3 dB couplers (7 and 8) are used as optical branching circuits that branch the light emitted from the optical input end to multiple output ends to create a 4X4.8X configuration shown in Figure 1.
Each 3.16x16 optical switch array was fabricated. Based on the results of measuring the insertion loss of each optical switch array above,
The insertion loss reduction effect explained with reference to FIGS. 3 and 4 shown above was confirmed.

Second Example This example was fabricated by integrating the 3cLB coupler used in the first example above with an NXN optical switch array unit on a semiconductor substrate. The effects obtained by this example were similar to those shown in the first example.

In addition, in the above embodiment, a 3dB coupler was used as the branch circuit, but when branching into many branches such as 4 branches or 8 branches,
It goes without saying that a star coupler may also be used.

It goes without saying that the optical switch array described above can be used as an optical switching processor, but it goes without saying that an optical switching system can be realized using the optical switch array used as the optical switching processor.

〔Effect of the invention〕

As described above, the optical switch array according to the present invention is an optical switch array having multiple inputs and multiple outputs, in which waveguide type optical switch units having the function of switching the output end of an incident light wave are coupled in multiple stages according to an electrical control signal. In the switch array, by configuring the above-mentioned optical switch array by combining circuits with cross-spar type connections and branch circuits, the insertion loss of semiconductor waveguide type optical switch arrays can be dramatically reduced. It is possible to realize an optical switch array with multiple inputs and multiple outputs that is integrated into a multi-input device with practical performance.

[Brief explanation of drawings]

Fig. 1 is a diagram showing the basic concept and configuration of an embodiment of the present invention, Fig. 2 is a schematic diagram of an N x N crossbar type connected optical switch array according to the prior art, and Fig. 3 is a schematic diagram of an MX
Figure 4 shows the relationship between the insertion loss and the value of M when an M optical switch array is configured, and FIG. FIG. 1... Optical switch unit 2... Input end 3... Output end 6... NXN crossbar type connection optical switch array 7.
8...branch circuit

Claims (1)

[Claims] 1. In an optical switch array having a multi-input multi-output terminal in which waveguide type optical switch units are coupled in multiple stages and has a function of switching the output terminal of an incident light wave according to an electrical control signal, The optical switch array is characterized in that the optical switch array is configured by combining circuits and branch circuits connected in a crossbar type. 2. The optical switch array according to claim 1, wherein the waveguide type optical switch unit is made of a semiconductor material. 3. The optical switch array according to claim 2, wherein the semiconductor optical switch unit operates by utilizing a change in refractive index due to carriers injected into the semiconductor. 4. The optical system according to claim 1, wherein the combination of the crossbar type connection circuit and the branch circuit is selected so that the total insertion loss of the optical switch array is minimized. switch array. 5. In the above optical switch array, the number of input terminals and output terminals is M each, and the optical switch units are connected in a crossbar type so that there are N inputs and N outputs, and between the above M and N. An optical switch array characterized in that the relationship M≧N holds true. 6. For the above optical switch unit, the propagation loss of the light wave within the unit is α dB, the branching and scattering loss within the unit is β dB, and the branch circuit is a one-to-one branch circuit, and the excess loss is γ dB. , the value of the number of input/output terminals N for each optical switch connected in the crossbar type is expressed by the following formula: n
The optical switch array according to claim 5, wherein the optical switch array is an integer closest to . n=2(3+γ)/(α+2β)l_n2 7, the value of N is an integer expressed as a power of 2 closest to the value of n, as described in claim 6. Optical switch array. 8. The optical switch array according to any one of claims 1, 4, and 5, wherein at least a portion of the optical switch array is formed of a semiconductor material. 9. The optical switch array has an input/output fraction M of 16.
An optical switch array according to any one of claims 1, 4, and 5, characterized in that the above is the case. 10. The optical switch array according to claim 9, wherein the input/output fraction M is particularly 32. 11. An optical switching processor comprising the optical switch array described above and an electrical control circuit. 12. The optical switching processor according to claim 11, wherein the electrical control circuit is an M×M switching circuit according to the number M of input and output terminals of the corresponding optical switch array. 13. The optical switching processor according to claim 12, wherein the input/output fraction M has a value of 16 or more. 14. The optical switching processor according to claim 12 or 13, wherein the input/output fraction M has a value of 32. 15. An optical switch using the above-mentioned optical switch processor.