JP4074565B2 - Optical signal pattern identification method and identification apparatus - Google Patents

Description

  The present invention relates to an optical signal pattern identification technique, and particularly in an optical packet switch network that is expected to be promising in the near future, an optical label (header) having address information of an input optical packet matches a local address signal of each node. The present invention relates to an optical-optical type optical signal pattern identification method and an optical signal pattern identification apparatus for quickly determining whether or not to perform.

  In recent years, with the explosive increase in data communication represented by the Internet, there has been an increasing demand for high-speed optical signals. However, after converting an optical signal of 10 Gbps or more into an electrical signal by a light receiving element, it has become a problem to process the electrical signal as it is by a conventional electronic circuit. For example, in optical packet communication, a router or the like decodes address information contained in an optical packet label to determine an output port, or delays the packet signal for an arbitrary time to avoid collision between optical packets. However, since the label recognition process and the memory processing function are conventionally constituted by a silicon-based LSI (Large Scale Integrated Circuit), the operation speed is 1 Gbps or less. End up.

  Therefore, a high-speed optical packet signal is converted into an electric signal by an O / E (optical / electric) receiving circuit using a light receiving element, and an electric clock signal is generated from the electric signal using InP or GaAs high-speed electronic circuit technology. A method is conceivable in which a clock signal is extracted by a device, and a high-speed electric signal is converted into a plurality of low-speed electric signals in parallel by the electric serial-parallel converter using the clock signal, and then label recognition is possible. However, in such a method, since clock generation and serial-parallel conversion all depend on an electronic circuit, a speed of about 40 Gbps is considered to be the limit. Further, when serial-parallel conversion is performed by InP or GaAs high-speed electronic circuit technology to convert a plurality of low-speed signals, the high-speed electrical signal is sequentially divided into half speeds (40 GHz → 20 GHz →. Therefore, a considerable number of stages are required, and problems such as clock extraction and phase control in each stage also occur. In addition, when these electronic circuits are used, the overall power consumption is expected to be considerably large. Moreover, in conventional clock extraction using an electronic circuit, it is necessary to apply feedback by a PLL (Phase Locked Loop) and lock the oscillation frequency of a VCO (Voltage Control Oscillator). Therefore, it is impossible to extract a clock instantaneously for a packet signal that is input automatically.

  As a method of decoding the address information of the high-speed optical label, all the bits of the optical label are converted into a low-speed electrical signal by the serial-parallel conversion described above, and then processed by the silicon LSI, and the optical label of the input optical packet There is a method for discriminating whether a node matches a local address held by a node. The latter method is more convenient for a bypass / drop device that determines whether an input optical packet is dropped and transferred to another network or passed (bypassed) at each node of the ring network. It is advantageous.

  As a pattern matching method for comparing the pattern of the optical label with the pattern of the local address signal, a method using an AND (logical product) circuit such as an electro absorption type modulator (EAM) can be considered. That is, the optical label is modulated by the electric signal of the local address when passing through the EAM set to normally-on. At this time, if the optical label is “1” and the local address signal is “1”, the optical signal is not output, and the optical signal is output only when the optical label is “1” and the local address is “0”. . That is, when the output optical signal from the EAM is “0”, it can be determined that the pattern of the optical label and the local address match, whereas when the output is obtained, the pattern of the optical label and the local address is determined. It can be judged as a mismatch. However, with this method, even when the optical label is “0” and the local address is “1”, there is a problem that the output light is not obtained despite the mismatch, and it is determined that they match. For this reason, this method has a restriction that the number of “1” bits of the optical label and the local address must be equal to each other. Furthermore, since the operation speed is limited by EAM, the limit is about 40 Gbit / s at present.

  On the other hand, an optical fiber loop was used as a method of overcoming the limitation on the number of “1” s that can be used in addresses and the limitation on the operation speed due to electrical signals, which are caused by using the AND circuit as described above. An all-optical pattern matching method using a non-linear optical loop mirror (NOLM) has been studied (see Non-Patent Document 1). In this method, since an exclusive OR (EX-OR) type logical operation is used instead of an AND type, the number of “1” is not limited, and an arbitrary optical label can be assigned. Furthermore, pattern matching of high-speed optical labels exceeding 40 Gbit / s is possible by all-optical processing.

K. L. Hall et al. “All-optical bit pattern generation and matching” Electron Letters Vol.32, No.13 P.1214-1215 (1996-7) Tatsushi Nakahara, Ryo Takahashi et al. "Label recognition and self-routing of ultrafast optical packets by self-synchronous all-optical serial-parallel conversion" IEICE Technical Report EMD2002-35, CPM2002-77, OPE2002-57, LQE2002-112 P.25 -29 (2002-08)

  As described above, in the pattern matching using the AND type circuit, the number of “1” of the optical label is limited, and further, the operation speed is limited by using the electric signal. Furthermore, in the EX-OR type pattern matching using an optical fiber, there is no restriction on the number of “1” and the operation speed is extremely high, but there is a problem that the entire apparatus becomes large.

  SUMMARY OF THE INVENTION An object of the present invention is to solve the above-described problems, and to realize an optical-optical type optical signal pattern identification method and an optical device capable of identifying an all-optical pattern of an EX-OR type that is extremely small and capable of operating at an ultra high speed. An object is to provide a signal pattern identification device.

In order to achieve the above object, an invention of an optical signal pattern identification method according to claim 1 is characterized in that a first circularly polarized optical signal composed of optical pulses of n bits (n is an integer of 2 or more) and the first circularly polarized light. A second circularly polarized light signal having circular polarization opposite to that of the optical signal and also comprising an n-bit optical pulse and an optical probe comprising the same n-bit optical pulse are respectively represented as i-th (i is 1). All integers from 1 to n) are synchronized and irradiated to one point of a planar optical switch composed of multiple quantum well layers, and downspin (or upspin) is caused by the first circularly polarized light signal. A first step of exciting carriers to excite carriers of spin opposite to the first circularly polarized light signal by the second circularly polarized light signal, and a difference between the upspin and downspin carriers. Polarized shape of the optical probe A second step of identifying whether or not the signal pattern of the first circularly polarized light signal and the signal pattern of the second circularly polarized light signal are the same by measuring the presence or absence of a change in It is characterized by having.
In order to achieve the above object, an optical signal pattern identification device according to a fourth aspect of the present invention is characterized in that a first circularly polarized optical signal composed of an optical pulse of n bits (n is an integer of 2 or more) and the first circularly polarized light. A polarization beam splitter having circular polarization opposite to that of the optical signal and passing through the second circularly polarized optical signal, which is also composed of an n-bit optical pulse, and probe light, which is also composed of an n-bit optical pulse, A λ / 4 plate for converting the first circularly-polarized light signal and the second circularly-polarized light signal that have passed through the polarizing beam splitter into reverse circularly polarized light, respectively, and the probe that has passed through the polarizing beam splitter Receiving light, a lens for condensing the first circularly polarized light signal and the second circularly polarized light signal that have passed through the λ / 4 plate, and light collected by the lens; Reflection type consisting of multiple quantum well layers for The first circularly polarized light signal, the second circularly polarized light signal, and the probe light are respectively i-th (i is an integer from 1 to n). Are applied to one point of the surface optical switch in synchronization with each other, downspin (or upspin) carriers excited by the first circularly polarized light signal, and the second circularly polarized light signal. Signal of the first circularly polarized light signal by measuring the presence or absence of a change in the polarization state of the optical probe caused by the difference from the spin carrier opposite to the first circularly polarized light signal excited by It is used for identifying whether the pattern and the signal pattern of the second circularly polarized light signal are the same.
In order to achieve the above object, the invention of an optical signal pattern identification device according to claim 6 is characterized in that a first circularly polarized optical signal composed of an optical pulse of n bits (n is an integer of 2 or more) and the first circularly polarized light. A polarization beam splitter having circular polarization opposite to that of the optical signal and passing through the second circularly polarized optical signal, which is also composed of an n-bit optical pulse, and probe light, which is also composed of an n-bit optical pulse, A λ / 4 plate and a −λ / 4 plate for converting the first circularly polarized light signal and the second circularly polarized light signal that have passed through the polarizing beam splitter into reverse circularly polarized light, respectively, and the polarization The probe light that has passed through the beam splitter, the first circularly polarized light signal that has passed through the λ / 4 plate, and the second circularly polarized light signal that has passed through the −λ / 4 plate are taken as one point. A lens for condensing, and the light is collected by the lens. A reflection-type surface optical switch composed of a multiple quantum well layer for receiving light, and the first circularly polarized light signal, the second circularly polarized light signal, and the probe light are respectively connected to the i th Down-spin (or up-spin) excited by the first circularly polarized light signal by irradiating one point of the planar optical switch with synchronized bits of the i th (i is an integer from 1 to n) ) And the change in the polarization state of the optical probe caused by the difference between the first circularly polarized light signal excited by the second circularly polarized light signal and the opposite spin carrier. Thus, the signal pattern of the first circularly polarized light signal and the signal pattern of the second circularly polarized light signal are used for identifying whether or not they are the same.

When carriers are excited in the multiple quantum well layer, exciton absorption saturation occurs due to spin-dependent phase space filling and spin-independent Coulomb interaction. Therefore, if there is a difference in the number of up-spin and down-spin of carriers excited inside, the probe light of the linearly polarized light (linearly polarized light is a combination of clockwise and counterclockwise circularly polarized light) clockwise and counterclockwise circularly polarized light. Each of the components is subjected to different saturable absorption effects due to spin-dependent phase space filling, so that the linearly polarized probe light changes to elliptically polarized light and is reflected from the optical switch. Therefore, if a polarizer orthogonal to the input probe light polarization is placed on the output side, probe light is not output when the number of upspin and downspin carriers is the same, but if there is a difference in the number of probes, Is output through the polarizer.
Therefore, a circularly polarized first circularly polarized light signal ( optical label signal ) composed of n bits (n is an integer of 2 or more ), and a second circularly polarized light signal ( local address signal ) having a reverse circular polarization. Similarly, an n-bit optical probe is irradiated to one point of a planar optical switch composed of multiple quantum well layers by synchronizing each i-th bit (i is an integer from 1 to n). When the up-spin and down-spin are respectively excited, if the signal patterns of the first and second circularly polarized light signals match, the internal spin balance is always maintained, so that the output of the optical probe can be obtained. Never but 1 when the bits of the number of mismatches were present at bits in the signal pattern of both, the optical probe output is obtained, it is determined the pattern of both do not match. That is, by measuring the presence or absence of a change in the polarization state of the optical probe caused by the difference between the up-spin and down-spin carriers, the signal pattern of the first circularly polarized light signal and the signal pattern of the second circularly polarized light signal are Can be identified.

  In the present invention, the polarization state of the probe caused by the difference in the number of up-spin (or down-spin) carriers in which the optical label is excited in the multiple quantum well layer and the down-spin (or up-spin) carrier in which the local optical signal is excited. By detecting the change, it is possible to determine whether the optical label and the local optical signal match or not, and it is possible to identify an extremely high-speed optical signal without being affected by the carrier lifetime or the spin relaxation time.

  Further, according to the present invention, since an EX-OR type operation in which probe light is detected is realized only when the pattern of the optical label and the local optical signal does not match, “1” that can be used in the optical label. The number of "" is not limited, and an optical label having an arbitrary pattern can be applied.

  In addition, the present invention can realize an optical signal pattern identification device that is extremely small and highly efficient as compared with a method using an optical fiber loop (see Non-Patent Document 1).

  In the present invention, the signal light 1 excites the carrier of one spin, the signal light 2 excites the carrier of the other spin, and the probe light monitors the spin balance. Match / mismatch can be recognized. Since the change in the probe light transmittance reacts only to the spin balance regardless of the total number of internal carriers, according to the present invention, it is extremely fast without being affected by the carrier lifetime or the spin relaxation time. It is possible to realize pattern recognition for optical packets. In the present invention, since the probe light is output only when the spin balance is lost, the so-called EX-OR type operation is performed. Therefore, the restriction on the number of bits of “1” in the optical label is eliminated. Furthermore, according to the present invention, it is possible to realize an extremely small and highly efficient optical signal pattern identification device as compared with a device using an optical fiber.

The best mode for carrying out the invention will be described below in detail with reference to the drawings.
(First embodiment)
FIG. 1 is a diagram showing a basic configuration of an optical signal pattern identification method according to the present invention, in which 11 is a condenser lens, 12 is an optical switch, 13 is a condenser lens, and 14 is a polarizer. FIG. 2 is a diagram for explaining the principle.

  Electrons and holes are present in a degenerated state of upspin (indicated by an upward arrow in FIG. 2) and downspin (indicated by a downward arrow in FIG. 2) for one energy state. Now, the wavelength of the signal light is adjusted to the exciton transition of electron-heavy hole, and the optical switch 12 composed of a multiple quantum well layer is irradiated. At this time, when circularly polarized signal light is incident, only one spin is excited, and when opposite circularly polarized signal light is incident, the other spin is excited.

  When the signal light is “0”, the polarization state of the linearly polarized probe light cannot be changed. Therefore, when the polarizer 14 orthogonal to the input probe light polarization is arranged on the output side, the output light is almost “0”. Become.

  On the other hand, when only one of the spins is excited with the circularly polarized signal light (optical signal) 1 or 2, the probe light of linearly polarized light (linearly polarized light is a combination of clockwise and counterclockwise circularly polarized light) Only the circularly polarized component that interacts with the spin feels changes in absorption and refractive index. That is, in this case, when linearly polarized probe light is irradiated, only the same circularly polarized component as the signal light is modulated by the signal light, and the probe light transmitted through the optical switch 12 becomes elliptically polarized light. 14 can be passed.

  On the other hand, when an optical switch is irradiated with circularly polarized signal light (optical signal) 1 and signal light (optical signal) 2 having the opposite circularly polarized light at the same time, upspin and downspin Are excited simultaneously, so that the saturable absorption effect due to phase space filling acts equally on both the clockwise and counterclockwise circular polarization components of linearly polarized probe light, so that the probe light changes its polarization state. Without passing through the optical switch 12, it is cut by the polarizer 14 and is not output.

  That is, as shown in FIG. 2, since the first three bits of the optical signal 1 and the optical signal 2 are equal, the number of up-spin and down-spin excited inside is equal and the transmittance of the probe light polarizer 14 is zero. However, at the moment when the mismatch between the two patterns occurs in the fourth bit, a difference occurs in the number of up spins and down spins, and the probe light can pass through the polarizer 14. At this time, the generated excess upspin carriers shift to downspin with a spin relaxation time of several tens of ps, so that the probe transmittance of the polarizer 14 is long even if the subsequent optical signal pattern is the same. The tail will be pulled. The downspin is excited by the seventh bit which is a mismatch bit opposite to the fourth bit, and when the number of spins is approximately the same, the transmittance of the probe becomes zero again.

  Thereafter, the mismatch bit appears again, and the probe light is output every time the spin balance is lost. That is, when the output of the probe light is obtained, the optical signal patterns of the two do not match, and when the probe outputs are all zero, it can be identified that they match. At this time, the total number of internal carriers has a nanosecond recombination lifetime, so it gradually accumulates, but the transmittance of the probe is not the total number of carriers, but the number of upspin and downspin carriers. Since it changes only in response to the difference between the two, it is possible to perform an ultra-high speed operation regardless of the carrier lifetime and the spin relaxation time. Here, an optical packet having the same number of bits as the signal light is used as the probe light, but a rectangular wave or CW (continuous-wave) light that lasts for the same time can also be used.

  Accordingly, in the configuration of FIG. 1, an optical label (also referred to as a header) given to an optical packet is input as an optical signal 1 and a local address signal having node address information is input as a signal light 2 to the node. It is possible to determine whether the address of the input optical packet is the same as the address of the node.

(Second Embodiment)
FIG. 3 shows an example of the configuration of the optical signal pattern identification device of the present invention. Here, 31 is a reflection mirror, 32 is a polarization beam splitter (PBS), 33 is a λ / 4 plate, 34 is a condenser lens, and 35 is an optical switch.

  Two signal light (optical signal) 1 and signal light (optical signal) 2 having orthogonal linearly polarized light are combined by a polarization beam splitter (PBS) 32 and converted into mutually circularly polarized signal light by a λ / 4 plate 33. Then, the light is irradiated onto the optical switch 35 at the same time. At this time, as described above, when the patterns and energies of both the signal lights 1 and 2 are equal, the internally excited spins are always the same, and the probe light is always reflected as linearly polarized light. Cut and not output.

  On the other hand, if even one bit does not match in the pattern of signal light 1 and signal light 2, the spin balance collapses at that moment, and probe light output is obtained, so both patterns do not match. Can be identified.

(Third embodiment)
FIG. 4 shows another configuration example of the optical signal pattern identification apparatus of the present invention. Here, 41 is a polarization beam splitter (PBS), 42 is a λ / 4 plate, 43 is a -λ / 4 plate, 44 is a condenser lens, and 45 is an optical switch.

  In the configuration example described in FIG. 3, when the signal light 2 is input to the PBS 32, it is necessary to input the signal light 2 from a different surface from the signal light 1. On the other hand, in the configuration of FIG. 4, the signal light 1 and the signal light 2 are incident from the same surface of the PBS 41, and the circularly polarized light beams that are opposite to each other using the λ / 4 plate 42 and the −λ / 4 plate 43 respectively. Has been converted. The linearly polarized probe light is output with the pattern of the signal light 1 and the signal light 2 being inconsistent and not in accordance with the same principle as described above.

  As described above, the present invention can realize pattern recognition for an extremely high-speed optical packet without being affected by the lifetime of the carrier and the relaxation time of the spin, and “1” in the optical label. The limitation on the number of bits is eliminated, and it is possible to realize an extremely small and highly efficient optical signal pattern identification device as compared with the conventional device using an optical fiber. Therefore, it can be expected that the optical signal pattern identification method and identification apparatus of the present invention can contribute to the development of a particularly promising optical packet switch network in the future.

It is a schematic diagram which shows the basic composition of the optical signal pattern identification method of this invention. It is a conceptual diagram explaining the principle of operation of the optical signal pattern identification method of FIG. It is a schematic diagram which shows an example of a structure of the optical signal pattern identification apparatus of this invention. It is a schematic diagram which shows the other structural example of the optical signal pattern identification apparatus of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Condensing lens 12 Optical switch 13 Condensing lens 14 Polarizer 31 Reflection mirror 32 Polarization beam splitter (PBS)
33 λ / 4 plate 34 Condensing lens 35 Optical switch 41 Polarizing beam splitter (PBS)
42 λ / 4 plate 43 -λ / 4 plate 44 Condensing lens 45 Optical switch

Claims (7)

The first circularly-polarized optical signal consisting of an n-bit (n is an integer of 2 or more) optical pulse and the circularly polarized light opposite to the first circularly-polarized optical signal, and the same as the n-bit optical pulse. Two circularly polarized optical signals and an optical probe consisting of an n-bit optical pulse are synchronized with each i-th bit (i is an integer from 1 to n), and from the multiple quantum well layer. The surface-type optical switch is irradiated with one point, downspin (or upspin) carriers are excited by the first circularly polarized light signal, and first circularly polarized light is emitted by the second circularly polarized light signal. A first step for exciting carriers of spin opposite to the signal;
By measuring the presence or absence of a change in the polarization state of the optical probe caused by the difference between the upspin and downspin carriers, the signal pattern of the first circularly polarized light signal and the second circularly polarized light signal And a second step of identifying whether or not the signal patterns are the same.   2. The first circularly polarized optical signal is an optical label attached to an optical packet, and the second circularly polarized optical signal is a local address signal having node address information. Optical signal pattern identification method.   3. The optical signal according to claim 1, wherein a condensing lens is used for irradiating the one point in the first step, and a polarizer is used for the measurement in the second step. Pattern identification method. The first circularly-polarized optical signal consisting of an n-bit (n is an integer of 2 or more) optical pulse and the circularly polarized light opposite to the first circularly-polarized optical signal, and the same as the n-bit optical pulse. A polarization beam splitter through which two circularly polarized optical signals and probe light consisting of n-bit optical pulses pass,
A λ / 4 plate for converting the first circularly-polarized light signal and the second circularly-polarized light signal that have passed through the polarization beam splitter, into circularly-polarized light in reverse directions, respectively;
A lens for condensing the probe light that has passed through the polarization beam splitter, the first circularly polarized light signal and the second circularly polarized light signal that have passed through the λ / 4 plate, at one point;
A reflective surface optical switch composed of multiple quantum well layers for receiving the light collected by the lens;
The first circularly polarized light signal, the second circularly polarized light signal, and the probe light are synchronized with the i-th bit (i is an integer from 1 to n) and the surface light is synchronized. Irradiating one point of the switch, the down-spin (or up-spin) carrier excited by the first circularly polarized light signal and the first circularly polarized light excited by the second circularly polarized light signal The signal pattern of the first circularly polarized optical signal and the second circularly polarized optical signal are measured by measuring the presence or absence of a change in the polarization state of the optical probe caused by the difference between the spin carrier opposite to the optical signal. An optical signal pattern identification device used for identifying whether or not the signal pattern is the same.   5. The optical signal pattern identification device according to claim 4, further comprising a reflection mirror for matching the first optical signal and the second optical signal before passing through the λ / 4 plate. The first circularly-polarized optical signal consisting of an n-bit (n is an integer of 2 or more) optical pulse and the circularly polarized light opposite to the first circularly-polarized optical signal, and the same as the n-bit optical pulse. A polarization beam splitter through which two circularly polarized optical signals and probe light consisting of n-bit optical pulses pass,
A λ / 4 plate and a −λ / 4 plate for converting the first circularly polarized light signal and the second circularly polarized light signal that have passed through the polarizing beam splitter into reverse circularly polarized light, respectively;
The probe light that has passed through the polarization beam splitter, the first circularly polarized light signal that has passed through the λ / 4 plate, and the second circularly polarized light signal that has passed through the −λ / 4 plate are 1 A lens for focusing on the point;
A reflective surface optical switch composed of multiple quantum well layers for receiving the light collected by the lens;
The first circularly polarized light signal, the second circularly polarized light signal, and the probe light are synchronized with the i-th bit (i is an integer from 1 to n) and the surface light is synchronized. Irradiating one point of the switch, the down-spin (or up-spin) carrier excited by the first circularly polarized light signal and the first circularly polarized light excited by the second circularly polarized light signal The signal pattern of the first circularly polarized optical signal and the second circularly polarized optical signal are measured by measuring the presence or absence of a change in the polarization state of the optical probe caused by the difference between the spin carrier opposite to the optical signal. An optical signal pattern identification device used for identifying whether or not the signal pattern is the same.   7. The first circularly polarized optical signal is an optical label attached to an optical packet, and the second circularly polarized optical signal is a local address signal having node address information. The optical signal pattern identification device according to any one of the above.

Images