RU2540800C2 - Method of switching optical packets and device for therefor - Google Patents

Abstract

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to means of designing digital systems. The method includes transmitting a label in the address part of an optical unit, using a clock channel with an allocated wavelength λ_N+1 and transmitting clock pulses which are common for all optical transmission channels and which form frames. The units consist of an address and a data (data packet) field. There is a label in the address field, the label representing a feature of a switch to which the message is addressed. There are protective intervals t₁ and t₂ before and after the label. A protective interval t₃ can be located at the end of a frame. A separate bit sequence corresponds to each switch, and in the absence of a data unit, an "Empty unit label" is written in the address to form a so-called "empty unit".

EFFECT: high rate of processing information with fewer electronic-optical conversions in the system and resultant distortions.

2 cl, 2 dwg

Description

1. The technical field to which the invention relates.

The invention relates to techniques for optical switching and can be used to create fiber-optic systems for transmitting information and local area networks with packet switching.

2. The level of technology

Currently, the most widely used method of constructing a digital primary network in which a signal is transmitted in optical form is Synchronous Digital Hierarchy (SDH) technology. A further development of optical information transmission technology was the Wavelength Division Multiplexing (WDM) technology, also called wavelength division multiplexing or spectral separation.

In the early 2000s, the principle of label-based IP packet switching was developed in the MPLS protocol (MultiProtocol Label Switching), which allows the formation of virtual packet transmission paths in the network of label switching routers (Label Switching Router, LSR). The same idea, transferred to the physical layer of the optical network, took the form of a protocol called MultiProtocol lambda Switching (MP1S). (L. Barash Multiprotocol lambda switching. Computer Review, No. 4, January 31 - February 6, 2001). Switching here is no longer based on the labels contained in the packet headers, but in accordance with the wavelengths at which traffic of one type or another is transmitted. This technology for processing optical signals is called lambda switching (the terms “photon switching” and “wavelength switching” are also used).

Despite the fact that the creation of optical packet switches is discussed in the scientific and technical literature, no analogues of the proposed switching method have been identified in the patent information. The proposed method relates to switching methods with a fully optical label exchange (AOLS - All-Optical Label Swapping), a particular case of which is the multiprotocol lambda switching method described in an article by V. Makkaveev. Photon switches. Components and Technologies Magazine, 2006, No. 2.

3. Disclosure of invention

The purpose of the proposed method for switching optical packets is to increase the speed of information processing, reducing the number of electron-optical transformations in the system and the distortions introduced by them.

In FIG. 1 shows the structure of a signal transmitted in an optical fiber of one transmission direction.

The signal is transmitted in N spectral channels that differ in wavelength similar to the WDM signal. To transmit information signals, wavelengths λ ₁ -λ _{N are used} .

A distinctive feature of this transmission method is:

a) label transmission in the address part of the optical unit, in contrast to multi-protocol lambda switching (MultiProtocol Label Switching), where the label is the wavelength at which the data block is transmitted;

b) the presence of a synchronization channel using the selected wavelength λ _{N + 1} . The synchronization channel transmits clock pulses common to all optical transmission channels and forming frames.

Information is transmitted frame by frame (in blocks), the time intervals of which are set by pulses transmitted in the synchronization channel.

Blocks consist of an address and a data field (data packet).

In the address field is a label, which is a sign of the switch to which the message is addressed. Before and after the label are protective intervals t _Z. At the end of the frame is also the guard interval t _Z.

Each switch has an individual bit sequence. In addition, if there is no data block in the address, the sequence of bits “Label of the empty block” is recorded, forming the so-called “empty block”. The specific label coding method is selected during system design.

Packet switching is carried out as follows.

Upon receipt of a synchronization pulse, the switch is transferred to the initial processing state of the frame. The guard interval t _З is determined by the switching time of the switch.

Then the switch processes the incoming label and switches the data block. The data packet is not converted into electronic form.

To compensate for the processing time and switching, a delay line must be introduced into the switch circuit.

Brief Description of the Drawings

Figure 1. The structure of the optical signal

Where:

3 - label;

4 - channel data transmitted on an optical carrier λ ₁ ;

5 - channel data transmitted on an optical carrier λ _N ;

6 - channel synchronization on an optical carrier λ _{N + 1} ;

7 - protective interval t ₃ ;

8 - data packet;

9 - data field

10 - address field;

11 - data frame duration.

FIG. 2. Switch structure

The diagram indicates:

12, 13 - optical divider of transmission directions 1-2 and 2-1, respectively;

14, 15 - optical splitter channel synchronization of transmission directions 1-2 and 2-1, respectively;

16, 17 - optical channel splitter data transmission directions 1-2 and 2-1, respectively;

18, 22 - optoelectronic converter channel data transmission directions 1-2 and 2-1, respectively;

19, 23 - optoelectronic converter channel synchronization of transmission directions 1-2 and 2-1, respectively;

20, 24 - delay line of the data channel of the transmission directions 1-2 and 2-1, respectively;

21, 25 - delay line channel synchronization of transmission directions 1-2 and 2-1, respectively;

26, 27 - switching element 2 × 2 channel data transmission directions 1-2 and 2-1, respectively;

28 - electron-optical router (OEO router);

29, 30 - control device of transmission directions 1-2 and 2-1, respectively;

31, 32 - switching element 2 × 2 data channels of transmission directions 1-2 and 2-1, respectively;

33, 34 - adder data channel and synchronization of transmission directions 1-2 and 2-1, respectively;

35 - decisive device backup system;

36 - switching element 2 × 2 redundancy system;

37, 38 - optical amplifier of transmission directions 1-2 and 2-1, respectively.

The structure of a photon switch (FC) that implements the indicated method for switching optical packets is shown in FIG. 2.

FC is designed to build fully optical networks of linear and ring topologies similar to SDH networks.

To facilitate perception, the diagram for each direction of transmission (from 1 to 2 and from 2 to 1) shows only one optical transmission channel for wavelength λ ₁ . In the presented scheme, only one information signal transmitted at a wavelength is considered.

The device of optical transmission channels for wavelengths λ ₁ -λ _N (using the technology of wavelength division multiplexing (WDM)) is similar to that shown and the switch circuit is proportionally increased.

The solid lines show the passage of the information signal and the synchronization signal, the dashed lines show the control signals of the switch.

The photon switch includes two transmission directions (1-2 and 2-1) and a backup device, as well as the terminal SDH multiplexer, and differs from it in that each direction is divided into an individual optical part and an electron-optical router 28 common to both directions.

The optical channel of each transmission direction (for example, 1-2) consists of a series of interconnected optical elements.

The received signal is fed to the optical splitter (12, 13) connected to the optical splitters (14, 16 and 15, 17).

One of the outputs of each optical splitter through an optoelectronic converter (18, 19, 22, 23) is connected to a control device (29, 30), the control signals from which are fed to the switching elements of the transmission directions (26, 31; 27, 32), respectively, and also a backup system resolver 35 and an electron-optical router 28 from each control device.

The second terminals of each optical splitter of the data channel through optical delay lines (20, 24) are connected to the inputs of the switching elements of the transmission directions (26, 27), the second input of which is connected to the electron-optical router 28.

The second outputs of each optical splitter of the synchronization channel through optical delay lines (21, 25) are connected to the inputs of the switching elements of the transmission directions (31, 32), the second inputs of which are connected to the control device (29, 30).

The first outputs of the switching elements of the transmission directions 26, 31 and 27, 32 are connected to the adders 33 and 34, respectively.

The second outputs of the switching elements of the transmission directions 26, 31 are connected to the electron-optical router 28.

The outputs of the adders 33, 34 are connected to the inputs of the switching element 36, the control action of which comes from the resolver of the backup system 35.

The outputs of the switching element of the backup system 36 are connected to optical amplifiers 37, 38, which form a transmission signal.

Switch operation description

Photon switch (FC) operates as follows.

In normal mode, the FC operates in the direction of transmission 1-2 and 2-1. An optical signal containing an information signal (on the wavelength λ ₁ ), a dedicated optical transmission channel for clock pulses (on the wavelength λ _{N + 1} ) is supplied to inputs 1 and 2.

Consider the passage of a signal in the direction of transmission 1-2.

Data and clock channels are separated in a divider (12), then part of the signal branches off in splitters (14, 16) and, after conversion to electrical form (in photodetectors 18, 19), enters a control device (UE) 29. The synchronization signal sets the switching elements to the initial position and prepares the UU for processing the labels of the next data block. The main part of both the information and the clock signal passes through the delay lines (20, 21), which are necessary to delay the signal for the duration of the label analysis and the generation of the control signal by the control unit. Then the information signal hits the optical switching element 26. If the UE has found a label that matches the address of the FC, an “empty block” or a block addressed to “everyone”, it switches the switching element and simultaneously outputs a signal to the OEO router 28, through which it issues to the second input of the switching element 26, the outgoing optical data unit, and in its absence, an “empty unit”. The switching element switches the incoming data block to the second output and then to the OEO router, and the block received from the router switches to the first output. Otherwise, the control signal is not generated, and the received optical unit is sent to the first output. The output signal is combined with the clock signal in the adder 33 and through the optical amplifier 37 is sent to the line.

In the absence of a signal at the input of UU 29, an alarm signal is supplied to the deciding device 35, through which protective switching is performed through the switching element 36. Simultaneously, through the switching element 31, the clock pulses generated by the UU 29 to control the recovery of the line are fed into the line.

The passage of the signal in the direction 2-1 is similar to that considered.

4. The implementation of the invention

The element base for creating such a switching circuit already exists. There are 2 × 2 basic optical modules on the market. An example is the solid state MEMS switch (Sercalo Miniature fiber optic MEMS switch).

In addition, in November 2004, Matsushita Electric Works received a patent for photon crystal-based baffles and light flux switches (United States Patent 6,822,784, Fukshima, et al., November 23, 2004, Light-beam deflecting device with photonic crystal, optical switch using the same, and light-beam deflecting method).

The proposed method will allow you to create fiber-optic networks with fully optical packet switching based on the existing structure of SDH networks by replacing SDH multiplexers with FC, which will allow:

increase the throughput of the data transmission system;

to reduce the overall dimensions and power consumption of switching equipment;

reduce the number of optoelectronic and electron-optical transformations in the system.

Claims (2)

1. The method of switching optical packets, which consists in the fact that information is transmitted frame by frame (in blocks), consisting of an address and a data field (data packet), the time intervals of which are set by pulses transmitted in the synchronization channel, moreover, to transmit information signals of data blocks, wavelengths λ _l -λ _N with label transmission in the address part of the data block; in the synchronization channel, the selected wavelength is used for clock pulses common to all optical transmission channels and forming frames; in the address field there is a label, which is a sign of the photon switch to which the message is addressed; when receiving synchronization pulses, the switch enters the initial processing state of the frame, and when a tag is found that matches the address of the photon switch, it switches the optical element to issue an optical data unit, characterized in that the label is transmitted in the address part of the optical unit and the presence of a synchronization channel using selected wavelength. 2. A photon switch that implements the method according to claim 1, comprising two transmission directions, the optical channel of each of which consists of a divider and a splitter for extracting a synchronization signal with a selected wavelength to set the switching elements in the initial state, a control device for detecting labels in address part of the data block that matches the address for switching the corresponding switching element, and issuing a signal to the electronic-optical router for both directions of transmission, I provide he received data block his exit from the photonic switch, characterized in that each direction is divided into individual optical portion and is common to both directions of the electron-optical router.

Images