JPH01143540A - Optical transmission data addressing system - Google Patents

Abstract

PURPOSE:To process separately a header and an address at an ultrahigh speed easily by means of one physical transmission line only and to decrease the data length to the same as an address length of parallel transmission by applying wavelength split optical transmission to a data and an address. CONSTITUTION:A data subject to wavelength multiplex where the data is processed by a wavelength lambdaa and address information is processed by two wavelengths lambdaa1 and lambdaa2 is sent and they are separated respectively by wavelength filters 101, 102, 103 and outputted to a distributer 12. A data A given from a data line 130 is outputted to a path 140 and other data are outputted to each output path as data 152-158 in the distributer 12. The data and address are subject to wavelength split optical transmission so as to attain ultrahigh speed data addressing while the data and address are separated easily. That is, the data and address information is sent in parallel on an optical transmission line in separate wavelengths and the separating processing of header and address is processed at ultrahigh speed by one physical transmission line and the data length is reduced to the same length as the address length.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an extremely high-speed and highly efficient addressing system for transmission data.

(Prior Art) Conventionally, addressing of data that can be transmitted has been performed by adding address information to each data as a header, as seen in packet communication systems.

FIG. 3 shows an example of a conventional addressing method in which address information is added to a header and transmitted.

Format 30 is a data format on transmission line 31. Each data has a ladder 301 in the 'D' data section 302 and 'H' containing address information. Here, 4
Data 32, 33, 34, 35 can be transmitted continuously, and each data has its destination address "00",
Route 3801 indicated by “10”, “01”, and “11”,
Addressed to 3803, 3802, and 3804.

The data and header are passed through the separator 36 onto 3601 and 3601 respectively.
The distributor 37 can be controlled by the contents of the 0-separated header separated onto 602, and each data can be sent to the output path indicated by the header. A method in which addressing control of each data is driven by the header in this way is called a header-driven method.

This header-driven method was mainly used for allocating data between exchanges and processors because it could autonomously control data addresses.

However, since this method transmits the data and address information in the same way, the amount of transmission increases by the amount of the header.
In other words, it has the disadvantage of increasing overhead.

Therefore, it is necessary to increase the data length and reduce the header ratio.

Furthermore, when transmitting at high speeds, the process of separating the header and data before analyzing the header and addressing the data becomes complicated, making header-driven control itself difficult to implement in ultra-high-speed transmission. .

To address this problem, addressing systems using optical wavelength multiplexing have been proposed in recent years. The method will be explained with reference to FIG.

40 indicates data on the optical transmission path 41, indicating that the data can be optically transmitted at wavelength λi.

Data 42 to 45 each have a wavelength λ1. λ8. λ2. The 6-segment splitter 46 transmitting light at λ4 connects each route 47o1 to 47.
Tweet on o4 each λ1. λ2. λ8. It has a wavelength filter of λ4 and addresses data of each wavelength.

That is, in this system, instead of adding address information to the header, the data itself is optically converted at wavelengths λ1 to λ4 according to the address information and transmitted. Therefore, instantaneous addressing is possible by filtering the wavelength when data is received, making it possible to realize ultra-high-speed addressing of optically transmitted data. However, in this method, if there are n addressing destinations, it is necessary to prepare n wavelengths, and there is a limit to the wavelengths that can be used.
A problem arises when the number of destinations n becomes large.

(Problems to be Solved by the Invention) As mentioned in the above section of the prior art, in the conventional header-driven addressing control, overhead occurs for the header, so it is necessary to increase the data length. However, since the header-driven method requires complicated processing to separate the header and data, it is difficult to use it in ultra-high-speed optical transmission systems.

In addition, in the addressing method using wavelength multiplexing,
Even if the data itself is wavelength-converted according to address information without adding a header, as many wavelengths as the number of destination addresses are required, making large-scale addressing control with many output routes difficult.

(Means for Solving the Problem) According to the first invention of the present application, it is a method for addressing optical transmission data to a plurality of destinations, and the data and destination address information represented by n bits of the data are , the data is assigned a wavelength λd, and the n-bit destination address information is assigned wavelengths λa1 to λa2 to each bit, the data and the address information are optically wavelength multiplexed and transmitted, and the data is transmitted at the wavelength λd. An addressing method is obtained in which the transmitted data is addressed to the destination indicated by the address information transmitted at the n wavelengths λa1 to λa2.

Further, according to the second invention of the present application, there is a method for addressing optical transmission data to a plurality of destinations, in which different wavelengths λd and λa are respectively assigned to data and destination address information represented by a plurality of bits of the data. The allocation and chanting data and the address information are transmitted by optical wavelength multiplexing, and the address information represented by a plurality of bits is transmitted by time division multiplexing at wavelength λ3, and the data transmitted at the wavelength λd is transmitted. , the addressing method is changed, characterized in that addressing is performed to a destination indicated by the address information represented by a plurality of bits transmitted by time division multiplexing at the wavelength λa.

(Operation) In the addressing systems according to the first and second inventions of the present application, data and address information are transmitted in parallel on an optical transmission path using different wavelengths. Therefore, separation of data and addresses can be achieved extremely easily simply by using a wavelength filter. Parallel transmission in an electrical system requires two physical transmission lines, but since the present invention uses optical wavelengths, parallel transmission in the present invention can be easily achieved with only one physical transmission line. .

Therefore, with the present invention, it is extremely easy to achieve ultra-high speed separation processing of headers and addresses, which was difficult in the conventional header-driven system. Furthermore, in the present invention, since addresses and data are transmitted in parallel, the data length can be shortened to the same length as the address length transmitted in parallel.

In the first invention, addresses can be multiplexed and transmitted in parallel using n wavelengths. Therefore, when 1 bit of data is received, n bits of address are instantly received.
This enables high-speed addressing to 21 routes. This means that the method of the present invention can address more routes than the conventional wavelength multiplex addressing method even if the number of wavelengths is the same. For example, if N wavelengths are used, addressing can only be done in N directions with the prior art, but according to the present invention, addressing can be done in 2N-1 directions.

On the other hand, according to the second invention, address information is time-division multiplexed and transmitted using a wavelength λ dedicated to addresses. Now, assuming that n-bit addresses are time-division multiplexed and transmitted serially, addressing to 21 routes becomes possible at the time when n bits are received. In other words, in the method of this second invention, one data can be addressed to 21 directions using only two wavelengths, which makes it possible to address more output routes than the conventional wavelength division multiplexing addressing method. becomes. In the second invention, it is sufficient if the data length is at least n bits, but if the data length is nXm bits, the n-bit address becomes m
Since it is possible to send a set, other wavelengths (λd, ~λd,
) can simultaneously perform addressing for multiple pieces of data sent.゛(Example) Next, the present invention will be explained in more detail with reference to Examples.

FIG. 1 is a conceptual diagram showing an embodiment of the first invention of the present application. Data on the optical transmission line 100 has a wavelength λd and address information λa, λ8 . It is wavelength-multiplexed and transmitted using two wavelengths. At this time, the data and address information can be separated by wavelength filters 101, 102, and 103, respectively. 1
11-118, 1111-1181, 1112-1
Up to 182 are data, address Alt address A, respectively.
Here, 2 bits of address are assigned to 1 bit of data.1For example, the address of data 111 is 2 bits 1111 and 1112.
00'', the distributor 12 is driven, and the data !1
The data “A I+” input from 130 is output to route 140. Similarly, the data from 112 to 118 are input to 15
2 to 158 are output to each output route.

This distributor 12 is the address! There are various possible mechanisms for driving the signals 131 and 132, such as decoding the 2-bit signals 131 and 132 to generate a 4-bit control signal and driving the gates (wavelength filters, etc.) of each route. . Further, although the data length is 1 bit here, it may be expanded to n bits or may be of variable length.

FIG. 2 is a conceptual diagram showing an embodiment of the second invention of the present application. Data and address information are transmitted on the optical transmission path 200 after being wavelength-multiplexed using wavelengths λd and λa, respectively. This is separated into input d and λ3 by wavelength filters 201 and 202.

At this time, the 2 bits of the address for each data have the wavelength λa
Transmission can be done in time division. Therefore, in this case, the data will be 2 bits or more. If the data and address can be transmitted in the same phase as shown in Figure 2, the data can be delayed until 2 address bits are received, or the distributor can be configured like a multi-stage distributor and the 1-bit address can be transmitted in the same phase. is distributed to each output route 241 to 2 in multiple stages each time it is received.
44. Alternatively, the address and data may be transmitted out of phase so that data reception begins when two address bits have been received.

Next, an embodiment in which the second invention is applied to transmit a plurality of data using a plurality of wavelengths and address each data using only λa will be described with reference to FIG.

Two types of data are wavelength-multiplexed and transmitted using λd and input d on the optical transmission line 500°. These are 501゜50
It is received by the second wavelength filter. Here, the data length is 4 bits. The address information for each data is 2 bits each, and is time-divided by the wavelength λa, so that two types of data can be transmitted alternately. Here address 521 and data 511, 522 and 512, 5
23 and 513 correspond to each other, and 524 and 514 correspond to each other. In FIG. 5, the phases of the address and data are shifted, so that data reception starts when the address has finished being received.

With this configuration, the 2 bits of address information of each data are input one after another to the distributor 53. Although the 0 distributor 53 is powered by 2 people and has 4 outputs, it is actually a parallel arrangement of distributors that each have 4 outputs powered by 1 person. are equivalent. This is because the wavelength λ
This is because the splitter 53 can be configured by inputting the light as is and preparing filters for λd and λd in parallel for each output route since d and λd are different.

Since it is possible to output four data by two people using three waves in this way, more efficient control is possible compared to the conventional addressing method using wavelength multiplexing.

FIG. 6 is a block diagram showing a specific example of the distributor 12 of FIG. 1. Data transmitted through the signal line 1300 wavelength input d is transmitted through wavelength filters 641, 642, 643, 644.
You can input it to . On the other hand, the address information λ"Isλa is input to O/E converters (optoelectrical converters) 621 and 622 through signal lines 131 and 132, respectively, and converted into a 2-bit electrical signal. The decoder 623 converts it into a 4-bit electrical signal. It becomes a filter drive signal.Each filter drive signal is 631
, 632, 633, and 634 drive each filter to receive the data λd.

The respective distributors 23 and 53 in FIGS. 2 and 5 have similar configurations.

(Effects of the Invention) In both the first and second inventions of the present application, data and addresses are transmitted using wavelength-division optical transmission, so data and addresses can be easily separated even in ultra-high-speed optical transmission systems. High-speed data dressing is possible. however,
Since the present invention performs parallel transmission using optical wavelengths, it is not necessary to increase the number of physical transmission lines as in the case of parallel transmission using electricity.

When one data is addressed in many directions, the first invention of the present application allows addressing in 21 directions with n+1 wavelengths, and the second invention of the present application clears n bits of address information in a time-division manner. Since the configuration is such that only one wavelength is used for addressing, 21
Directional addressing becomes possible.

Furthermore, according to the first invention, since n-bit addresses can be received simultaneously, instantaneous addressing is possible and it is suitable for high-speed operation. On the other hand, according to the second invention, if m sets of address information are transmitted in the time-division direction, it is possible to address m data wavelengths with one wavelength.

As described above, both the first and second inventions of the present application are extremely effective in efficiently addressing data transmitted at extremely high speeds.

[Brief explanation of the drawing]

1 and 2 are conceptual diagrams illustrating the operation of an addressing circuit implementing the addressing method according to the present invention.
FIG. 2 shows an embodiment of the second invention of the present application in which address information is time-division multiplexed and transmitted using wavelength λa. FIG. 3 is a conceptual diagram illustrating a conventional addressing method using header-drib type, FIG. 4 is a conceptual diagram illustrating a conventional addressing method using wavelength multiplexing, and FIG. 5 is a conceptual diagram illustrating a conventional addressing method using wavelength multiplexing. A conceptual diagram showing another embodiment, FIG. 6 is a block diagram showing a specific example of the distributor 12 of FIG. 1.

Claims (2)

[Claims] (1) A method of addressing optical transmission data to multiple destinations, in which a wavelength λd is assigned to the data and destination address information represented by n bits of the data, and the destination address of the n bits is For information, wavelengths λa_1 to λa_2 are assigned to each bit,
The data and the address information are optically wavelength multiplexed and transmitted, and the data transmitted at the wavelength λd is addressed to the destination indicated by the address information transmitted at the n wavelengths λa_1 to λa_2. Addressing method. (2) A method of addressing optical transmission data to multiple destinations, in which different wavelengths λd and λa are assigned to data and destination address information represented by multiple bits of the data, respectively, and the data and address information are the address information represented by a plurality of bits is transmitted by time division multiplexing at a wavelength λa, and the data transmitted at the wavelength λd is transmitted at the wavelength λa.
An addressing method characterized in that addressing is performed to a destination indicated by the address information represented by a plurality of bits transmitted by time division multiplexing.