JPH084271B2 - Optical communication network and local area network - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical communication network constituted by connecting star couplers to each other and a multimedia-compatible one capable of transmitting data communication, voice signals and video signals constituted by the optical communication network. Regarding local area networks.

[0002]

2. Description of the Related Art Local Area Network (LAN)
Has recently become widespread as a data communication network between computers, workstations, etc. arranged in a relatively short distance, and one of them is called Ethernet (registered trademark) developed by Xerox Corporation. .

FIG. 9 is a diagram showing an Ethernet (registered trademark) network. In FIG. 9, 1 is a coaxial cable, 2 is a tap (branch point), and 3 is a node (terminal, station). Each node 3 is connected to the coaxial cable 1 by a tap 2. When the number of nodes 3 to be connected increases, a tap 2 is newly provided and a new node 3 is connected to it.

On the other hand, in recent years, the development of optical communication using an optical fiber as a transmission medium is remarkable. The excellent characteristics of optical fiber, such as wide band, low loss, and high noise resistance, are suitable for local area networks. However, it is difficult to form a network similar to Ethernet (registered trademark) because it is not possible to provide taps one after another with an optical fiber like a coaxial cable.

Therefore, a network has been proposed in which transmission and reception of nodes are divided into separate terminals and all the nodes are distributed by a star coupler. For example, E.
G. Rawson et. al. , IEEE Tran
action on Communication
s, Vol, COM-26, No. 7, Jul 19
78, p983-p990, "Fibrenet: M
ultimode Optical Fibers f
or Local Computer Network
s "etc.

FIG. 10 shows an example of the above proposal of an optical communication network using a star coupler. In FIG. 10, 3
Is a node, 4a and 4b are optical fibers, 5 is a mixing rod type star coupler, and 6 is a terminal.

The signal from each node 3 is converted into an optical signal by each light emitting element 7 and supplied to the star coupler 5 through each optical fiber 4a. These optical signals are all mixed once by the star coupler 5, then distributed to each light receiving element 8 via each optical fiber 4b and converted into electric signals again, and these electric signals are supplied to each node 3.

As a result, the signal transmitted from one node can be transmitted to all the nodes (multicast property), and a communication network similar to Ethernet (registered trademark) is constructed. be able to.

However, the optical communication network using such a star coupler has a drawback in that even if an attempt is made to add nodes to expand the network, only the number of terminals included in the star coupler can be added. This is because when the star couplers are connected together, a closed loop is formed, and phenomena such as oscillation and damped oscillation occur.

In order to solve such a problem, as described in Japanese Patent Application No. 2-98370 filed by the present applicant, a pair of star couplers should be connected to the same node. The transmission coefficient between the input terminal and the output terminal thus formed may be zero. By doing so, it becomes possible to connect the star couplers one after another in order to expand the network scale. In addition, bidirectional communication is possible between the two nodes connected to the network configured in this way, and collision detection can be easily performed by utilizing this bidirectionality.
Furthermore, Japanese Patent Application No. 2-4090 filed by the applicant
As described in the specification of No. 70, the number of optical fibers, optical connectors, and relay amplifiers is halved by allowing one terminal to carry out the input and output functions without dividing the input terminal and the output terminal. It can be reduced.

Generally, in a communication network, what kind of transmission control procedure (hereinafter referred to as "protocol") is used when transmitting a signal. The protocol adopted in the above-mentioned Ethernet (registered trademark) is a contention system.

In the network of FIG. 9, let us consider a case where signal transmission is performed by a contention-based protocol.
In order to transmit a signal from a certain node 3, it is first checked whether or not there is a transmission signal from another node on the bus, that is, the coaxial cable 1. If there is no signal on the bus, start transmission, and if there is any signal on bus, do not transmit and wait. Therefore, the bus will win the line as soon as possible, and the line will be secured. Therefore, such a protocol is called a contention-based protocol. The protocol actually adopted in Ethernet (registered trademark) is CSMA / CD (Carrier).
Sense Multiple Access / Co
This is called "liliction detection" and basically follows the above control procedure.

The contention-based protocol is suitable for data communication because the bus is used without waste of time. However, the local area network that complies with the protocol of the contention system is not suitable for transmitting signals that require real-time property such as voice signals and video signals. For these audio signals and video signals, once the transmission is started, a predetermined amount (a few kilobits) of signals must be transmitted at least before a predetermined short time (for example, about one hundredth of a second) has elapsed. And images are not smooth. However, under the contention-based protocol, even if transmission is started once, the bus may be occupied by another node on the way.
If the bus is relatively idle, real-time performance can be ensured even under the contention-based protocol, but this is a probabilistic problem and there was a problem from the stability of communication.

In order to solve such a problem, as described in Japanese Patent Application No. 2-69999 filed by the present applicant, a plurality of transmission channels are provided to perform data communication and real-time communication. There is a method of transmitting and via separate transmission channels.

In the network of a plurality of transmission channels, as shown in FIG. 11, a plurality of buses 9a and 9a are provided.
b is provided, and each node 3 is connected to all the buses 9a and 9b. Of these, the bus 9a is a transmission channel for data communication, and the other buses 9b are transmission channels for real-time communication. Also, a bus 9b for real-time communication
The time-division control device 10 is provided corresponding to the above. Further, the time division control device 10 receives the instruction for starting real-time communication from the transmission channel for data communication in order to receive the instruction from the bus 9
It is also connected to a.

When real-time communication is required, each node uses the bus 9a, which is a transmission channel for data communication, and the time-division controller 10 under the line contention protocol.
Request line allocation. The time-division control device 10 assigns lines by time-division multiple access (TDMA) of the token bus system in consideration of the use status of the bus 9b managed by itself.

[0017]

However, the above-mentioned method of transmitting the data communication and the real-time communication by the separate transmission channels has a problem that the cost required for constructing the network becomes large because a plurality of buses are provided. there were. For example, when the bus is constructed by coaxial cables, two coaxial cables are naturally required. This not only doubles the cost of the coaxial cable, but also doubles the space required for laying, and two taps and transceivers must be prepared.

Therefore, an object of the present invention is to construct an optical communication network and a local area network having a plurality of transmission channels with a simple structure.

In order to solve the above-mentioned problems, an optical communication network of the present invention has a plurality of terminals, an optical signal is input / output from the same terminal, there is no distribution of the optical signal to its own terminal, and , A terminal of a passive star coupler to which an optical signal is distributed is connected to terminals other than its own terminal, and a terminal of the passive star coupler is connected to a node through an optical fiber. A means for transmitting and receiving via a common optical fiber using different wavelengths for each channel is provided, and a plurality of transmission channels are configured by wavelength multiplexing between the passive star couplers. .

Further, in the local area network of the present invention, at least one of the plurality of transmission channels is a contention system transmission channel, and at least one of the plurality of transmission channels is a time division system transmission channel. An optical communication network, a plurality of nodes respectively connected to the optical communication network, and a time division control device provided in association with each of the time division transmission channels via the line contention transmission channel. And a time-division control device that performs transmission via the time-division transmission channel in response to all the requests.

[0021]

According to the optical communication network based on the present invention,
Since optical signals of different wavelengths are used in each channel, a plurality of transmission channels can be configured by one optical fiber.

According to the local area network of the present invention, the time division transmission channel and the contention transmission channel are assigned to different channels, respectively.
The optical fiber of the system enables stable transmission of signals that require real-time processing such as voice signals and video signals, in addition to transmission by the contention system with high line utilization efficiency, and transmission by the time division system.

[0023]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing a configuration example of an optical communication network of the present invention. In this embodiment, a plurality of 8-terminal passive star couplers 11 are used to connect each node 3 to the terminals of each star coupler 11, and the star couplers 11 are interconnected via a relay amplifier 12 to form a network. I'm building. Each node 3 and the star coupler 11 are connected via an optical fiber 4. The optical fiber 4 has a multimode core diameter of 50 μm and an outer diameter of 12
A 5 μm graded index optical fiber is used. Similarly, the star coupler 11 and the relay amplifier 12 are also connected via the optical fiber 4.

The 8-terminal star coupler 11 has a structure as shown in FIG. An optical integrated circuit of an optical waveguide is formed on a substrate 13 made of glass or polycarbonate to obtain a required distribution ratio. In FIG. 2, 14 is 2 × 2
The optical equalizer, 15 is a 1 × 2 optical unequal demultiplexer / multiplexer, and 16 is a 2 × 2 optical unequal distributor. Further, 17 is an optical fiber, R is a radius of curvature of the optical waveguide, and x _n and y _n are input / output components. Note that, for example, 1 × 2 means that there is one optical path on one end side and two optical paths on the other end side.

2 × 2 light equalizer 14 is shown in FIG. 3 and 1 × is shown in FIG.
The schematic structure of the two-optical unequal demultiplexer / multiplexer 15 and the 2 × 2 optical unequal distributor 16 is shown in FIG.

In the 2 × 2 light equalizer 14, as shown in FIG. 3, the widths of all the optical waveguides are set to be equal. Further, in the 1 × 2 optical unequal demultiplexer / multiplexer 15 and the 2 × 2 optical unequal splitter 16, as shown in FIGS. 4 and 5, the width of the optical waveguide varies depending on the required distribution ratio. Is different. The 2 × 2 optical unequal splitter 16 is composed of a 1 × 2 optical equal demultiplexer / multiplexer 16a and a 1 × 2 optical unequal demultiplexer / multiplexer 16b.

The structure of the star coupler 11 is, as specifically described in Japanese Patent Application No. 2-409070, a method of forming an optical waveguide by diffusing thallium (Tl) ions or the like into a glass substrate. The method can be realized by a method of irradiating the polycarbonate with ultraviolet rays to selectively polymerize it to form a refractive index distribution to form an optical waveguide, or a method of forming a combination of fusion-bonding type couplers.

In the star coupler 11 shown in FIG. 2, an optical signal 18 input via an optical fiber 17 is used.
Are sequentially branched at a predetermined ratio by a 1 × 2 light unequal demultiplexer / multiplexer 15, a 2 × 2 light unequal divider 16, and a 2 × 2 light equalizer 14, and are supplied to another optical fiber 17. . Optical signals from other optical fibers are combined at a predetermined ratio and supplied to a certain optical fiber 17.

The node 3 has a structure as shown in FIG. The transmission signals for two channels from the main body portion 19 of the node 3 are converted into optical signals by the light sources 20a and 20b, and then passed through the mode scrambler 21 and the 1 × 2 optical equal demultiplexer / multiplexer 22 for wavelength division multiplexing. The signals are combined by the demultiplexer / multiplexer 23 and transmitted to the network from the port 25. The mode scrambler 21 has optical waveguides arranged in a zigzag manner, and by uniformly distributing the optical signals from the light sources 20a and 20b into each mode of the multimode optical fiber, the passive star coupler It works to prevent the distribution ratio from varying.

The light emission wavelengths of the light source 20a and the light source 20b are different. In this embodiment, the light emission wavelength λ _a of the light source 20a is λ _a = 780 nm and the light emission wavelength λ of the light source 20b is λ _b = 870.
nm. Although the light sources 20a and 20b are made of AlGaAs Fabry-Perot type semiconductor lasers, they may be light emitting diodes. Hereinafter, the channel transmitted at the wavelength λ _a is referred to as channel A, and the channel transmitted at the wavelength λ _b is referred to as channel B.

The received signal from the network is port 2
After passing 5, the signal is transmitted to the wavelength division multiplexer / demultiplexer 23 and demultiplexed. The demultiplexed optical signal is input to the photodetectors 24a and 24b through the 1 × 2 optical equal demultiplexer / multiplexer 22. The light receivers 24a and 24b are silicon avalanche photodiodes (Si-APD). About half of the optical signal is transmitted to the light source 20a and the light source 20b, but this becomes a loss.

The wavelength division multiplexer / demultiplexer 23 is connected to the channel A.
The optical signal of wavelength λ _a is distributed to the side of
It is an element that functions to distribute the optical signal of _b . The wavelength division multiplexer / demultiplexer 23 is of the interference filter type, which is already commercially available. FIG.
Since the star coupler 11 shown in (1) is a passive type, even if signals of a plurality of wavelengths are mixed, signals are distributed without interfering with each other.

The relay amplifier 12 has a structure as shown in FIG. The optical signal input from one port 25 is separated into wavelengths by the wavelength division multiplexer / demultiplexer 23, and the optical receiver 24
After conversion into electric signals by a and 24b, amplification and waveform shaping are performed by the regenerative amplifier 26, and the electric signals are converted again into optical signals of respective wavelengths by the light sources 20a and 20b, and then wavelength division multiplexing / demultiplexing / combining. The waves are combined by the wave filter 23 and transmitted to the port 25 on the opposite side. The relay amplifier 12 is a bidirectional relay amplifier as is clear from FIG. 7. Relay amplifier 1
The 2 can be provided with so-called 1R, 2R, 3R, etc. functions as required. Here, the 1R function means that the signal is simply amplified, the 2R function means that the signal is shaped, and the 3R function means that the signal is re-encoded.

An amplifier for converting an optical signal into an electric signal and amplifying it once as shown in FIG. 7 is called a relay regenerative amplifier.
As the relay amplifier 12, a semiconductor laser amplifier which is a bidirectional optical amplifier may be used. Regarding semiconductor laser amplifiers, Shimada and Nakagawa “Research Status of Semiconductor Laser Amplifiers Used in Optical Communication Systems”, O plus
E, August 1989 issue, p106-p111. In this case, since the wavelength band with a gain of the semiconductor laser amplifier is about 50 nm (-3db), it is necessary to set the wavelength difference between λ _a and λ _b to about 30 nm. Therefore, when a semiconductor laser amplifier is used as the relay amplifier 12, for example, λ _a = 780 nm and λ _b = 810.
It is necessary to use a distributed feedback (DFB) semiconductor laser having a stable oscillation wavelength for the light sources 20a and 20b. This is because the Fabry-Perot type semiconductor laser has a temperature dependence of wavelength of about 0.3 nm / ° C. The wavelength of the optical signal does not have to be around 800 nm, and for example, the wavelength of 1.3 μm band or 1.6 μm band may be used. In particular, an optical signal of a specific wavelength is used to configure this optical communication network. There is no reason why you should use it.

Regarding the optical fiber, one having a different core diameter may be used, or a single mode optical fiber may be used instead of the multimode optical fiber. However, in these cases, the light source, the light receiving element, the star coupler and the like must be redesigned appropriately. As a bidirectional optical amplifier,
Rare earth doped fiber amplifiers are known as well as semiconductor laser amplifiers. Of the rare earth doped fiber,
An erbium (Er) -doped fiber amplifier having a gain in the 1.6 μm band has a gain wavelength band of 2 nm (−3d
Since it is as narrow as B), wavelength multiplexing is not impossible but difficult. On the other hand, it is known that the praseodymium (Pr) -doped fiber amplifier having a gain of 1.3 μm has a gain wavelength band of about 30 nm (−3 dB) (for example, Oishi et al. “Pr ^{3 +} -doped fluoride”). 1.3 μm band optical amplification characteristics of fiber ”, 1991 Autumn Meeting of the Institute of Electronics, Information and Communication Engineers, C-221 (1991)).
Therefore, as the rare earth-doped fiber amplifier applied to the present invention, the praseodymium-doped fiber amplifier is more suitable than the erbium-doped fiber amplifier. It is needless to say that the optical amplifier applied to the present invention is not limited to the above examples, and it is desirable that the optical amplifier has a wide amplifiable wavelength band. In the above description, the optical signal format is return-to-zero (RZ). In optical communication, a non-return to zero (NRZ) signal is often used, and thus distinction is required.

Next, an embodiment in which a multi-channel local area network is realized by using the optical communication network of the present invention will be described.

In the local area network of this embodiment, in the above multi-channel optical communication network, channel A is a channel that complies with the contention-type protocol for data communication, and channel B is a time division multiplex for voice signals and video signals. This can be realized by setting the channel according to the connection (TDMA) protocol. The signal transmission rate of channel A is 10 Mbps, the encoding format is Manchester code, and the protocol is IEEE80.
Compliant with 2.3. However, collision detection is determined to be a collision when a carrier is sensed during transmission. The signal transmission rate of channel B is 100 Mbps, the encoding format is 4B5B code, and the protocol is token passing (token bus method). The carrier frequency is 125 Mbps due to the nature of the 4B5B code in which 4-bit information is encoded into a 5-bit code string. The token is generated only by the time division control device 10 shown in FIG. The time-division control device 10 is no different from the general node 3 in terms of hardware.

Each node 3 first requests the channel allocation of the channel B to the time division control apparatus 10 via the channel A. The time division control device 10 allocates a line according to a certain algorithm based on the request and the state of use of the line.

FIG. 8 shows the signal flow on channel B, showing the signal flowing on channel B at some point in the network, with the horizontal axis representing time. Lines are assigned by sending out short packets 27 called tokens on the channel B at regular intervals. A packet is a group of code strings and is called a packet by analogy to a parcel. The address of the node that can be transmitted is written in the token, and when the token in which the transmission permission of the own node is written is received, the corresponding node immediately sends out the packet 28. In this case, the length of the packet is set in advance to a fixed length τ ₁ .

In this embodiment, the time-division control unit 10 has 1000 time slots (slot length T = 1 m) per second.
s) and allocate it to each node requesting circuit allocation. So the token is 1ms in this example
It is sent every time. Normally, 20 time slots are assigned to one node per second, so the signal delay time is about 50 ms. The maximum round trip propagation delay time of the network is 46.5 which is the standard of Ethernet (registered trademark).
Since μs is set, a dead time τ ₀ of about 250 μsec must be expected for each time slot as a time required for the token to travel through the network and a time required for control, and the packet length τ _{1 that} can be transmitted by each node
Was about 750 μs. Therefore, one packet contains about 75 kbits of information. Therefore, information of 1.5 Mbps can be transmitted in 20 time slots per second. Therefore, the line efficiency is 7
It will be 5%. The transmission rate of 1.5 Mbps with 20 time slots as one unit is "INS Net 1
This is because the connection with the circuit switching / packet switching service of ISDN called "500" is considered. That is,
In the "INS Net 1500", 24 channels of 64 kbps are prepared for 24 channels, and the total transmission amount is about 1.5 Mbps.

Since the optical communication network shown in FIG. 1 has bidirectionality, in the case of one-to-one communication, the time division control device 10 generates a token which permits transmission to two nodes at the same time. In the case of broadcasting, that is, in the case of one-to-many communication, the time division control device 10 generates a token that permits transmission to only one node. In the case of one-to-one communication, as described in the specification of Japanese Patent Application No. 2-98370, there is confidentiality, that is, the correct content is not transmitted to nodes other than the communication concerned node. Further, in the case of one-to-one communication, twice as many signal transmissions as in the case of one-to-many communication are performed. It is possible to add the above confidentiality and to double the transmission capacity,
This is a feature of the network shown in FIG. 1 and the above communication control system.

In this embodiment, the above-mentioned "INS Net 150" is used.
The above configuration was made because the connection with "0" and the transmission of the DVI video signal were mainly considered. However, when the transmission of the audio signal is necessary, the design of the time slot is changed to, for example, a unit of 64 kbps. It may be designed so that signal transmission can be performed. DVI is a technical method of band-compressed video signals and is currently being developed.

Further, in the above embodiment, only one transmission channel for signals requiring real-time property is provided, but it may be further increased if necessary. Further, the time division control device 10 may be responsible for controlling the increased channels, or another time division control device may be provided and a time division control device may be provided for each transmission channel. The latter is preferable from the viewpoint of improving the reliability of the network.

Although the present embodiment has been described above as a local area network compatible with multimedia, this local area network is digital P if the viewpoint is changed.
It has the same function as BX (private telephone exchange). Not only that, there is an advantage that it is possible to perform a broadcast (broadcast) that is difficult with a general digital PBX. Therefore, it goes without saying that the present application includes the use of the configuration of the present invention as a digital PBX.

In the above embodiment, the optical communication network performs wavelength multiplexing of two wavelengths, but the wavelength multiplexing degree can be further increased if necessary. When the semiconductor laser amplifier is used as a relay amplifier, the range of wavelength band with gain is from 50 nm (−3 db: 0.8 μm band) to 80 n.
Since it is about m (-3db: 1.3 μm band), for example, 1 nm
If wavelength multiplexing is performed every time, it is possible to amplify from 50 channels to about 80 channels with one relay amplifier.

In the above embodiment, the local area network has been described as an example.
Since 0 independent channels can be configured, the optical communication network according to the present invention can be applied to a wide area network such as an optical cable television (CATV) network.

[0048]

As described above, according to the optical communication network of the present invention, by one optical fiber,
Multiple transmission channels can be constructed, and the network can be easily expanded and reduced. As a result, the amount of optical fiber cables and optical connectors used can be reduced, and the space required for laying the optical fibers and the construction cost can be reduced as compared with the case where the network is configured by a plurality of optical fiber cables.

Further, according to the local area network of the present invention, a so-called multimedia local area network capable of simultaneously handling data communication and voice signals, video signals, etc. is realized in a network constituted by one optical fiber. be able to.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration example of an optical communication network of the present invention.

FIG. 2 is a plan view of an 8-terminal passive star coupler used in the optical communication network of FIG.

FIG. 3 is a plan view of a 2 × 2 light equalizer.

FIG. 4 is a plan view of a 1 × 2 optical unequal demultiplexer / multiplexer.

FIG. 5 is a plan view of a 2 × 2 light unequal distributor.

6 is a diagram showing an internal configuration of a node used in the optical communication network of FIG.

7 is a diagram showing an internal configuration of a relay amplifier used in the optical communication network of FIG.

FIG. 8 is a timing chart showing a state of time division control on a channel according to a time division multiple access protocol.

FIG. 9 is a diagram showing a general network.

FIG. 10 shows a network built using star couplers.

FIG. 11 is a diagram showing a multi-channel local area network.

[Explanation of symbols]

1 coaxial cable, 2 taps, 3 nodes, 4, 4
a, 4b optical fiber, 5 star coupler, 6 terminals, 7 light emitting element, 8 light receiving element, 9a, 9b bus, 1
0 time division control device, 11 8 terminal passive type star coupler, 12 relay amplifier, 13 substrate, 14 2 × 2 optical equal distribution, 15 1 × 2 optical unequal demultiplexer / multiplexer, 16 2 ×
2 optical unequal distributor, 17 optical fiber, 18 optical signal,
19 node body part, 20a, 20b light source, 21
Mode scrambler, 22 Optical demultiplexer / multiplexer, 23
WDM demultiplexer / multiplexer, 24a, 24b light receiver, 25
Port, 26 regenerative amplifier, 27 tokens, 28 packets, R radius of curvature of optical waveguide

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location G02B 6/28 U

Claims (5)

[Claims] 1. An input / output of an optical signal having a plurality of terminals
It is done from one terminal, there is no distribution of optical signal to its own terminal,
Moreover, the terminals other than the self terminal are connected to the terminals of the passive star coupler to which the optical signal is distributed, and the passive
Type star coupler terminal through optical fiber to node
To each of the multiple channels on the node.
Common optical fiber using different wavelengths to each other
An optical communication network characterized in that a means for transmitting and receiving via the above is provided, and a plurality of transmission channels are configured by wavelength multiplexing between the passive star couplers. 2. The optical communication network according to claim 1, wherein at least one of the plurality of transmission channels is a contention system transmission channel, and at least one of the plurality of transmission channels is a time division system transmission channel. When,
A plurality of nodes respectively connected to the optical communication network, and a time division control device provided corresponding to each of the time division transmission channels, in response to a request via the contention transmission channel. A local area network comprising: a time division control device that performs transmission through the time division transmission channel. 3. The optical communication network according to claim 1.
The star coupler corresponds to the terminal and
With a star coupler that has an internal configuration with a multiplexer
Optical communication network characterized by the fact that there is. 4. The local area network according to claim 2.
In, the time division control device transmits on the time division transmission channel.
The address of at least one station that allows
The station that has been sent a token and is authorized to transmit immediately after the receipt of the token.
Characterized by transmitting packets on a divided transmission channel
Local area network. 5. The local area network according to claim 4.
In the above, the token is covered by the address of the station that is allowed to send.
The two authorized stations to send the
Then send the packet on the time division transmission channel
Characteristic local area network.