JPH05207043A - Local area network - Google Patents

Abstract

PURPOSE:To connect the scanner and the printer of a color laser copying device by an optical fiber on the basis of a protocol different from FDDI to construct a network. CONSTITUTION:CMI-demodulation is executed by a demodulator 1, and a synchronizing signal for multiplexing is generated by the violation of bit 1, and the free-running clock of a PLL 2 is used for the demodulation, and a demodulated result is returned again to the PLL 2, and phase locking is executed. In a shift register 11, a serial picture signal is converted into the parallel signal of 8 bits. A control code serial/parallel-converted by the shift register 4 is fetched by DMA transfer and interruption operation from a microprocessor 7. A multiplexer 9 constitutes a frame by synthesizing a control signal from the shift register 6 and a picture signal delayed by a delay circuit 5. In a modulator 10, a signal obtained by delaying the frame synchronizing signal taken out of the demodulator 1 by several clocks is used as the synchronizing signal of the frame.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a local area network, and more particularly to a local area network in which a color laser copier original reading device (scanner) and a color laser copier printer are connected by an optical fiber. Is.

[0002]

2. Description of the Related Art Color laser copier (hereinafter referred to as CLC
It is needless to say that the scanner and the printer do not have to have a separate structure and may be integrated into a stand-alone color copying machine. However, considering the systemization of various printers, scanners, image processing computers, etc., there is a great deal of merit in configuring the scanner and the printer as separate units. The interface that connects the conventional CLC scanner and printer is designed to achieve high-speed transmission with 8-bit parallel transmission in order to transmit a large-capacity image signal at high speed. A signal line is required, and if a control signal is also included, a cable having a diameter of about 10 mm that bundles 25 or more signal lines is required.

[0003]

However, in the above-mentioned conventional transmission, the cable length is limited to within a few meters from the viewpoint of electrical signal attenuation and unnecessary radiation noise, and the system expandability is also limited. poor. For example,
It is not easy to output from one scanner to ten printers at the same time. Therefore, by using an optical fiber instead of an electrical cable, the above-mentioned problems concerning transmission distance and unnecessary radiation noise are solved, and the concept of a local area network is introduced as a link of the optical fiber. Although the scalability can be dealt with, there are the following problems in combining these conventional techniques.

That is, the amount of information of a color image signal is extremely huge. For example, when reading A4 size in full color of 300 dpi, the amount of data is 47 Mbytes. When this data is stored in the semiconductor memory, the price of the memory is considerably high. Therefore, the conventional CLC does not have a large capacity memory by driving the scanner and the printer in synchronization. I have to.

On the other hand, the current LAN protocol has a problem that the real-time property of communication is impaired because all data is packetized and transmitted. Therefore, it is necessary to temporarily store the data to be transmitted in the memory and transmit the data one by one by a device such as a microprocessor that can judge the condition. Therefore, there is a problem that in a normal LAN for transmitting data in packets, even a high-speed one such as FDDI cannot be used for transmitting the CLC image signal. The present invention has been made in view of the above points, and an object of the present invention is to provide a color laser copier original reading device (scanner).
Is to connect a color laser copier printer with an optical fiber, and to provide a local area network adopting a protocol different from FDDI.

[0006]

The present invention has the following structure as one means for solving the above-mentioned problems. That is, a plurality of image reading terminals and a plurality of image output terminals are provided, and the plurality of image reading terminals and the plurality of image output terminals are connected to each other by an optical fiber so that image information and control information can be obtained. In the local area network for transmission,
Dividing means for electrically time-dividing one optical fiber into a plurality of channels, and image information and control information are independent of each other among the plurality of channels divided by the dividing means. And a control means for controlling the transmission of the image information. The control means controls the transmission of the image information by the channel of the control information.

Further, a plurality of image reading terminals and a plurality of image reading terminals
An image output terminal, the plurality of image reading terminals,
Connect the image output terminals to each other by optical fiber
A local area network for transmitting image information and control information
In a network, one optical fiber can be electrically connected
Dividing means for time division into different channels, and image information and control
Information is divided into multiple channels by the dividing means.
Inside, channels are transmitted independently of each other
The transmission of the image information is controlled by the channel of the control information.
Control means to control and change of image information on output side of each terminal
The recognition device that recognizes the usage status of the
Receiving a call command for image information between terminals and
Image information channel of own terminal is in use
When it is recognized that the image information
Transfer with additional information indicating that the channel is in use
And means for doing so.

Further, it has a plurality of image reading terminals and a plurality of image output terminals, and the plurality of image reading terminals,
In a local area network for transmitting image information and control information by mutually connecting the plurality of image output terminals with an optical fiber, a dividing means for electrically time-dividing one optical fiber into a plurality of channels. , The image information and the control information are transmitted independently of each other among the plurality of channels divided by the dividing means, and the transmission of the image information is controlled by the channel of the control information. A control means and a recording means for recording the state of each terminal are provided, and the control means issues a packet corresponding to the state of the specific terminal based on the recording of the recording means to make one loop of the optical fiber loop. , Notifies other terminals of the state of the specific terminal.

[0009]

With the above-mentioned structure, it functions to make the transmission of image information and control information efficient. In addition, the connection control of the channel is smoothed, and the state of the terminal connected to the network can be accurately grasped, and the function of smoothing the connection control to the terminal is also achieved.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. FIG. 1 shows a network structure in which a plurality of scanners and a plurality of printers are connected in a loop by an optical fiber 100 in a local area network according to an embodiment of the present invention. In addition, FIG.
A computer 2 between the scanner 202 and the printer 204.
1 shows a connection form in which an interface unit (IPU) 203 for providing an interface with 01 is interposed. If the IPU function is made into a card according to the bus of the host computer, it can be mounted on the host computer and the host computer and the network by the optical fiber can be directly connected. In this case, up to 123 terminals can be connected.

Further, FIG. 3 is an example showing a connection form of a set of conventional scanner, printer, IPU, and computer. In practice, it is desirable to keep the number of terminals on the network at about 20 per loop because of the traffic. For example, it is assumed that there are about 10 scanners and about 10 printers. In this case, if one set of scanner / printer occupies the network,
In the worst case, another 9 sets of scanners / printers may be kept waiting.

However, it is possible to simultaneously communicate with 10 sets of scanners / printers adjacent to each other. If only one scanner is connected, up to 122 printers can be connected without any problem. Generally, when communicating via a network, one page of buffer memory is required. However, if an architecture that assumes real-time communication is adopted, the buffer memory becomes unnecessary by activating the scanner side and printer side at the same time. Become.

In the network of this embodiment, 123 Mb
It is specified that the CLC scanner and the printer are connected only by a pair of optical fibers of ps. Also, the image signal and the control signal are time-division multiplexed in one optical fiber,
The optical fiber has a single-ring form to make it full-duplex. This is inferior in reliability at the time of failure as compared with the FDDI having a double ring, but in the network of the present embodiment, it is assumed that a large amount of image data is sent though it is a small packet. The data transmission efficiency is higher than the exchange FDDI. Further, the optical fiber is strong against external noise, and even if an error of several bits occurs in the transmitted image signal, it does not cause a fatal problem. Therefore, in the network of this embodiment, for example, FCS ( An error detection mechanism such as a frame check sequence) is omitted.

4B / 5B used in FDDI
Since the coding method is inefficient in use of the line, the CMI code violation synchronization method is adopted instead. This system has the advantages over the FDDI in that the clock extraction is easier, the line utilization efficiency is higher, and the TDM (time division multiplexing) is easier. Here, 125
Since the optical fiber of Mbps is time-division (TDM),
The signals shown below appear to be individually wired. Image signal <I-Channel> Printer start signal <I-Channel> Page synchronization signal <I-Channel> Line synchronization signal <I-Channel> Control command signal (ASCII) <C-Channel> Of these, control command The signal is a synchronous RS
-232C, and transmits various control commands of layer 3 or higher in OSI.

Further, these signal groups are <I-Channel
<el-element> and <C-Channel> element, where <I-Channel> is the bulk
A bearer channel through which data flows, <C-Ch
<annel> is a control channel for controlling the connection of <I-Channel>. In addition, here, 100 Mbp
In order to handle ultra-high speed C-Channel data of s or more, a processor for I-Channel that controls the data may be, for example, an 8-bit one-chip microcomputer.

<Physical Interface> The standard of the optical fiber network according to this embodiment is defined as follows. Loop shape: Single loop Material: Quartz glass Core / clad diameter: 50/125 micron Fiber type: GI-50 Connector type: FC type Wavelength: 1.3 micron Clock: 144Mbps Image signal: 17.8Mbyte / s
ec (max) Control signal: 140 Kbps Modulation signal: CMI Frame synchronization: Bit 0 violation 1 Frame length: 128 octets Multiframe: 16 frames Access method: Network distributed maximum number of nodes: 123 Maximum distance between nodes : 1Km Maximum total loop length: 123Km Call control format: Connection type Switching format: Circuit switching Error control: None (image data) Parity check (control packet)

<Optical Fiber Interface Data Link> One frame is divided as shown in FIG. 4 in order to time-division-multiplex an image signal and a control signal. Where Tx
D0 to TxD7, and Rx Terminal ID
Are both transmitted using the bits of Serial Data. Since both of them have an 8-bit structure, 16 consecutive frames are required to send a total of 16 bits of data. FIG. 5 shows the frame structure. In FIG. 5, F [n] (n = 0 ... 15) indicates that it is a relative frame number counted from the frame of M3 = 1. The 16 frames from F [0] to F [15] are called a frame set. The CMI coding and the bioration synchronization method are publicly known techniques, and thus detailed description thereof will be omitted here.

<Address Control (Physical Address)> Each terminal such as a scanner, a printer, an IPU, a host computer, etc., has an independent physical address (Terminal) as follows.
ID number). That is, Terminal ID 0 invalid address Terminal ID 1 master terminal address (existing) Terminal ID 2-124 general terminal address (existing) Terminal ID 125 spare Terminal ID 126 system address (virtual) Terminal ID 127 broad cast (virtual) book In the system of the embodiment, the terminal having the address of ID = 1 is the master terminal and the other terminals are the general terminals. This master terminal can also have the function of a general terminal. The master terminal can also be installed as an independent device.

The architecture of the optical loop / local area network according to this embodiment is a network distributed type. That is, each terminal has a network function. Also, customarily, a part of a node that receives an optical signal is called an "incoming line", and a part that transmits an optical signal is called an "outgoing line". And each node is the I-Channel of the outgoing line from it.
It manages only the e-traffic. From this, the network of the present embodiment is characterized as being basically a network-distributed type. Needless to say, the master terminal has more processing than the general terminal, but call setup, busy control, call disconnection, and network-specific work are distributed to each terminal. ..

The method of processing a command byte received by a general terminal is as follows. 1. Same as own terminal ID number Rx Terminal I
The command byte of D is read and is not transmitted to the next terminal. 2. Rx Terminal I different from own terminal ID number
The command byte of D is ignored, and it is retransmitted to the next terminal without processing. 3. When the outgoing line is blocked, when a call setup packet that uses the outgoing line is received, a predetermined bit is set and transmitted to the next terminal. (The call setup packet is sent in broad cast mode.)

The master terminal also performs the following processing. That is, 1. Block command bytes for terminals that are not physically present and terminals that are stationary. 2. Regular terminals that are in operation are searched periodically, and if there is no response from them, it is considered to be in the stationary state (DM). 3. Accepting initialization from a stationary terminal. 4. Recovery operation at the time of network congestion. <Topology> The terminal ID numbers are connected in order from the smallest number as shown in FIG. 6, and the packet and the data flow in the direction of the arrow in the figure.

<Call control> For call control, for example, the terminal A
The sequence of normal communication when (ID = 3) sends an image signal to terminal B (ID = 7) is as follows. 1. Terminal A sends a "call" packet to terminal B. (Broad cast) 2. Each terminal (ID = 4, 5, 6) between terminals AB is <I
-Check the usage status of Channel>, and if it is free, pass the "call" packet without processing, and if it is busy, pass the "call with a busy flag" packet ("call not possible"). 3. Terminal B is busy with the received "calling" packet.
If the flag is not set, after preparing the printer,
The "incoming call acceptance" packet is returned to the terminal A, and if the busy flag is set, the "disconnect" packet is returned to the terminal A. 4. When the terminal A receives the "incoming call acceptance", it sends the image data to <I-Channel>. 5. Terminal A sends a "disconnect" packet to Terminal B. (Broad cast) 6. Each terminal (ID = 4, 5, 6) between terminals AB is <I
-After taking the procedure to release Channel>, pass the "cut" packet unprocessed. Temporarily, <I-Chan
If there is a problem opening el>, pass the "cut with error flag" packet. 7. Terminal B confirms that the error flag of the "disconnect" packet is not set, and returns the "disconnect acceptance" packet to terminal A. (To set the busy flag, invert the MSB of the parity byte.)

The flow of the above sequence is shown in FIG. Although the flow shown in the figure focuses only on the terminals A and B, a detailed sequence in which other terminals are taken into consideration is shown in FIG. The communication end / reception end command marked with * in FIG. 7 is a concept of the upper layer, and is transmitted by an “information packet” described later.

In FIG. 8, since the "calling" packet is transferred while being examined by each terminal, it can be seen that there is a delay each time it passes through the terminal. Since the "incoming call" acceptance packet has a one-way loop, it is sent to the terminal 7,
It makes a round and returns to the terminal A. Further, since there is no software delay with respect to image data, the image data arrives from the terminal A to the terminal B in substantially real time.

In FIG. 9, the terminal between the terminals A and B is <I
FIG. 7 shows a sequence diagram when using -Channel>. In this case, the terminal 4 receives the "call" from the terminal A.
By setting the busy flag of the packet, the packet is changed to a "call not possible" packet and transmitted, and terminal B is informed that this communication is impossible.

<Packet> The types of packets are as follows. Calling packet: Terminal sends <I to other terminals
Make a request to use Channel>. Incoming call acceptance packet: Shows that it is ready to receive. Disconnection packet: Indicates that the use of <I-Channel> is terminated. Disconnection acceptance packet: Indicates that <I-Channel> has been released. Information packet: Indicates that the packet includes information of the upper layer. Reject packet: A packet when the network cannot accept the terminal request.

<Packet Structure> FIGS. 10 to 12 show the structures of main packets. <State Transition Diagram> The general terminal takes one of the following states. Stop: Power is off DM: Not recognized by master terminal (Disconnect Mode) Idle: Is acknowledged by master terminal by issuing INIT Waiting for response: Waiting for permission to use I-Channel State Ready: State in which I-Channel is ready to be used Transmission length / Reception: State in which image signal is communicated by I-Channel FIG. 13 shows the mutual relationship as a state transition diagram.

<Timer> The general terminal has the following timers and manages the state transition. T. 1: The timer is started after the calling terminal issues a calling packet. Even if this timer times out, if neither an incoming call acceptance nor a disconnection request is returned, the outgoing call packet can be issued again. The number of retransmissions of this packet is N. It is limited to once, and it is judged to be busy at that time. T. 2: The general terminal is T. INIT once every two seconds
Can issue packets. This is repeated until DISK or UA is received, and when the UA packet is received, the state shifts to the idle state. When it receives a DISK packet, it stays in the DM state and the T.S. After waiting for 3 seconds or more,
T. 2 is reset and the retransmission of the INIT packet is started again. Here, T. 1 = 60 seconds T.I. 2 = 5 seconds T.S. 3 = 60 seconds N.Y. 1 = 3 times.

<Transmission of Image Data> Image data is transmitted using the I channel (D0 to D1015 in the frame). The first pixel of each scanning line is written in D0, and when the average transmission rate of image data is less than the average transmission rate of I channel (17.8 Mbyte / sec), there is no data to be sent. Since there is, a NULL frame is sent in that case. A FIFO of about 128 bytes is required for this speed matching.

<Description of Circuit Configuration> Next, modulation / demodulation of an optical signal according to this embodiment will be described. FIG. 14 is a block diagram showing modulation / demodulation of an optical signal in the local area network according to the embodiment. In the figure, an input optical signal is converted into an electric signal by a demodulator 1 including a photoelectric conversion element (avalanche diode), and CMI demodulation is performed. The synchronization signal for multiplexing the signals is produced by the bit 1 bioration. A free-running clock of PLL2 is used for demodulation, and phase synchronization is achieved by returning the demodulation result to PLL2 again. The clock created by this PLL 2 is a counter 3 that controls the entire sequence.
Also supplied. Since CMI modulation / demodulation is known, its detailed description is omitted. Further, the modulation system here is not limited to the CMI system, and may be, for example, the 4B5B system.

The input signal demodulated by the demodulator 1 contains an image signal, a control signal, and the above-mentioned bioration bit. Therefore, for example, in order to take out only the image signal, the counter circuit 3 makes a signal whose logic is "true" at a portion corresponding to the image signal portion, and the shift register 11 is driven by the clock of only that portion. To The shift register 11 also performs serial / parallel conversion of the image signal at the same time, converts the image signal into an 8-bit parallel signal, and sends it to a printer (not shown). Incidentally, 8
The conversion to the bit parallel signal is made in consideration of the compatibility with the conventional printer engine, and the image signal may be sent as it is.

Of the output of the demodulator 1, the control signal portion of M2 shown in FIG. 5 is sent to the shift register 4. This M2 is a bit assigned to transmit a serial / control signal, and transmits a receiving terminal ID number and serial transmission data as shown in FIG. Since the serial control signal that can be sent in one frame is one bit, 16 frames are required to transmit the data shown in FIG. This is called 16 multi-frame transmission. Also, for the command, one byte is transmitted in 16 multi-frames. Then, the command is sent in a series of 5 to 10 bytes.
This is called a packet. For example, since the calling packet has a 6-byte structure, 96 frames are required to receive this packet.

The head of the packet is a serial / synchronization signal (M
3). When sending the first byte of this packet, the M3 bit in the frame is also set to "1" and sent. This allows the receiving side to know the start of packet reception. Then, when the packet is started to be sent, the packet must be sent out without interruption. However, when there is no packet to be transmitted, the invalid byte (NULL byte) is continuously transmitted. Therefore, the receiving side needs a circuit for ignoring this NULL byte. Therefore, according to the present method, it is possible to transmit the packets together.

The control code serial / parallel converted by the shift register 4 is input to the microprocessor 7. The input method is DMA transfer or read operation from the microprocessor by interruption. Also, since the serial data is sent with the receiving terminal number, the microprocessor 7 can immediately retransmit the packet to the shift register on the output side when the packet is not addressed to the own terminal. However, when the packet related to the own terminal is being transmitted, the completion of the transmission of the packet is waited for, and then the packet addressed to another terminal is transferred. Then, the packet is transmitted via the shift register 6, and in sequence,
The operation opposite to that of shift register 4 is performed.

The delay circuit 5 is a circuit for delaying the <I-Channel> image signal and is not provided for a general terminal. In other words, it is provided only for the master terminal which exists only one in one LAN. The use will be described later. The multiplexer 9 combines the control signal (C-Channel) from the shift register 6 and the image signal (I-Channel) from the delay circuit 5 to form a frame in the format shown in FIG. The switching of the multiplexer 9 is performed by the switching signal generated by the counter circuit 3. In the case of a transceiver for a general terminal that does not have the delay circuit 5, an image signal (I-Channel) unrelated to the terminal itself is retransmitted without processing immediately after reception. Therefore, the data delay is only the minimum fixed delay required in the modulator / demodulator. This state is called "through" data.

On the other hand, control signal (C-Channel)
With respect to the above, the channel data is not "through" because a non-constant delay by the microprocessor 7 is introduced and the packet is processed. Further, since the processing speed of the microprocessor is slow, in order to send the data according to the format shown in FIG. 4, the data is sent with a delay of exactly one frame.

The transceiver of the local area network according to the present embodiment is, for example, 7.
Since it takes 1 microsecond, when the maximum of 123 terminals are connected, it takes at least 873 microseconds for the command to go around the loop. Since this value does not take into consideration the processing speed of the microprocessor of each node, a considerable delay actually occurs.

Such a discrepancy between the I-Channel data and the C-Channel data is not a serious problem in the specifications of the network, but it is C-Ch.
If the call control of the annel is delayed, I-Channel eventually
Since it also affects the throughput of l, the smaller the delay, the better. Therefore, a method of making the delay due to the microprocessor almost zero will be described later. The multiplexer 8 switches <I-Channel> to <I-Channel>.
When -Channel> is being used for another node, the multiplexer 8 is passing the signal from the delay circuit 5. On the other hand, when the terminal itself is transmitting, the signal from the shift register 12 is passed.

In the case of the LAN according to this embodiment, <I-Ch
The switching of “annel>” may be gradually performed, and invalid data may be allowed to flow. However, since it is actually for future expansion, <I-Channel
The switching of l> is also performed in synchronization with the frame. Then, the frame combined by the multiplexer 9 is converted into a CMI-converted optical signal by the modulator 10 including a laser diode and transmitted to the next node. As the frame synchronization signal used for the CMI bioration in the modulator 10, the frame synchronization signal extracted from the demodulator 1 is delayed by several clocks. This delay of several clocks is for synchronizing with the delay of the data due to the latch inserted to absorb the rounding of the waveform generated in the multiplexers 8 and 9.

<Explanation of essential part of network> As described above, the delay circuit 5 of FIG. 14 is provided only in the master terminal. This is gradually delayed in a general terminal <I-Chan
This is because the master terminal adjusts the cumulative delay of nel> so that the total delay amount becomes one frame (or a multiple thereof). Generally, when an optical fiber is connected on a loop to form a LAN, it is usually impossible to connect in the circuit switching mode, and therefore packet switching is used. However, in this embodiment, since the delay circuit 5 is provided, circuit switching can be realized. Below, regarding the main part of the delay circuit,
This will be described with reference to the block diagram shown in FIG.

FIG. 15 is a detailed block diagram of the delay circuit 5 of FIG. In the figure, the I-Channel data input to the shift register 20 is converted into an 8-bit parallel signal there. This is because the access time of the random access memory (hereinafter referred to as RAM) 22 is as slow as 35 nanoseconds, and the apparent data rate is lowered by performing the processing in parallel. Since the parallel output from the shift register 20 flows at the I-Channel clock rate, the D type latch (hereinafter referred to as the D latch) 2
1 holds data for 8 clocks.

The data rate in this embodiment is 55.5 nanoseconds, and the RAM of the C-MOS type can be used. Since this RAM is used to delay the data for a certain time, the address An need only be incremented regularly. That is, the RAM 22 sets the delay amount of the I-Channel data to 1 byte (8 bytes) as described later.
Control in bit units. Also, the output On of the RAM 22
Are latched by the D latch 23 for waveform shaping.

I-Chn waveform-shaped by the D latch 23
The nel data is delayed in barrel shifter 24.
Unlike the normal shift register, the barrel shifter 24 can control the delay amount from the outside. The control signal is the barrel select signal. In the present embodiment, the barrel shifter 24 has at most 0
Since it is only necessary to obtain a delay amount of 7 bits, a delay amount of 8 bits or more is realized in combination with the RAM 22.

The output of the barrel shifter 24 is received by the waveform shaping D latch 39 and input to the FIFO 41. Since the signal that has passed through the optical fiber loop contains jitter, the delay amount is rarely a positive multiple of 1 bit, and its value changes with temperature changes and the optical fiber is physically Even if it is bent, it changes. Therefore, a RAM that can obtain the delay amount of any positive bit number
The combination of 22 and the barrel shifter 24 is not sufficient, and the FIFO 41 absorbs the variation of the delay amount of 1 bit or less.

Here, the address control circuit of the RAM 22 will be described. The RAM 22 performs writing and reading while the I-Channel has passed the time corresponding to 8 clocks.
The write time and the read time are equivalent to 4 clocks. Therefore, the data selector 2
5 alternately switches the address information of two systems every four clocks.

The address for writing is made by the counter 26, and the address for reading is made up of the adder 28 and the D latch 27. All the outputs of this D latch 27 are added by the adder 28.
By feeding back to one input Bn, inputting a binary "1" into the other input An and applying a clock to the D latch 27, the system operates as a binary counter. When the value applied to the input An is set to "2", D
The output of the latch 27 is incremented by "2" and
If "0" is applied to, the output of the D latch will stop at the value at that time.

D latches 29, 30, 32, and NAND
The gate 31 receives the control signal “Increment” from the microprocessor 7 from logic “L” to “H”.
When it changes to, it becomes "H" for one clock. As a result, the value "2" is supplied to An, and the output of the D latch 27 advances by 1 with respect to the counter 26. Similarly, by the D latches 35, 36, 38 and the NAND gate 37,
When the "Decrement" signal changes from the logic "L" to "H", "0" is applied to An and the D latch 2
The output of 7 lags counter 26 by one.

<Delay Amount Measuring Circuit> In the above delay circuit,
Since the amount of delay to be set varies depending on the number of terminals in the optical loop, it has to be actually measured. Hereinafter, the circuit for measuring the amount of open loop delay of the optical loop will be described with reference to FIG. Incidentally, in the figure, the same components as those of FIG.

The measurement of the delay amount shown in FIG. 16 is performed at the master terminal. In the master terminal, modulator 1
As a clock of 0, a built-in clock 50 is used.
The switch 51 is a data selector for switching the clock to the modulator 10 between the master terminal and the general terminal. When the optical output of the modulator 10 returns to the demodulator 1 through the loop in such a configuration, the frame synchronization signals SYNC1 and SYN are output from both the modulator 10 and the demodulator 1.
C2 is obtained respectively. Both SYNC signals are applied to the SR latch 52, and its output Q is from SYNC1 to SYNC.
During the period up to 2, it becomes a logical "H". The counter 53 is RS
The flip-flop 52 measures the amount of one-round delay by counting the number of clocks which are logically "H".

Further, the counter 53 is reset by the frame synchronization signal SYNC1 of the modulator 10 and simultaneously starts counting. Since the output of the counter 53 is read by the microprocessor 7, the SY signal is output simultaneously with the end of counting.
It is latched by the parallel shift register 54 by NC2. The delay amount obtained here is a relative delay amount between the transmission frame and the reception frame, and the value does not exceed the transmission time of one frame. The value set in the delay circuit 5 is obtained as follows. That is, [set value] = [number of clocks required to transmit one frame] − [measured one-round delay amount] ...
(1)

As described above, according to this embodiment,
A local area network is composed of optical fibers, and one optical fiber is divided into a packet channel and a circuit channel by time division, and signals with slow command lines are transmitted in the packet channel. By efficiently transmitting high-speed signals of the data system by the circuit channel, the interface between the full-color scanner and the printer can be realized without expensive page memory, and 10 to 20 scanners can be realized. The printer / printer can be connected to each other like a LAN by using this interface.

In the block diagram showing the modulation and demodulation of the optical signal shown in FIG. 14, the shift register 11 is used as the LAN interface for the printer and the scan interface is used as the interface for the printer and the scanner without using the system that also serves as the interface of the printer. The shift registers 12 may be separately provided in each LAN interface. Then, at this time, the attributes are clarified by the terminal identifier in the packet so that the commands for the scanner and the printer are not mixed up.

Further, in the block diagram shown in FIG. 16, inputs SYNC1 and S of the RS flip-flop 52 are provided.
Cross YNC2, parallel shift register 5
By replacing the SYNC2 input to the enable terminal of 4 and the SYNC1 input to the clear terminal of the counter 53, the calculation by the above formula (1) becomes unnecessary,
It is possible to match the value set in the delay circuit 5 with the actual measurement value. That is, [set value] = [actual measured value]
... (2). The present invention may be applied to a system including a plurality of devices or an apparatus including a single device. Further, it goes without saying that the present invention can be applied to the case where it is achieved by supplying a program to a system or an apparatus.

[0054]

As described above, according to the present invention,
By dividing one optical fiber into a channel for transmitting image information and a channel for transmitting control information so as to match each signal speed by time division, there is an effect that a signal can be efficiently transmitted. .. Particularly, in the invention described in claim 3, the connection control of the channel is facilitated, and in the invention described in claim 5, the signal transmission accurately grasping the state of the terminal connected to the network is performed. The effect is that it will be possible.

[Brief description of drawings]

FIG. 1 is a diagram showing a network structure in which a plurality of scanners and a plurality of printers are connected in a loop by an optical fiber in a local area network according to an embodiment of the present invention.

FIG. 2 is a diagram showing a connection form in which an interface unit for providing an interface with a computer is interposed between a scanner and a printer;

FIG. 3 is a diagram showing an example of a connection configuration of a conventional set of a scanner, a printer, an IPU, and a computer,

FIG. 4 is a diagram showing a frame structure of time division multiplexing;

FIG. 5 is a diagram showing an M2 bit multi-frame structure;

FIG. 6 is a diagram showing a terminal ID number and a data flow,

FIG. 7 is a diagram showing a call control communication sequence when an image signal is sent,

FIG. 8 is a diagram showing details of another call control communication sequence;

FIG. 9 is a terminal between terminals A and B, which is an I-Channel.
Diagram showing the sequence in use,

[Figure 10]

FIG. 11

FIG. 12 is a diagram showing the structure of a main packet,

FIG. 13 is a state transition diagram of a general terminal,

FIG. 14 is a modulation / demodulation block diagram of an optical signal in a local area network according to an embodiment.

FIG. 15 is a block diagram showing a main part of a delay circuit,

FIG. 16 is a diagram showing a circuit for measuring a loop delay amount of an optical loop.

[Explanation of symbols]

 1 Demodulator 2 PLL 7 Microprocessor 8, 9 Multiplexer 10 Modulator 24 Barrel shifter 26 Counter 27 D latch 28 Adder 52 RS flip-flop 54 Parallel shift register

Claims (6)

[Claims] 1. A plurality of image reading terminals and a plurality of image output terminals, wherein the plurality of image reading terminals and the plurality of image output terminals are mutually connected by an optical fiber to provide image information. In a local area network for transmitting control information, a dividing means for electrically time-dividing one optical fiber into a plurality of channels, and a plurality of channels for dividing image information and control information by the dividing means. , A control means for controlling transmission of the image information independently of each other, wherein the control means controls transmission of image information by a channel of control information. .. 2. A means for adding information for specifying a destination terminal of the information to the information transmitted by the channel of the control information, and when the information is not addressed to the own terminal, the information is taken into the own terminal. 3. The local area network according to claim 1, further comprising means for immediately retransmitting to another specific terminal. 3. A plurality of image reading terminals and a plurality of image output terminals, wherein the plurality of image reading terminals and the plurality of image output terminals are mutually connected by an optical fiber to provide image information. In a local area network for transmitting control information, a dividing means for electrically time-dividing one optical fiber into a plurality of channels, and a plurality of channels for dividing image information and control information by the dividing means. , The control means for transmitting the image information independently of each other and controlling the transmission of the image information by the channel of the control information, and recognizing the usage status of the channel of the image information on the output side of each terminal. When the recognizing means and its own terminal receive a call command for image information of other terminals and the recognizing means recognizes that the image information channel of its own terminal is in use. Local Area Nets workpiece and having a means for transferring by adding information indicating that the image information channel for the expression call command is in use. 4. The local area network according to claim 3, wherein the own terminal discards the call command and returns a channel busy response to the terminal which issued the call command. .. 5. A plurality of image reading terminals and a plurality of image output terminals are provided, and the plurality of image reading terminals and the plurality of image output terminals are mutually connected by an optical fiber to provide image information. In a local area network for transmitting control information, a dividing means for electrically time-dividing one optical fiber into a plurality of channels, and a plurality of channels for dividing image information and control information by the dividing means. , Control means for transmitting the image information independently by mutually independent channels and controlling transmission of the image information by the channel of the control information, and recording means for recording the state of each terminal, the control means Is characterized by issuing a packet corresponding to the state of the specific terminal based on the recording of the recording means to make one loop through the optical fiber loop and notifying other terminals of the state of the specific terminal. Local area the net work to. 6. The local area network according to claim 5, wherein when a specific terminal is not participating in communication, control commands relating to the terminal are ignored.