JPS5864847A - Data communication system - Google Patents

Abstract

PURPOSE:To economically transmit data in a premises via telephone pair lines installed in the premises, by providing a folding device connected with a plurality of communication devices with two-line type semiduplex lines and the communicating device connected to data devices. CONSTITUTION:A communication device 21 adds an address and a check code of a data device 1i to a data received from a data device 11 for packet. The device 21 transmits the packet to a device 400 via a two-line type semiduplex line 31 except during the reception of packet from a folding device 400. The device 400 makes only the packet effective, which reaches the device 400 fastest when the transmission of packet is started almost at the same time from some communication devices, folds back the packet to the other communication devices, rejects the reception of packets arrived with a delay and informs the rejection of reception for the communication devices transmitting the packets. When a communication device 2i detects that the packet is addessed to the device itself, the data section of the packet is transmitted to the device 1i. The other communication devices abolish the packets.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a communication system that allows data communication to be carried out economically within a relatively small area such as a campus.

In general, a network that exchanges data within a relatively small area, such as a campus, is called a local network, and is usually constructed of a computer and a group of terminals connected to it. - In recent years, a variety of new tights local networks have been realized that allow data communication without using exchanges, and a representative example is the Ethernet network developed by Ki Xerox.
As shown in Figure 1, this Ethernet is a system in which a large number of communication devices communicate with each other using one coaxial cord as the entire medium, and C8MA/C
D (Carrier Sense Multiple Ac
cess with Co11isinnDetect
Communication is carried out based on communication principles conventionally used in wireless local networks called ion. In FIG. 1, ll"li is a data device, 21 to 21 are communication devices, 3 is a 1oJI [Il:l-), 41 to 4i1d are trans-nodes, and 5 is a termination circuit.

The present invention is based on a communication principle similar to that of Ethernet, and in order to clarify the difference between the two, the communication method in Ethernet will be explained first.

In FIG. 1, for example, when the data device 11 sends data, the communication device 21 converts the data into 7 bits and transmits the state of the coaxial cord 3, which is the communication medium, to the transceiver 4.
If no carrier is detected on the coaxial cord 3, it is assumed that the carrier is empty and the coaxial cord 3 is immediately connected to the
If a carrier is detected, it waits for the carrier to become empty and then sends out a data node. Also, the state of the coaxial cord 3 is monitored during data transmission, and when it detects that another communication device has started transmitting a data packet, data transmission is stopped, and the above transmission starts again from the beginning after an appropriate time interval. Perform the steps. The other communication devices 22 to 2i monitor E' while transmitting data on the coaxial cord 3 through their respective transceivers 42 to 41, and detect a node addressed to them by the communication part of the communication device. Then, all the packets are lost and transferred to the data device connected to itself. Note that the termination circuit 5 is provided to prevent reflections at both ends of the coaxial cord.

In this way, Ethernet does not have a central control device like the host in conventional computer networks, and each communication device can exchange data based on its own judgment.It also uses coaxial cords. Because it is capable of high-speed transmission, it is highly scalable and has a communication capacity that allows thousands of terminals to communicate by resonating a single coaxial cord. You can build a network.

However, since systems such as Ethernet require the use of coaxial cables, it is necessary to install new coaxial cables when constructing a network. The initial cost for installation work is relatively high, and the transceiver circuit must be installed closely to the coaxial cord to avoid deterioration of the transmission characteristics of the coaxial cord due to bridged taps. There are drawbacks such as the restriction that the installation position of the communication device is limited to the vicinity of the coaxial cord, or the disconnection of the coaxial cord will make all communications impossible. The present invention eliminates the above drawbacks. In order to do this, data communication based on the O8MA/CD system can be performed using, for example, a wire pair already installed in the premises for telephone use as a communication medium, which will be described in detail below.

FIG. 2 shows an embodiment of a data communication network according to the present invention, in which the same components as those of the conventional example shown in FIG. 1 are denoted by the same reference numerals. That is, 11-11 are data devices, 21-21 are communication devices, 31-31 are two-wire half-duplex lines (hereinafter simply referred to as exciters), 400 is a return device, 411-4i1 are gate circuits, 412-31 are 4i2
are signal detection circuits, 413 to 4i3, 414 to 4i4
is a signal line, 415 to 4i5 are gate control circuits, 41
6 to 4i6 are signal lines, 61 is a return line, 62 is a common control circuit, and 63 is a control line.

The communication method will be explained below using FIG. 2. - As an example, consider the case where the data device flk:11 transmits data to the data device 1i. Note that the data devices 11 to 1 herein are
1 refers to various data devices such as teletypes, printers, or computers. The data device 11 first transfers data to the communication device 21. There are no particular restrictions on the interface or data transmission procedure for data transfer in this section, and it is sufficient to select a method that is convenient for both the data device and the communication device.

On the other hand, communication between the communication devices 21 and 21 via the line 31.3i and the return device 400 is performed based on an interface and a communication procedure specific to this network. This communication method is the main feature of the present invention, and its operation will be explained in detail below.

The communication fjt21 divides the data received from the data device 11 into data blocks of about 100 bytes, and each data block has the destination communication device 2I and data device I.
Packet 2 with I7 address and check code added
The first communication frame fk21 to be created sends packet 2 to the return device &400 through the entire line 31, except when it is receiving a packet from the return device 400. If the packet y is being received and is intercepted, the packet l is sent out after the reception is completed. Note that "receiving a communication device" does not necessarily mean that a packet addressed to oneself is being received, but also includes a case where a packet 2 addressed to another communication device is being intercepted.

Although the following explanation describes the operation of the return device [400], a particularly important point in the function of the return device 4oo is that when several communication devices start transmitting packets at almost the same time, the most Only the packet that arrived at the loopback device early is valid and is looped back to all the communication devices of the two packets, and for the nomikerato that arrived late, the packet is refused to be accepted, and the packet is sent to the communication device that sent the packet. The point is that it has a thousand attacks to tell that packet acceptance has been rejected.

The packet transmitted from the communication device 21 is added to the gate circuit 411 of the loopback device 400 via the line 31 and then sent to γL.
Ru. At this time, the signal detection circuit 412 attached to the gate circuit 411 detects that the packet has arrived through the signal line 413, and changes the state of the signal line 414 from 110 to 1. Is the gate control circuit 415 attached to the gate circuit 411 a signal? When N414 changes from O to 1, the mode is switched to reception mode, and the state of signal 416 changes from O to 1.

The gate circuit 411 is controlled by a signal line 416, and the gate opens in the direction of passing the packet to the Keratra return a61 (added from the line 31), and the packet is sent to the gate circuit 4.
11, is added to the other gate circuits 421-4 via the return line 61, and is sent to the communication devices 22-21 via the circuits 432-3I. Each communication device 22-2I
checks the address and check code of the received nomikerato, and if it is addressed to itself, sends the packet. In this illustrative example, the communication device k 2 i takes in the packet, decomposes the packet into the Nonoket 2, and extracts the necessary data and partial sound data device 11.
@Other communication devices discard the received packet due to a mismatch in the delivery address.

On the other hand, the output line 414 of the signal detection circuit 412 is also applied to the common control circuit 62. The common control circuit 62 includes signal extraction circuits 412 to 41 associated with each gate circuit 411 to 411.
The state of the output lines 414 to 414 of 2 is monitored, and when it is detected that the state of one or more output tubes has become 1, the control line 63 is
Change from kO to 1. Control $63 is each gate control circuit 4
15 to 4i5 are connected by the following formula, and each gate control circuit 415 to 415 is
Signal detection circuits 412 to 4i2 connected to each
It does not accept a change in the status of the output devices 414 to 4i4 from 0 to 1. In other words, a packet from one communication device is not received by the return device 400 and is returned.
'& Ij means that even if a packet is sent from another communication device, it will not be accepted by the return device. In addition, each communication device 22-21 does not receive the signal on the lines 31-31 even while transmitting a packet. Therefore, if the packet sent by itself is rejected by the return device 400 due to other first-arriving packets, the communication device 2
2 to 21 monitor the mixed signal of the first packet returned by the return device 400 and the packet sent by itself, and detect that the packet sent by itself was not received. This can be done and the packet transmission will be stopped.

An example of a circuit realizing these functions is shown in FIG. In FIG. 3, the gate circuit 411 consists of a receiver 411a that receives packets from the line 31, a gate 1411b that sends the received packets to the return line 61, and a gate 411C that sends the packets from the return line 61 to the line 31 side.
Each gate of 11b and 4110 has a gate control circuit 41.
Controlled by control line 416 from 5, only 10,000 is open at all times.

The signal detection circuit 412 is a fixed multi-pie break capable of writing triggering, and is triggered through a signal line 413 to keep the output at 1 every time the output of the line receiver 411a changes. In this case, the signal format to be transmitted is P.
1 like E code (phase encoding code)
If a code with a state transition for each bit is used, the duration of the monostable maf vibrator is set slightly longer than the total 1 bit (7) period, so that the output of the signal detection circuit 412 is You can keep it at 1.

The common control circuit 62 includes an OR circuit shown by reference numeral 62a in FIG. 3, and when one or more of the input lines 414 to 414 becomes 1, the output becomes 1.

The gate control circuit 415 consists of two AND gates 415a and 415b and an RS flip-flop 415C% delay circuit 415d. AND game 1' 415a generates a pulse with a width corresponding to the gate delay of the OR circuit 62a when the output of the signal detection circuit 412 transitions to 1 and the outputs of all other signal detection circuits are O, and the RSS flip flow rough occurs. 4.5ct = set. , xND gate 415b generates a pulse corresponding to the delay width of delay circuit 415d when output 414 of signal detection circuit 412 transitions from 1 to 0 (at the rear end of the packet), and resets all R8 flip-flops 415c. Output line 416 of RS fritz flop 415C. is set to 1 while receiving a packet.

Gate 411b in gate circuit 411 is opened to allow packets from line 31 to pass to return line 61111.

An example of the timing of the above operation is shown in FIG. A- in Figure 4
The F signal corresponds to the same symbol in the third (¥1). A is an example of the received PE code, B is the output of the signal detection circuit 412, C is the output of the common control circuit 62, D, B, F is an RS flip-flop 415 in the gate control circuit 415, respectively.
These are the set input, reset input, and output of d.

With the operation described above, data can be sent and received between any two data devices, and if two or more communication devices send packets at the same time, only one packet will be automatically selected in the return device. In extremely rare cases, two or more packets are sent back to the loopback device 4 with a time difference of about the gate delay time of the common control circuit 62 (several 10 n5ec). When a packet arrives, the mechanism that selects only one packet may not work, and the okerats may collide in the return device, causing interference. In such a case, the communication device that sent the packets cannot detect that a collision has occurred, but the receiving communication device detects bit errors due to interference and discards those packets. Therefore, this is normal packet transmission. This is equivalent to a case where a transmission error occurs during the process, and the probability of such a collision occurring is sufficiently low that it is hardly a problem. A twisted wire pair is used as a communication medium.IC
If you use a suitable commercially available cable driver/receiver, for example, if the distance between the communication device and the return device is set to a maximum of IKm, it will be 120 kb + t/S.
The communication speed of culm is obtained. At this speed 08MA/C
Multiply by null-jitter 08 that can be realized with method D, and further add 10% of the overhead portion such as Nonoket's address part.
If (0, 1) is subtracted, the effective communication capacity is 120kbit.
/S X 0.8 X O,9” 86 kbit/S
Therefore, if the average communication speed of the data device is k 1.2 kh/S, then communication of 70 channels can be performed simultaneously, which is 70 out of 86 k / 1.2 k. This communication capacity is considered to be sufficient for a small-scale local network.

In addition, if appropriate line equalization means are used, a communication distance of 5 km and a communication speed of 200 to 300 using twisted wire pairs for telephone communication can be achieved.
It is known that the AIO characteristics can actually be increased as the kbit/S increases, and the range of application can be expanded using this method. Since it can be configured easily, the cost per line is low and reliability is sufficient. In addition, a power outage of an individual communication device or a disconnection of a communication line is treated as the same as no signal, so that it does not affect communication between the stations.

As explained above, according to the present invention, the initial cost is low and a sufficient communication response can be obtained as a small-scale local network by making use of several telephone wires in the premises and using a return device with a simple circuit configuration. This has the advantage of being able to construct a highly reliable data communication system that has sufficient capacity and does not affect the overall communication even if a line disconnection occurs.

[Brief explanation of the drawing]

Figure 1 is a configuration diagram of a conventional data communication network, Figure 2
The figure is a configuration diagram of a data communication chitwork showing one embodiment of the present invention, FIG. 3 is a circuit diagram showing details of a return device, and FIG.
The figure is a time chart for explaining the operation of the circuit of FIG. 3. 11-11...Data equipment, 21-21...Communication device, 31-31...2-wire half-duplex circuit, 400...
Folding device, 411-4i1 gate circuit, 412
~412...Signal detection circuit, 413~413. 414~4i4... Signal line, 415~415 ・
...Gate control circuit, 416-416 ...Signal line,
61... Return line, 62... Common system # circuit, 63...
...Control line. Patent applicant: Oki Electric Industry Co., Ltd. Patent attorney Yoshi 1) Takashi Sei Figure 1 Figure 2 Figure 3 Figure 4

Claims (1)

[Claims] It has n communication devices connected to the data device, a return device that connects all the communication devices, and a two-wire semi-2N line that connects each communication device and the return device, and the communication device is generated from the data device. All data is converted into packets, and when another nokerato is not sent from the loopback device, the packet sent to the loopback device is received and inspected, and a mismatch with the original packet is detected. n and the Nonoket Sendout Meeting stops, ? r After the transmission of other packets from the return device is stopped, the packet is retransmitted from the beginning, and when a communication device that is not currently sending a packet receives a packet ' from the return device, if the packet is addressed to itself, it receives the data of the packet. to a data device connected to you,
If the packet is not addressed to itself, it is discarded, and when the return device detects that one communication device has started sending packets, it returns the packet i to the packet i, unless it is already accepting another packet. We will accept the request and return it internally, and the communication will be processed.
The packet is transferred to all communication devices except for the packet, and until the transfer of the packet is completed, the a packet' from other communication devices is
A data communication method characterized by not accepting l.