JPH0354495B2 - - Google Patents

Description

[Detailed description of the invention]

(1) Technical field of the invention The present invention generally has 30 ports and a communication speed of 9600bps.
A method and system for interconnection between terminals and processing units in the range of 240 ports and 1200bps communication speed, more specifically, without making any special changes to existing operating systems or developing new operating systems. The present invention relates to a mutual communication method and a mutual communication system that can support the above mutual communication. (2) Technology background In information processing network systems used in environments consisting of many laboratories such as universities, it is difficult to run various operating systems used by one or several users. A large number of small and medium capacity computer systems are provided. The university environment described above is the most suitable for implementing a local computer network (LCN), since the devices that make up the system are usually conveniently located in one or more buildings in close proximity. It is. (3) Prior Art and Problems However, many of the terminals or processing devices used in each laboratory are of different types, making the installation and operation of the above-mentioned LCN complicated. In other words, different processing device hardware typically requires different network hardware, and the operating system requires a large number of LCN software to achieve this diversity. In addition to requiring multiple rewrites, a special drive system had to be developed, creating many problems. (4) Purpose of the Invention In view of the above-mentioned conventional problems, the present invention provides an interface to a host computer in a special location where compatibility between hardware and software can exist, thereby providing software for an operating system. A method for interconnecting terminals and processing devices that allows access to any processing device connected to a network of many terminals without modification, development or use of complex network operating systems; and The purpose is to provide a mutual communication system. (5) Structure of the Invention According to the present invention, the purpose is to perform frame synchronization using a vertical synchronization signal constituting an NTSC synchronization signal, perform slot synchronization using a horizontal synchronization signal, and perform one or more slots of one frame as a unit. In a method for mutual communication between a terminal and a processing device or between processing devices using a time division multiplexing method in which corresponding slots of a plurality of consecutive frames are assigned as one time slot, the time slot is used for utility data. A communication device between a terminal device and a processing device or between a processing device and a device comprising a field and a control bit field, and controlling the connection between each terminal device and a processing device or between each processing device based on the bit data of the control bit field. one or more devices in a coaxial cable that connects a sync generator generating an NTSC sync signal at one end and a terminator at the other end to provide a baseband channel. A cable insertion device for supplying coaxial cable access to a tap device is installed, and one or more tap devices connected to a terminal or a processing device are connected to the cable insertion device, and the synchronization occurs. Frame synchronization is performed using the vertical synchronization signal included in the NTSC synchronization signal generated by the device, and slot synchronization is performed using the horizontal synchronization signal. The time slot is allocated as one time slot, and the time slot consists of a utility data field and a control bit field. Based on the bit data of the control bit field, connection control is performed between each terminal and the processing device or between each processing device. This is achieved by providing an interconnection system between terminals and processing devices or between processing devices using a time division multiplexing method. (6) Embodiments of the invention Examples of the invention will be described below with reference to the drawings. The mutual communication method and mutual communication system between a terminal device and a processing device (hereinafter, the system and method will be simply referred to as the present system) according to the present invention are as follows:
NTSC (National Television System)
It is based on a fixed transmission time slot assignment derived from synchronization signals and a baseband bus structure. Figure 1 shows the International Organization for Standardization (ISO) of this system.
This is a structural diagram showing the communication between the layer based on this system and each device configuring this system.
RG-11 coaxial cable transmission medium; 2 is a cable insertion device; 3 and 4 are e.g. RS-422 transceiver cables; 5 is a tap device based e.g. on a Motorola M6802 microprocessor;
is an arrow indicating interconnection with a terminal device or host computer, for example by RS-232; A indicates a physical layer; B indicates a data link layer; and C indicates a higher level functional layer. In Fig. 1 above, among the three layers at the lowest level, the physical layer provides a baseband channel of, for example, 500 kb/sec through the coaxial cable transmission medium 1, and the signal function is for medium-band one-division CATV.
It shall be compatible with frequency division multiplexing in either Community Antenna Television (Community Antenna Television) systems or dual cable broadband LCNs. Additionally, this system is centrally synchronized.
Since it is a TDMA (Time Division Multiple Access) mechanism, the physical layer A also extracts the synchronization signal from the cable 1. Two synchronization signals and a master reset signal received from cable 1 and a serial data stream are carried to data link layer B. Another serial bit stream is sent from data link layer B to physical layer A for transmission onto the cable. As shown in Figure 1, the physical Layer A hardware, consisting of the receiver, remote component cable drivers, synchronization extraction circuitry, and master reset detector,
Installed in a separate assembly from Link B and higher level components. This assembly, called cable insertion device 2, has its own power supply and provides coaxial cable access for up to four tap devices (the part of the access module excluding the cable insertion device). is inline with RG-11 coaxial cable 1. By sharing a single cable insertion device 2, the consumption of cable insertion power supply and cable connection hardware is offset and takes advantage of the fact that several tap devices are likely to be placed in close proximity. Branching to tap devices is supported by RS-422. Optical isolation on the RS-422 receiver minimizes ground loop problems. At data rates of 500 kb/sec used on the receiver and transmission data lines, the distance between the cable insertion device 2 and the tap device 5 can be up to 50 meters. Tap device 5 contains the data link layer B of the system and many of the higher level (network and transport) procedural functions. Most of the functionality at the data link level is a special 20-bit
This is done via an LS-TTL receiving/transmitting device. Such data link functions include addressing, address identification, data wrapping, and serialization/deserialization. Most of the details of link management are performed by the tap device microprocessor in accordance with signals provided by the data link hardware. Higher level functions performed by the microprocessor include end-to-end connection monitoring, data flow control and buffering, and client layer communications. The system accepts asynchronous serial data on the RS-232 line, indicated by arrow 6, as either 8 data bits without parity or 7 data bits with parity. Details such as number of stop bits, line data rate, and use of EIA modem control lines can be selected through switches and software. The preferred mode of operation is 8 data bits, no parity, 1 stop bit, 9600 bits/second. Transfer of messages between the higher level layers C is typically done through ASCII strings, and existing host terminal software is used to communicate with the tap device. Tap devices connected to terminals then have software that allows the user to interact with the network via menus and question-and-answer sequences, and tap devices connected to computers exchange control and status information with connected hosts. Use a single ASCII character to Also available are commands that allow the user to run listen-only modes, such as monitoring special links, or enabling XON/XOFF character processing. The latter is called flow control mode, which allows you to use lower speeds (e.g. 9600bps) on the net.
(below) equipment can be used. for example,
A 1200bps dial-up modem can be connected to a tap device and used to access a computer's ports running at higher speeds of 9600bps. Without the flow control feature, buffer overflow would occur at the 1200 bps end of the link. Allows for flow control,
Once equipped with a host O/S that supports XON/XOFF character processing, the tap device at the slower end monitors the input buffer of its network and sends XOFF characters to prevent overflow. It becomes like this. When the buffer is empty, an XON character is sent to the host O/S.
Data transmission from resumes. Incidentally, the physical layer of the present system provides station-to-station data transport means and synchronization signals necessary for proper operation of the TDMA communication method. Figure 2 shows the physical form of this system. In the figure, 11 is a sync generator that generates an NTSC sync signal, 12 is a terminator, 13 is a coaxial cable,
Reference numeral 14 indicates a cable insertion device (inserter), and 15 indicates a tap device. A coaxial cable 1 providing the data transmission medium for the system has a synchronous generator 11 at one end and a terminator 12 at the other end, configured as a data bus and providing access to a coaxial cable 13. Cable insertion device 14 is installed in-line with the cable to minimize impedance mismatch and subsequent reverberations. Data and synchronization are then combined on coaxial cable 13 according to a ternary signal scheme. FIG. 3 is a signal waveform diagram for explaining the above three-way signal mechanism. Referring to the figure, the synchronization signal 21 is approximately
3V positive TTL (Transistor−Transistor Logic)
The binary data 22 is -2V and 0
volts (ground), in which a low logic value is represented by -2V and a high logic value is represented by 0V.
It is expressed as The synchronous generator 11 used is a standard feeding cable driver capable of special ternary states.
It is a CATV device. That is, when two synchronization signals (vertical and horizontal) appear, drive the net,
When one of the synchronization signals is not present, the driver is in a ternary state and each cable insertion device 1
4 to drive data onto the cable. Terminator 12 terminated at the opposite end of the cable from sync generator 11 provides a low logic -2V level. Cable insertion drives the cable to ground (0V) to assert high logic. Vertical and horizontal sync pulses extracted from the cable provide frame and slot sync to the individual tap devices, respectively. Vertical sync pulses (occurring every 16.67 ms) define a frame of 262.5 TDM slots. Horizontal sync pulse (63.4μs
each occurrence) defines an individual slot. FIG. 4 is a waveform diagram showing the above TDM frame and slot format, in which reference numeral 23 is a frame synchronization signal defined by a vertical synchronization pulse;
In addition, 24 indicates a slot synchronization signal defined by a horizontal synchronization pulse, and there are 8 sets of 30 slots per frame, and 2 slots that are not used.
6, and two half-slots or full slots 25, for a total of 242.5 slots, of which 8 sets of 30 slots are assigned to each of the 30 stations in the system, and the other 2.5 Slots are left unused. The vertical and horizontal sync pulses are separated by a set of retriggerable one shots. -1V
A high impedance differential receiver with a decision limit of shall be designed so that the cable will not be rendered unusable in the event of a power failure. The outputs of optically separated receivers that receive data from tap devices that are transmitted across the bus are combined in a wire-OR, again taking advantage of the fact that only one station can transmit at a time. connected to. Thus, each station receives 20 bits per second.
This means that a total of 480 bit packets can be transmitted. This 20-bit packet is 500
Sent at a bit rate of kilobits per second or 40 μs per packet. Frame and slot synchronization are centrally processed, but bit synchronization within a packet is asynchronous on a tap device basis. Note that additional circuitry for generating the bus master reset signal is also part of the cable driver. A master reset is detected by all stations on the network whenever the frame sync signal extends beyond the full frame time. bus·
Master reset is not part of normal network operation and is used only as a means of recovery from major system failures. Referring again to FIG. 2, most of the physical layer hardware is contained within cable insertion device 14. The cable entry is designed to provide cable access from one to four tap devices 15, each of which has an RS-422 The optically isolated data lines 16 interface to the cable insertion, with four of the five data lines, designated 3 in FIG. The other one is directed to the tap device 15.
5 to a cable insertion device 14. Since only one station can transmit at a time, one cable driver is sufficient for all four tap devices connected to cable insertion device 14. Similarly, only one set of receiver, synchronization extractor and reset detection circuit is required. However, since this system uses a terminated coaxial cable bus as shown in FIG. 2, cable branching is not possible. Therefore, the system cables must be routed so that they pass in reasonable proximity to all possible sites of use. However, since the tap device 15 may be separated from the cable insertion device 14 by a considerable distance, the RS-422 interface between the tap device 15 and the cable insertion device 14 is not flexible enough. be. The total length of the coaxial cable 13 from one end to the other is limited by propagation delay. Since the system is centrally synchronized, it is important that the time references for each tap device 15 on the network are not unduly skewed. In this system, a deviation of up to 2 μs is the limit allowed from end to end. This gives a maximum cable length of 400 meters (1300 feet). Next, the data link layer and higher level procedural functions will be described. The procedural functions are implemented by some special purpose hardware and a microprocessor that make up the tap device, and the data link procedures are based on packet formats and procedures that ensure orderly link management. FIG. 5 is a diagram showing the bit structure of a packet in this system. Referring to the figure, a packet 31 has three fixed subfields and has a constant length. The packets 31 all contain 8-bit data bytes 32 with one or two zeros;
Contains two control bits 34 and addressing or auxiliary control information 33 in the utility field (provided that zero or one data byte is sent, otherwise the utility field is data bytes). Note that the code 35 is the start bit, 3
6 indicates a stop bit. The contents of one (or both) data bytes 32 are not interpreted at the data link level. The two control bits 34 are decoded to ascertain the contents of the remainder of the packet as described in Table 1. When both control bits 34 are cleared (00), the link is in the disconnected/idle state. Utility field is null system address
FF16, but this is never a legal address for any tap device. In this state, the tap device can receive or originate calls from other tap devices. When a link exists between two tap devices, control bits 34 will be (11), (01), (10) depending on the number of data bytes 32 (0, 1, 2) in packet 31. changes between states. When control bit 34 indicates zero or one data byte (11 or 01), the address of the receiving station is sent in the utility field and another station that may attempt to connect to one of the obstructed stations is sent. station can see both stations involved in the exchange. For example, if stations 5 and 19 are connected to each other, station 5 will send a "19" in its utility field and station 19 will send a "5" in its utility field. When one of the two connected stations resets control bit 34 to (0,0), the connected link is severed. The station that first sets up a disconnect immediately enters the disconnect in progress state, while the other station waits until it has received three consecutive sets of (0,0) control bits before disconnecting. This reduces spurious cuts due to noise-infringed control bits.

[Table] Although Figure 6 is a state transition diagram for explaining the procedure for connecting and disconnecting links in this system, it is one of many possible alternative mechanisms for illustrating the procedure. It is not limited to this. The link control procedure in this system will be explained below with reference to the same figure. The software for the tap device takes interrupt service into consideration, and consists of 2.1 kilobytes of code written in, for example, assembly language. The many arrows in the diagram indicate control flow paths for receiving and originating calls from other tap devices, performing listen-only (spy) functions, disconnecting calls, and starting calls from one tap device to another. always starts from the rest or idle state 41 of the tap device. First, by typing "L" the tap device enters the LOSON state 42 and the user is asked for the address of the tap device to be paged. If a valid destination code is typed, the tap device will
The LINKUP-PENDING state is entered, but otherwise the tap device performs a restart procedure.
LINKUP-PENDING state 43 is the same as all pending states, after the vertical synchronization signal is generated.
LINKUP status becomes 44. During LINKUP, a connection request packet is assembled and transmitted. After that, the tap device enters the WAIACK-PENDING state 45, and after the occurrence of the vertical synchronization signal, the tap device enters the WAIACK-PENDING state 45 and waits until the WAIACK
Enter state 46. In the WAIACK state 46, the tap device waits for confirmation of a connection request from the station with which it wishes to communicate. A connection request has three consequences. The first is that the called station is offline, the second is that the called station is occupied, and the third is that it is successfully connected. When a station is offline, no data is received in its slots, and another vertical sync signal is eventually generated to indicate the end of frame. Therefore, when a vertical synchronization signal is received during WAIACK state 46,
The tap device concludes that the called station is offline, prints a message to that effect, and restarts. When the called station receives data, it means either the connection was successful or the station is communicating with another tap device. The first data received during WAIACK state 46 causes the tap device to
Place in CNTPCT state 47. In this state, the combination of control bits and utility field address contents determines whether the connection attempt was successful (control bits (1, 1), the utility field contains the address of the originating station) or whether the called station Is it blocked?
Or it is decrypted to see if there is a response. If the called station is occupied, the source tap device indicates to whom the occupied station is connected by writing a utility field containing the station address and a control bit indicating zero or one data byte. Wait until confirmed as indicated by. If the connection is considered successful, the tap device will
The CONNECT state 48 is entered and data can be sent and received from other stations. A control bit error causes the tap device to
Note that BRKCNT state 49 is entered. 3 connected errors in control bits 4
9, 50, 51 are link disconnection and restart 52
occurs. BREAK at either end of the link
Receipt of the sequence also immediately causes the link to be disconnected and restarted. Incoming calls from other tap devices can only be accepted when the tap device is in the IDLE state 41. In the IDLE state, hardware address search logic is enabled and each packet transmitted on the net is searched for a station address match. As soon as an address match is received, the MATCH state 53 is entered, no further matches are possible, and the identity of the calling station is verified by latching the slot number where the match occurred. The called station then sends SEND−ACK−PENDING
State 54 is entered in preparation for verifying an incoming call by echoing the address of the calling station. Following the vertical sync signal, SEND−
The ACK state 55 is entered, an address echo is sent, and immediately thereafter the CONNECT state 48 begins, allowing the calling station to exchange data with it. The time it takes to perform the entire connection sequence can be as short as 1, depending on when the request for the vertical synchronization signal is made and the outcome of the connection attempt.
The frame time (16.7 ms) varies as long as 3 frames. The implementation of this procedure is fairly evenly distributed between special 20-bit receiver/transmitter hardware and microprocessor software. Due to the simplified procedure, the software complexity is also kept to a minimum (approximately 2 kilobytes of code is required). Next, higher level functions will be explained. The most important of the higher-level functions are user/host handlers, collectively called TT handlers.
It is computer interface software. As shown in FIG. 1, input and output from the tap device 5 occurs via an RS-232 serial line 6, and commands to the tap device and responses from the tap device are
Sent into ASCII. As a result, a suite of software is required to put addresses and commands into a format suitable for controlling the tap device. Similarly, the results of a connection attempt, notification of an incoming call, rejection of an invalid command, or destination address must be communicated to the user or host computer in a convenient format. For example, if a user wants to connect from a CRT terminal to a time-scheduling computer port, the user must first press the BREAK key on his terminal to cause the tap device to perform a restart, and the terminal displays the following START button. Print out the message. TERRANET V4.0; THIS IS STATION
“N” ENTER COMMAND (H FOR HELP)≫ Just replace N with the numerical address of the user's tap device on the network. User then “1”
(for logon), the tap device returns the following answer: ENTER LOGON CODE≫ Here, the user enters the computer and
Enter the station address of the port. The tap device rejects non-numeric entries or entries that are out of range. If the address entered by the user is considered valid, the tap device will attempt to connect to the desired resource. At this point, one of the following three images will be printed depending on the result of the connection attempt.
If no data of any kind is detected in the transmission slot of the called station,
The tap device will print SORRY, STATAION IS OFFLINE. If the called station is busy (already connected to another station M), the response will be SORRY, BUSY WITH STATION “M”. The tap device's response when the third connection attempt is successful is STATION CONNECTED…. A data transparent link connects the user's terminal and computer resources.
All characters received at either end are sent across the net unchanged. The only means to terminate the link (either at the called or originating station) is to send a BREAK sequence across the RS-232 link to either tap device. on the RS-232 line
Since the BREAK condition is not an ASCII character, data transparency is guaranteed. A link break message is printed at the station opposite the one where BREAK was initialized, and a start up message is printed at both ends as the tap device resets. The TT handler also performs data buffering from the RS-232 line to the net and from the net to the RS-232. Each ring buffer is 256 bytes long. A total of four ring buffers are involved in the connection between two tap devices. Other higher level functions, such as flow control characteristics, are realized by the host computer O/S.
Supports XON/XOFF character processing, slow speed
from that net to a tap device with an RS-232 link.
It is based on the assumption that if the RS-232 buffer becomes more than half full, it will be possible to block the flow of data from the host. Finally, regarding error control, in many network systems error detection is a function of the first two layers of the procedure. However, the system provides no error detection or correction (other than rejecting packets containing contextually incorrect control bits). Instead, the responsibility for error control rests with the host O/S software. Data transfers are not performed on a "best effort" basis and collisions cannot occur in such situations. Each destination circuit provided by the system provides a reliable data communication path from location to location with predictable throughput and delay characteristics that are independent of other traffic on the network. As mentioned above, this system has the highest speed per second.
Data is transferred from end to end at a rate of 9600 bits with an average delay of about 1.0 ms. Preliminary testing of the system in terms of bit error rate measured an error rate of less than 1 in ¹⁰⁹ bits. As described above, the system according to the present invention provides an interface to a host computer when potential hardware and software compatibility exists, such as through a conventional RS-232C asynchronous serial interface. This solves the conventional problem and allows access to any computer system connected to the system from a large number of remote user terminals. The system also uses the existing host operating system software for terminal handling, and the network is connected to the host computer using the old terminal interface. In operation, it provides a complete dual nominal circuit from the terminal to the host computer at transmission speeds of 300 to 9600 bps, and this circuit is completely data transparent. On the other hand, the system allows medium speed machine-to-machine communication via serial terminal ports and the execution of appropriate application programs on the communicating machines. This facilitates the use of computer-based data acquisition and workstations with larger computing facilities, as well as the exchange of files between incompatible operating systems. Further, a large number of systems can be connected by means of a gateway tap device that enables interconnection of the systems. (7) Effects of the Invention As explained in detail above, the network system according to the present invention does not require any special control program for computers or input/output devices when connecting terminals, and can be easily connected using an RS-232C connector. Because the tap device, terminator, synchronous generator, and cable insertion device that make up the system can be made small and lightweight, both the network hardware and the software necessary to support its use are simplified. It is highly effective in providing a simple and inexpensive local network system, which can reduce the cost per tap device. Furthermore, by using a specific slot of a plurality of consecutive frames as one time slot, a relatively large amount of data can be transmitted at one time.

[Brief explanation of drawings]

Figure 1 is a structural diagram showing the layers and communications of this system, Figure 2 is a physical configuration diagram of this system, Figure 3 is a synchronization signal waveform diagram, and Figure 4 is a waveform diagram showing TDM frame and slot formats. , FIG. 5 is a bit configuration diagram of a packet, and FIG. 6 is a state transition diagram in this system. 1, 13... Coaxial cable, 2, 14... Cable insertion device, 3, 4... RS-422 transceiver cable, 5, 15... Tap device, 11...
Sync generator, 12...terminator, 16...data line, 21...sync pulse, 22...binary data, 23...frame synchronization signal, 24...slot synchronization signal.

Claims (1)

[Claims] 1. Frame synchronization is performed by a vertical synchronization signal that constitutes an NTSC synchronization signal, slot synchronization is performed by a horizontal synchronization signal, and correspondence between consecutive frames is performed using one or more slots of one frame as a unit. In the interconnection method between a terminal device and a processing device or between processing devices using a time division multiplexing method in which slots for processing are assigned as one time slot, the time slot is composed of a utility data field and a control bit field, and 1. An interconnection method between terminals and processing devices or between processing devices, characterized in that connection control is performed between each terminal device and the processing devices or between the processing devices based on bit data of a control bit field. 2 Multiple types of bit data in the control bit field can be set, and connection control between the terminal and the processing device or between the processing devices is performed based on the content of the bit data and the output timing of the vertical synchronization signal. A method for mutual communication between a terminal device and a processing device or between processing devices as claimed in claim 1, characterized in that the method is characterized in that: 3 Connect the sync generator that generates the NTSC sync signal at one end and the terminator at the other end to connect the coaxial cable to one or more tap devices in a coaxial cable that supplies the baseband channel. It consists of a configuration in which a cable insertion device for providing cable access is installed, and one or more of the tap devices connected to the terminal or processing device are connected to the cable insertion device, and the NTSC synchronization generated by the synchronization generator is connected to the cable insertion device. Frame synchronization is performed using the vertical synchronization signal included in the signal, and slot synchronization is performed using the horizontal synchronization signal.
With one or more slots of a frame as a unit, corresponding slots of consecutive frames are assigned as one time slot, and the time slot consists of a utility data field and a control bit field, and the control bit field An interconnection system between terminals and processing devices or between processing devices using a time division multiplexing method that controls connections between each terminal and processing devices or between each processing device based on bit data.