Data Communications Glossary & Technical Reference

Please select the letter of the glossary category you wish to search for A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, R, S, T, U, V, W, X.

Click here to go back to my Technical Page

Technical Reference Section

Cable Distance Guide

Code Conversion Chart

Data Circuit Troubleshooting Guide

Fibre Optic Technology

Interfaces

Local Area Networks (LAN)

Modem Selection Guide

Modular Jacks and Line Cords

PC Video Standards

Protocols

A

ACK (Acknowledgement) - A communication control character transmitted by a receiver indicating that the message was received correctly.

Active Device - In current loop applications, a device capable of supplying the current for the loop; active devices generate or regenerate data signals.

Accoustic Coupler - A device that enables data to be transmitted over the telephone network via a telephone handset.

ADCCP - Advanced Data Communications Control Procedures. A link level protocol similar to HDLC.

Address - A unique sequence of bits, a character, or a group of characters that identifies a network station user, or application; a unique identification for the destination of data; used mainly for polling purposes.

Alternate Routing - An alternative communications path used if the normal one is not available. There may be one or more possible alternative paths.

Analog Data - Data in the form of continuously variable physical quantities (compare with Digital Data).

Analog Transmission - Transmission of a continuously variable signal as opposed to a discreetly variable signal. The normal way of transmitting a telephone, or voice. signal has been analog, but now digital encoding using PCM is coming into use.

ANSI (American National Standards Intstitute) - The principal standards development organisation in the USA.

API (Application Program Interface) - A set of formalised software calls and routines that can be referenced by an application program to access underlying network services.

APPC (Advanced Peer-to-Peer Communications) - Also called Logical Unit 6.2 (LU 6.2); a network node definition by IBM, featuring high level program interfacing capabilities on a peer-to-peer basis.

Appletalk - A proprietary computer networking standard promulgated by Apple Computer used for connecting macintosh computers and other peripherals, particularly Laser-Writer printers. Operates at 230Kbps.

Application Layer - The highest of the seven-layer OSI model structure. Contains all user or application programs.

Application Software - Programs that perform useful functions in the processing or manipulation of data. Includes database managers, word processors, spreadsheets, and other programs that enable the useful manipulation of data.

ARCNET - Attached Resource Computer Network. The Datapoint 2.5Mbps local area network, one of the first and still one of the most popular networks.

ARP (Address Resolution Protocol) - A Transmission Control Protocol/lnternet Protocol (TCP/IP) process that maps IP addresses to Ethemet addresses. Required by TCP/IP for use with Ethernet.

ARQ: (Automatic Request for Repetition) - A system employing an error detecting code and so conceived that any false signal initiates a repetition of the transmission of the character incorrectly received.

ASCII (American Standand Code for Information Interchange) - An 8-level code accepted as a communication standard in North America to acheive compatibility between data services. ASCII uses 7 binary bits for information and the 8th bit for parity purposes.

Asynchronous Transmission - Transmission in which the time intervals between transmitted characters may be of unequal length. Transmission is controlled by the start and stop bits at the beginning and end of each character. Also referred to as start-stop transmission.

Attenuation - Decrease in magnitude of current, voltage, or power of a signal in transmission between points. May be expressed in decibels.

Audio Frequencies - Frequencies that can be heard by the human ear.

Automatic Calling Unit (ACU) - A dialling device which permits a business machine to automatically dial calls over the public switched telephone network.

Auto-Configure - The ability of a device to configure itself to the link by examining the data on the link.

Return to the top of the page.

B

Backbone Network - A transmission facility designed to interconnect low-speed distribution channels or clusters of dispersed user devices.

Balun (BALanced/UNbalanced) - An impedance matching device used to connect balanced twisted pair cabling with unbalanced coaxial or other cabling systems.

Band Splitter - A multiplexer designed to split a line's available bandwidth into independent narrower band sub-channels.

Bandwidth - The range of frequencies available for signalling. The difference expressed in cycles per second (hertz) between the highest and lowest frequencies of a band.

Baseband - A signal frequency below the point at which the signal is modulated as an analog carrier frequency. In modulation, the frequency band occupied by the aggregate of the transmitted signals when first used to modulate the carrier.

Baseband Modem - A line driver or short-haul modem.

Batch Processing - A data processing technique in which input data is accumulated offline and processed in batches.

Baud - A unit of signalling speed. The speed in baud is the number of line changes (in frequency, amplitude, etc) or events per second. At low speeds, each event represents only one bit condition and baud rate equals bps. As speed increases, each event represents more than one bit (e.g. dibit), and baud rate does not tnuly equal bps. In common usage, "baud rate" and "bps" are often used interchangeably.

Baudot Code - A code for the transmission of data in which 5 bits represent one character. This code is used most often in tele-type-writers.

BEL - A control character which is used to call for attention, and often sounds a buzzer or bell on the receiving device.

Bell - American AT&T standards for devices that transmit over telephone lines.

BERT - Bit Error Rate Test.

BCC (Block Check Character) - The result of a transmission verification algorithm accumulated over a transmission block, and normally appended at the end. (e.g. CRC, LRC).

BCD (Binary Coded Decimal) - A coding scheme in which each decimal digit or character is represented by a series of binary digits (bits). Four bits are required to represent numerics and six bits are required for alphanumerics.

Bisynchronous Transmission (Bisync) - A data link control procedure developed by IBM using character synchronisation. Also see COP.

Bit - A contraction of the term binary digit. A bit can be either 0 or 1 and is the smallest possible unit of information making up a character or word in digital code.

Bit Duration - The time it takes one encoded bit to pass a point on the transmission medium. In serial communications, a relative unit of time measurement, used for comparison of delay times (e.g. propagation delay, access latency) where the data rate of a transmission channel can vary.

Bit Error Rate / Block Error Rate Testing (BERT/BLERT)- An error checking technique that compares a received data pattern with a known transmitted data pattern to determine transmission line quality.

Bit Order - The order of transmission of the bits in a byte or data character. Typically LSB first.

Bit Oriented - Used to describe communications protocols in which control information may be coded in fields as a single bit.

BPS (Bits per Second) - The basic unit of measure for serial data transmission capacity. Kbps for kilo (thousands). Mbps for mega (millions). Gbps for giga (billions).

Bit Sense - Nominal or inverted corresponding to assignment of a mark or space to a particular voltage level.

Bit Shift - The ability to inspect data characters formed with the first bit in different positions in the byte.

BPS - Bits Per Second. See Bit Rate.

Break-Out Box - A testing device that permits the user to switch, cross, and tie interface leads, and often contains LED's for monitoring leads.

Broad band - Communication channel having a bandwidth greater than a voice grade channel, and therefore capable of high-speed data transmission.

BSC - See Bisynchronous Transmission.

BOP - Bit Oriented Protocol (e.g. SDLC, HDLC, X.25)

Breakout Box (BOB) - A testing device that permits the user to monitor the status of the various signals between two communicating devices, and to cross and tie interface leads using jumper wires.

Bridge - A device that connects different local area networks at the data-link layer.

Broadband - Describes transmission equipment and media that can support a wide range of electromagnetic frequencies. Typically the technology of CATV transmission, as applied to data communications, that employs coaxial cable as the transmission medium and radio frequency carrier signals in the 50 to 500 MHz range.

Broadcast - A method of transmitting messages to two or more stations at the same time.

Buffer - A storage device used to compensate for a difference in rate of data flow, or time of occurence of events, when transmitting data from one device to another.

Bus - A data path shared by many devices with one or more conductors for transmitting signals, data, or power.

Byte - A unit of information, used mainly in referring to data transfer, semiconductor capacity, and data storage. Also referred to as a character, made up of a group of eight bits.

Byte-Oriented - Similar to Bit-Oriented. Control information may be coded in fields of one byte (character) length.

Return to the top of the page.

C

CAN (Cancel) - A character indicating that the data preceeding it is in enor and should be ignored.

Carrier - A high frequency signal which is modulated to carry information over long distances through space or via cable.

Carrier Detect (CD, DCD) - An RS-232/V.24 modem signal that indicates to an attached terminal that the modem is receiving a signal from the remote modem. Same as Received Line Signal Detect.

Carrier System - A means of obtaining a number of channels over a single path by modulating each channel on a different carrier frequency and demodulating at the receiving point to restore signals to their original form.

CCITT - The International Telegraph and Telephone Consultative Committee. An international organisation concerned with devising and proposing recommendations for international telecommunications. Now referred to as ITU.

Character - Letter, figure, number, punctuation or other sign contained in a message.

Character Oriented - Describes a communications protocol or a transmission procedure that carries control information encoded in fields of one or more bytes.

Checksum - The total of a group of data items or a segment of data that is used for error checking purposes. Both numeric and alpha fields can be used in calculating a checksum, since the binary content of the word can be added. Checksums can detect single-bin errors and some multiple-bit errors.

Circuit - A means of both-way communication between two points, comprising associated send and receive channels.

Circuit, 4-Wire - A communication path in which 4 wires (2 in each direction) are utilised.

Circuit, 2-Wire - A communication path using 2 wires, and may be used for one way, half duplex, or full duplex communications.

Clamping Voltages - The "sustained" voltage held by a clamp circuit at a desired level.

Clear To Send (CTS) - Modem interface signal that indicates to the temminal that it may begin transmission.

Clock - A timing device used to indicate intervals of time. A clock circuit may be used in computers and data sets to synchronise and co-ordinate the manipulation and transmission of data.

Cluster - A group of terminals or other devices in a single location. Devices are sometimes connected to a "Cluster Controller". Also see Controller.

Coaxial Cable - A popular transmission medium usually consisting of one central wire conductor (two for twinaxial cable) surrounded by a dielectric insulator and encased in either a wire mesh or an extruded metal sheathing.

Common Carrier - An organisation that provides communications transmission facilities to the public. In Australia, these organisations are Telstra and Optus.

Communications Protocol - The rules governing the exchange of information between devices on a data link (e.g. SDLC, LAT, TCP/IP, X.25).

Communications Server - An intelligent device providing communications functions. An intelligent, specially configured node on a local area network designed to enable remote communications access for LAN users.

Composite Link - A line or circuit connecting a pair of multiplexers, which carries the multiplexed data.

Compression - Any of several techniques that reduce the number of bits required to represent information in data transmission or storage (thus conserving bandwidth or memory), permitting the original form of the information to be reconstructed. Also called "compaction".

Concentrator - Any communications device that allows a shred transmission medium to accomodate more data sources than there are channels currently available within the transmission medium. See also Statistical Multiplexer.

Conditioning - The addition of equipment to a leased voice-grade channel to provide minimum values of line characteristics required for data transmission.

Console - The device used by a computer operator to monitor and control system performance and operation.

Contention - A method of line control in which the terminals request to transmit, and if the channel is free, transmission goes ahead. If it is not free, the terminal waits until it becomes free.

Control Character - A character whose occurrence indicates a control function in the associated business machine or computer. Control characters are used for line control operations, formatting, terminal and device control or as information separators, like CR (carriage return), VT (vertical tab) and EOT (end of transmission).

Controller - A type of communication control unit that manages the details of line control and routing of data through a network.

COP (Character-Oriented Protocol) - A protocol that uses control characters to perform the functions of data link control.

CPU (Central Processing Unit) - The section of a computer that directs the sequence of operations and initiates the proper commands to the computer for execution.

CR (Carriage Return) - An ASCII or EBCDIC control character that moves a device's cursor or print-head to the left margin.

CRC (Cyclic Redundancy Check) - A checking method in which the numeric binary value of a block or frame of data is divided by a constant divisor. The quotient is discarded and the remainder serves as the check sequence. The check sequence is a series of bits added to the end of a frame or block used to detect errors.

Cross Talk - The unwanted transfer of energy from one circuit to another circuit.

Crossover Cable - A data cable designed to link two DTE devices together by directing the transmitted signals from one to the receive lines in the other.

CRT (Cathode Ray Tube) - A picture tube similar to that used in a television. A term often used to describe a display terminal.

CSMA/CD (Carrier Sense Multiple Access / Collision Detection) - A local area network access method in which contention between two or more stations is resolved by collision detection. When two stations transmit at the same time, they both stop and signal that a collision has occurred. Each then tries again after waiting a predetermined period of time, usually several microseconds.

CTS (Clear To Send) - An RS-232 modem interface control signal used in a switched half-duplex circuit to notify the computer that it has line control. In a full-duplex circuit, this signal is constant.

Current Loop - A method of interconnecting terminals and transmitting signals whereby a "mark" is represented by current on the line and a "space" represented by the abscence of current.

Return to the top of the page.

D

Data Code - A set of symbols or rules for use in representing information. (e.g. ASCII, EBCDIC, Baudot).

Datagram - A finite-length packet with sufficient information to be independently routed from source to destination. Datagram transmission typically does not involve establishing an end-to-end session, and may or may not entail confirmation/acknowledgment of delivery.

Data Integrity - A term used to describe the amount of data being passed over a link, without errors.

Data Link - Any serial data communications transmission path, generally between two adjacent devices or nodes, and without any intermediate switching devices.

Data Link Layer - Second layer in the OSI model. The network processing entity that establishes, maintains and releases data-link connections between (adjacent) elements in a network. Controls access to the physical medium (Layer 1).

Data Set - See Modem or DCE.

Data Set Ready (DSR) - A modem interface control signal which indicates to the DTE that the modem is ready for transmission.

Data Transfer Rate - The average number of bits, blocks or characters per unit of time transferred from a data source to a destination device.

Data Terminal Ready (DTR) - A modem interface signal which indicates to the modem that the DTE device is ready for transmission.

DC (Device Control) - A type of control character primarily intended for controlling data flow from an attached device. (e.g. DC1, DC2, DC3 etc).

DCE (Data Communications Equipment) - The equipment installed at the user's premises, which provides all the functions required to establish, maintain, and terminate a connection.

DDS (Digital Data Service) - A data service offered by Telstra in Australia designed for the transmission of synchronous digital data over leased line connections.

DDCMP - Digital Data Communication Message Protocol. A uniform discipline for the transmission of data between stations in a point-to-point or multipoint, synchronous or asynchronous network.

DECNET - Digital Equipment Corporation's proprietary network architecture that works across all of the company's machines.

Dedicated Line - A dedicated circuit, a non-switched channel. Also called a private or leased line.

Demodulation - The process of retrieving data from a modulated carrier wave. The reverse of modulation.

Delay - In communications, the time between two events. Also see Propagation Delay, Response Time.

DES (Data Encryption Standard) - A scheme approved by the National Bureau of Standards that encrypts data for security purposes. DES is the data communications encryption standard specified by most government agencies and financial institutions.

Destination Field - A field in a message header that contains the address of the station to which the message is being sent.

Dial Network - A term that refers to the public switched telephone network (PSTN).

Dial-Up - The use of a dial or pushbutton telephone to initiate a station to station telephone call.

Dibit - A group of two bits. The four possible states for a dibit are 00, 01,10 ,11.

Differential Phase Shift Keying (DPSK) - A modulation technique, as used in ITU V.22 modems.

Digital Data - Information represented by a code consisting of a sequence of discrete elements.

DIP Switch - A switch used to open and close leads between two devices. This type of switch is often mounted directly on the circuit board of a device, and combinations of the switches are used for configuration.

Distortion - The unwanted change in waveform that occurs between two points in a transmission system.

DMA (Direct Memory Access) - A method of moving data from a storage device to RAM.

DOS (Disk Operating System) - A set of programs that instruct a disk-based computing system to manage resources and operate related equipment.

Downline Loading - The process of sending configuration parameters, operating software or related data from a central source to a remote device.

Downtime - The period during which computer or network resources are unavailable to users because of a failure.

Driver - A software module that, under control of the processor, manages an l/O port to an external device, such as a serial RS-232 port to a modem.

Drop Cable - In local area networks, a cable that connects the main network cable and the data terminal equipment (DTE).

DSE - Data Switching Exchange. The set of equipment installed at a single location to switch data traffic.

DSR - See Data Set Ready.

DTE (Data Terminal Equipment) - A device which acts as a data source (e.g. a terminal or computer system).

DTMF (Dual-Tone Multiple Frequency) - The audio signaling frequency on touchtone, pushbutton telephones.

DTR - See Data Terminal Ready.

Dumb Terminal - Generally describes an asynchronous terminal with no local processing capacity.

Duplex Transmission - See Full Duplex.

Return to the top of the page.

E

EBCDIC - Extended Binary Coded Decimal Interchange Code. An 8 bit coding structure used to represent data in certain data processing and communication terminal devices. Enables the representation of all graphics and control codes in a logic format.

Echoplex - A method of checking data transmission accuracy whereby the received data is returned to the sending end for comparison with the original data.

EIA (Electronic Industries Association) - A consultative group of manufacturers recognised as the standards writing group in the United States for electronic equipment.

EISA (Extended Industry Standard Architecture) Bus - A 32-bit adaption of the 8 and 16-bit buses originally developed by IBM. It is starting to be used in high-end PCs and file servers with 80386 and 80486 processors. The EISA bus was a joint development of Compaq and other PC manufacturers as an alternative to the proprietary IBM Micro Channel Bus.

EMI/RFI (Electromagnetic Interference/Radio Frequency Interference) - Unwanted electromagnetic emissions, generated by lightning or by electronic or electrical devices, that degrade the performance of another electronic devices. Interference may be reduced by shielding.

Emulate - Using one system to imitate another system. The emulating system executes the same programs, uses the same data, and produces the same results as the original system.

Encoding / Decoding - The process of organising information into a format suitable for transmission, and then reconverting it after transmission.

ENQ (Enquiry) - A transmission control character used as a request for a response from a remote station.

EOT (End Of Transmission) - A transmission control character used to indicate the conclusion of transmission.

EPROM (Electrically Programmable Read Only Memory) - Read only, non-volatile semiconductor memory that is erasable under ultra-violet light and reprogrammable.

ESC (Escape) - A control character used to provide additional control functions to characters which are appended to it.

Ethernet - A popular local area network design and the product of Xerox Corp., characterised by 10Mbps baseband transmission over shielded coaxial cable and employing CSMA/CD as the access control mechanism. Standarised by the IEEE as specification IEEE 802.3

ETB (End of Transmission Block) - A transmission control character used to indicate the end of a transmission block of data.

ETX (End of Text) - A transmission control character that indicates the end of text transmission.

Error Correction - A system employing an error correcting code and so arranged that some or all signals detected as being in error are automatically corrected at the receiving terminal before delivery.

Equaliser - An electronic circuit designed to improve the frequency response of an analog line.

Even Parity - A "dumb" terminal data verification method in which each character must have an even number of "on" bits.

Expanded Memory - A PC's address memory (any amount above 640K) that conforms to the LIM (Lotus, Intel & Microsoft) memory specification.

Extended Memory - A PC's address memory (any amount above 640K) that does not conform to LIM memory specification. May be converted to expanded memory by software drivers.

Return to the top of the page.

F

Facsimile (FAX) - Transmission of pictures, maps, diagrams, etc. The image is scanned at the transmitter, reconstructed at the receiving station, and duplicated on some form of paper or film.

FCS (Frame Check Sequence) - In bit-oriented protocols, a 16-bit field that contains transmission error-checking information, usually appended at the end of a frame.

FDDI (Fibre Distributed Data Interface) - An ANSI standard for fibre-optic links with data rates up to 100Mbps. The standard specifies multimode fibre, 50/125, 62.5/125 or 85/125 core cladding, an LED or laser light source, and 2 km for unrepeated data transmission at 40 Mbps.

FDM - See Frequency Division Multiplexer.

FEP(Front End Processor) - A dedicated computer linked to one or more host computers or multiuser minicomputers. Performs data communications functions and serves to off-load the attached computers of network processing. In IBM SNA networks, an IBM 3704, 3705, 3725 or 3745 communications controller.

Fibre Loss - The attenuation (deterioration) of the light signal in optical fibre transmission.

Fibre Optics - Transmission technology in which modulated light-wave signals, generated by a laser or LED, are propagated along a glass or plastic medium, and then demodulated to electrical signals by a light sensitive receiver.

File Server - In local area networks, a station dedicated to providing file and mass data storage services to other stations on the network.

Firmware - A computer program or software stored permanently in memory.

Flag - In communications, a bit pattern of six consecutive "1" bits (character representation is 01111110) used in many bit-oriented protocols to mark the beginning (and often the end) of a frame.

Flow Control - The procedure for regulating the flow of data between two devices. Flow control prevents the loss of data once the receiving device's buffer becomes full.

Flying Lead - A lead which exits the back of the connector hood on the outside of the cable jacket.

Frame - In bit-oriented protocols, data is formatted in frames for transmission. Each frame consists of a start flag followed by an address field, control field, data field, frame check sequence, and a stop flag.

Frequency Division Multiplexer - A device which divides the available transmission frequency range into narrower bands, each used for a separate channel.

Frequency Shift Keying - A frequency modulation technique in which one frequency represents a mark and a second frequency represents a space.

Front-End Processor - A communications computer associated with a host computer. It may perform line control, message handling, code conversion, error control, and applications functions, such as control and operation of special purpose temminals.

FSK - See Frequency Shift Keying.

FTAM (File Transfer, Access and Management) - An OSI application which provides transparent access to files stored on dissimilar systems.

FTP (File Transfer Protocol) - An upper level TCP/IP service that allows copying of files across a network.

Full Duplex - Refers to a communications system or equipment capable of transmission simultaneously in two directions.

Return to the top of the page.

G

Gain - Increased signal power, usually the result of amplification. The opposite of attenuation.

Gateway - A conceptual or logical network station that serves to interconnect two otherwise incompatible networks, network nodes, sub-networks or devices. Performs protocol conversion operation across numerous communications layers.

Ground - An electrical connection or common conductor mat, at some point, connects to the earth.

Group Addressing - In transmission, the use of an identification field that is common to two or more stations.

GUI (Graphical User Interface) - An operating system or environment which displays program choices and options on the screen as icons, or picture/symbols. Users enter commands by pointing at icons with a mouse or other pointing device and clicking a button on the mouse.

Return to the top of the page.

H

Half Duplex - A circuit which provides transmission alternately in either direction.

Handshake - Exchange of predetermined signals occurring when the connection between two data sets is first established.

Hardware - Physical equipment, as opposed to the computer program or method of use.

HDLC (High-level Data Link Control) - A link level bit-oriented protocol. Level 2 of X.25

HDX - See Half Duplex.

Header - The control information added to the beginning of a message transmission block or packet.

Hertz (Hz) - A measure of frequency or bandwidth. The same as cycles per second.

Hexadecimal Number System (Hex) - A number system with a base of 16. In Hex, the first ten digits are 0 - 9, then the last six digits are represented by the characters A - F.

Return to the top of the page.

I

ICMP (Internet Control Message Protocol) - The TCP/lP process that provides the set of functions used for network layer management and control.

Idle Character - A character that is sent, when there is no information to be sent.

IEEE - Institute of Electrical and Electronics Engineers.

Impedance - The resistance to the flow of alternating current in a circuit.

In-House - A system whose remote terminals are all situated in one building or localised area.

Intelligent Terminal - A terminal that contains a processing element. It can perform local data processing and storage activities.

Interactive Processing - Time-dependent (real-time) data communications. A user enters data and then waits for a response from the destination before continuing.

Interface - A shared boundary. The connection point between business machines and the data set, modem or communications channel.

IP (Internet Protocol) - The protocol used in gateways to connect networks at the OSI Network level (Level 3) and above.

IPARS - International Passenger Airline Reservation System. An IBM term originally, but fast becoming a generic term for any airline protocol.

ISA (Industry Standard Architecture) - A bus standard developed by IBM for expansion cards for the first IBM PC. The original bus supported a data path only 8 bits wide. IBM subsequently developed a 16-bit version for its AT class computers. The 16-bit AT ISA bus supports both 8 and 16 bit cards. The 8-bit bus is commonly called the "PC/XT bus", while the 16-bit bus is called the "AT bus".

ISDN (Integrated Services Digital Network) - An ITU recomendation to standardise operating parameters and interfaces for a network that accomodates a variety of mixed digital transmission services. Access channels include basic rate (128 Kbps) and primary rate (2.048 Mbps).

ISO - International Standards Organisation. The body which promotes the development of worldwide standards.

ITB (Intermediate Block Character) - A transmission control character that terminates an intermediate block.

Return to the top of the page.

J

Jitter - The slight movement of a transmission signal, in time or phase, that can introduce errors and loss of synchronisation in high speed synchronous transmission.

Jumper - A piece of wire which is used to tie together pins on one end of an interface cable.

Return to the top of the page.

K

Kilo (k) - 1,000 units of measure (e.g. kHz, kbps).

Kilo (K) - 1,024. Used in specifying memory size, denoted by capital "K" (e.g. Kbyte).

Return to the top of the page.

L

LAN - See Local Area Network.

LAP (Link Access Procedure) - The data-link level protocol specified in the ITU X.25 interface standard. Originally LAP, supplemented with LAPB (LAP-Balanced) and LAPD.

LAPD (Link Access Procedure-D) - Link-level protocol devised for ISDN connections, differing from LAPB in its framing sequence.

Latency - The time interval between a network station's seeking access to a transmission channel and the granting or receiving of access. Equivalent to waiting time.

Lattisnet - Manufactured by Synoptics Communications, it was the first implementation of Ethernet at full 10 Mbps over twisted pair and fibre optics in a star topology.

Layer - In the OSI reference model (seven basic layers), a collection of related network processing functions.

Lead - An individual pin or wire circuit in a physical interface.

Leased Line - A telecommunications channel leased between two or more service points in one exchange or different exchanges, usually at a monthly rate.

LED (Light Emmitting Diode) - A semiconductor light source that emits visible light or invisible infra red light.

Line Driver - A signal converter that enhances a digital signal to ensure reliable transmission over an extended distance.

Line Turnaround - The reversing of transmission direction from the sending device to the receiving device, as used in a half-duplex circuit.

Link - A transmission path beween two stations, channels, or parts of communications systems.

Link Layer - Layer 2 of the OSI reference model, also known as the data-link layer.

LLC (Logical Link Control) - A protocol developed by the IEE 802 committee for data-link layer transmission control.

Loading - Adding inductance (load coils) to a transmission line to minimise amplitude distortion.

Local Area Network - A data communications system confined to a limited geography area (up to about 10km) with moderate to high data rates (100kbps to 50Mbps). The area served may consist of a single building, a cluster of buildings or a campus. The network uses some type of switching and data addressing scheme, and does not use common carrier circuits, although it may have gateways and bridges to other networks.

Local Loop - The part of a communication circuit from the subscriber's equipment to the line terminating equipment in the exchange.

Loopback - A Loopback test is a type of diagnostic test in which data is returned to the sending device after passing through all, or a portion, of the network link. This type of test permits comparison of data received with the data that was sent.

Loss - Reduction in signal strength, expressed in decibels (dB). opposite of gain.

LRC (Longitudinal Redundancy Check) - A method of providing a level of confidence that data is being received correctly by performing a parity check across the data message.

LSB - Least Significant Bit.

LSI (Large Scale Integration) - A term used to describe a multifunction semiconductor device, such as a microprocessor, with a high density of electronic circuitry (up to 1000 circuits) on a single silicon chip.

LU 6.2 - In SNA, a set of protocols that provide peer-to-peer communications between applications.

Return to the top of the page.

M

MAC (Media Access Control) - A media-specific access control protcol within IEEE 802 specifications. Currently includes variations for Token Ring, token bus, and CSMA/CD. The lower sub-layer of the IEEE's link layer, which compliments the Logical Link Control (LLC).

Macintosh - A family of Apple microcomputers that represents the first wide-scale deployment of icons, windows, a mouse cursor device, and a consistent user interface.

Magnetic Medium - Any data storage medium and related technology, including disks, diskettes, and tapes, in which different patterns of magnetisation are used to represent bit values.

Mainframe - A large computer system which has the ability to control a number of peripheral devices, and execute programs stored in memory devices.

Manchester Encoding - Digital encoding technique (specified for the IEEE 802.3 Ethernet baseband network standard) in which each bit period is divided into two complimentary halves. A negative-to-positive (voltage) transition in the middle of the bit period designates a binary "1", while a positive-to-negative transition represents a "0". The encoding technique also allows the receiving device to recover the transmitted clock from the incoming data stream (see clocking).

MAP (Manufacturing Automation Protocol) - A General Motors originated suite of protocols, the implementation of which tracks the seven layers of the OSI model.

Mapping - In network operations, the logical association of one set of values, such as addresses on one network, with quantities or values of another set, such as devices on another network (e.g. name-address mapping, internetwork route mapping, protocol-to-protocol mapping).

Mark - In communications, an impulse on a data circuit used to signify a "1" binary condition.

Master Station - A unit having control of all other terminals on a multipoint circuit for purposes of polling and/or selection.

MAU (Multistation Access Unit) - A wiring concentrator used in Token Ring local area networks.

Medium - Any material substance used for the propagation or transmission of signals, usually in the form of electrons or modulated radio, light, or accoustic waves.

Mega - Designation for one million 1,000,000 (e.g. Mbps)

Menu - Describes the method of configuring the setup of a device using prompts or choices directing the user down a decision tree.

Micro Channel - A proprietary bus developed by IBM for its PS/2 family of computers' internal expansion cards.

Microprocessor - An electronic integrated circuit, typically a single-chip package, capable of receiving and executing coded instructions (e.g. Zilog Z80, Intel 80486, Motorola 68030)

MIPS (Million Instructions per Second) - A general comparison gauge of a computers raw processing power.

Mnemonic - A coded representation of an address, operation, or term which is intended to be easily remembered. (e.g. SNRM for Set normal Response Mode, STX for Start of Text, etc)

Modem - Contraction of the term Modulator-Demodulator. A modem is used to convert a digital signal from a computer to an analog signal, so that it may be transmitted over a network.

Modem Eliminator - A device used to connect a local terminal to a computer port, instead of using a pair of modems. Allows DTE to DTE connection, and provides the necessary control signals that would normally be generated by the DCE.

Modulation - The technique of modifying some of the characteristics of a waveform.

Monitor - Monitors are used to observe, evaluate, and/or intercept transmissions in a data communication system.

MSB - Most Significant Bit.

MTBF (Mean Time Between Failures) - The average length of time for which a system, or a component, works without failure.

MTTR (Mean Time To Repair) - When a system or component develops a fault, this is the average time taken to correct the fault.

Multidrop Line - Line or circuit interconnecting several stations. Also called a multipoint line.

Multimode - Essentially, an optical fibre designed to carry multiple signals, distinguished by frequency or phase, at the same time.

Multiple Routing - The process of sending a message to more than one recipient, usually when all destinations are specified in the header of the message.

Multiplexer - A device designed to combine several signals into a composite stream for economic transmission. Techniques employed are frequency division (FDM), time division (TDM) and statistical multiplexing.

Multipoint Line - A single communications line or circuit interconnecting several stations supporting terminals in several different locations. Use of this type of line usually requires some kind of polling mechanism, with each terminal having a unique address. Also called a multidrop line.

Multitasking - The concurrent execution of two or more tasks or applications by a computer. May also be the concurrent execution of a single program that is used by many tasks.

Return to the top of the page.

N

NAK (Negative Acknowledgement) - A communication control character transmitted by a receiver indicating that it received some information incorrectly.

Nanosecond - One billionth of a second.

NETBIOS (Network Basic Input/Output System) - Software developed by IBM, provides the interface between a PC's operating system, the I/0 bus, and the network.

Netware - A local area network operating system manufactured by Novell.

Network - A series of points interconnected by communications channels.

Network Architecture - A set of design principles, including the organisation of functions and the description of data formats and procedures, used as the basis for the design and implementation of a network.

Network Interface Controller - Electronic circuitry that connects a workstation to a network, usually a card that fits into one of the expansion slots inside a personal computer. It works with the network software and computer operating system to transmit and receive messages on the network.

Network Layer - Layer 3 in the OSI model. The logical network entity that services the transport layer, responsible for ensuring that data passed to it from the transport layer is routed and delivered through the network.

Network Management - Customer network management is the systematic approach to the planning, organising, controlling and evolving of a communications network, while optimising cost perfommance.

Network Operating System - The software used to connect devices, share resources, transfer files and perform network activities. Usually, there are two parts to a network operating system: server and workstation. The workstation part puts the PC on the network. It reroutes transmissions over the network when needed. The server part runs on a PC known as the server. The server software makes disks, software and ports available to other PCs on the network. In some networks, workstations and servers can be the same machine so all users can share some of their resources as well as access those of other users.

Network Topology - The physical and logical relationship of nodes in a network. Networks are typically of either a star, ring, tree, or bus topology, or some hybrid combination.

NFS (Network File System) - An extension of TCP/IP that allows files on remote nodes of a network to appear locally connected.

Node - The node is that point in the transmission system where lines or trunks from many sources meet. That location in a data network where switching is done, e.g. in a packet switched network such as Austpac, the node is where the equipment does the packet switching function is located.

Noise - Random electrical signals, introduced by circuit components or natural disturbances, which degrade the performance of a communication channel.

NRZ (Non Return to Zero) - Pulses in alternating directions for successive "1" bits. No change from existing bias for "0" bits.

Null Modem - A device that connects two DTE devices directly by emulating the physical connections of a DCE device.

Return to the top of the page.

O

Off Hook - A modem automatically answering on a dial-up line is said to go "Off Hook", in a similar manner to picking up the telephone handset.

Off Line - Condition in which a terminal or other device is not connected to a computer or is not actively transmitting via a network. Opposite of on-line.

On Line - A device which is actively connected to a computer or processor is said to be "On Line".

Operating System - The software of a computer that controls the execution of programs, typically handling the functions of input/output control, resource scheduling, and data management.

Optical Disk - A very high density information storage medium that uses light to read and write digital information.

OS/2 - The second generation operating system developed by IBM and Microsoft for use with the Intel 80286, 80386 and 80486 microprocessors. Unlike its predecessor (PC MS-DOS), OS/2 is a multitasking operating system.

OSI - Open System Interconnection. OSI refers to the standardised procedure for exchange of information among terminal devices, computers, people, networks, processors, etc. that are "open" to one another for this purpose by virtue of their mutual use of these procedures. OSI is itself a reference model, being developed jointly by ISO and ITU. It provides a common basis for the co-ordination of standards development for the purposes of systems interconnection and allows existing systems and standards to be placed in perspective within the overall 7-layer model shown below:

Level 7	Application
Level 6	Presentation
Level 5	Session
Level 4	Transport
Level 3	Network
Level 2	Data Link
Level 1	Physical

Overhead - In communications, all information, such as control, routing, and error checking characters that is in addition to user data. Includes information that carries network status or operational instructions, network routing information, and retransmissions of user data messages that are received in error.

Return to the top of the page.

P

PABX (Private Automated Branch Exchange) - Telephone service provided exclusively for a customer's use, consisting of exchange lines, a switchboard, and extension or "local" telephones which may be interconnected with the trunks or with each other. See also Data PABX.

Packet - Data grouped for transmission through a data network such as a X.25 network.

Packet Switching - The transfer of data by means of addressed packets whereby a channel is only occupied for the duration of transmission of the packet. The channel is then available for transfer of other packets. In contrast with circuit switching, the data network determines during, rather than prior to, the transfer of a packet.

PAD (Packet Assembler/Disassembler) - Equipment providing packet assembly and disassembly facilities.

Parallel Processing - Concurrent or simultaneous execution of two or more processes, or programs, within the same processor, as contrasted with serial or sequential processing.

Parallel Transmission - Transmission mode that sends a number of bits simultaneously over separate lines to a device such as a printer. Usually unidirectional.

Parity Bit - A bit added to characters so that the total of "1" or "0" bits in a character will always be either even (even parity) or odd (odd parity).

Parity Check - The addition of non-information bits that make up a transmission block to ensure that the total number of 1s is always either even (even parity) or odd (odd parity). Used to detect data transmission errors.

Pass-Through - Describing the ability to gain access to one network element through another.

Passive Device - In current loop applications, a device that must draw its current from connected equipment.

PC (Personal Computer) - A generic term for a single-user microcomputer. Also refers to the IBM Personal Computer, the first microcomputer to be widely accepted in business and still a standard for compatibility.

Phase Modulation - One of three ways of modifying a sine wave signal to make it "carry" information. The sine wave, or carrier, has its phase changed in accordance with the information to be transmitted.

Physical Layer - Within the OSI Model, the lowest level (Level 1) of network processing, below the link layer, concerned with the electrical, mechanical, and handshaking procedures over the interface that connects a device to a transmission medium.

Pixel - Picture element, the smallest unit of a graphics or video display.

Point-To Point - Describing a circuit that interconnects two points directly, where there are generally no intermediate processing nodes, computers, or branched circuits, although there could be switching facilities.

Polarity - Any condition in which there are two opposing voltage levels or changes, such as positive and nagative.

Polling/Selection - The process of "calling out" to remote stations from a central point on a sequential, systematic basis. The polling operation is to request or collect data. The selection operation is to distribute data.

Port - A computer interface capable of attaching to a modem or temminal.

Presentation Layer - Layer 6 of the OSI model. Provides standards for restructuring data into the required format, character set or language.

Print Server - An intelligent device used to transfer information to a series of printers.

PROM (Programmable Read Only Memory) - A nonvolatile memory chip that permanently stores information.

Propagation Delay - The time necessary for a signal to travel from one point on a circuit to another.

Protocol - A formal set of conventions governing the format and control of inputs and outputs between two communication devices or processes.

Protocol Converter - A device for translating the data transmission code of one computer or peripheral to the data transmission code of another computer or peripheral, thus enabling equipment with different data formats to communicate with each other.

Pubic Data Network - A network established and operated for the specific purpose of providing transmission services for the public (e.g. Austpac).

Public Switched Data Network (PSDN) - A network providing switched links to many users (e.g. the telephone network).

Pulse Code Modulation (PCM) - Modulation of a pulse train in accordance with a code.

Pulse Dialling - The older form of telephone dialling, sometimes known as decadic dialling, that uses breaks in DC current to indicate the number being dialled.

Pulse Trap - A device that monitors any data interface lead for changes in logic levels.

Return to the top of the page.

R

RAM (Random Access Memory) - Semiconductor read/write volatile memory. Any data stored is lost if power is removed.

Real Time - A Real Time computer system is one that controls an environment by receiving data, processing it, and returning the results sufficiently quickly to affect the functioning of the environment at that time. Similar to an On-Line System.

Redundancy - In data transmission, the portion of a message's gross information that can be eliminated without losing essential information. Also, duplicate facilities.

Redundancy Check - An automatic or programmed check based on the systematic insertion of components or characters used especially for checking purposes.

Repeater - In digital transmission, equipment that receives a pulse train, amplifies it, retimes it, and then reconstructs the signal for retransmission.

Request To Send - A modem interface signal sent from the terminal to the modem, indicating the terminal has something to transmit.

Response Time - The elapsed time between the generation of the last character of a message at the terminal and the receipt of the first character of the reply. It includes terminal and network delay, and computer processing time.

Ring Indicator (Rl) - A dial-up modem interface signal which indicates to the terminal that an incoming call is present.

Ring Topology - A network topology in which each node is connected to two adjacent nodes.

RISC (Reduced Instruction Set Computing) - Internal computing architecture in which processor instructions are pared down so the most can be performed in a single processor cycle.

ROM (Read Only Memory) - Nonvolatile semiconductor memory manufactured with predefined data content, permanently stored.

Routing - The assignment of the communications path by which a message or telephone call will reach its destination.

RS-232 - An EIA standard for interfacing DTE with DCE. A 25 pin physical interface. Equivalent to ITU V.24

RS-422 - A balanced electrical implementation of RS-449 for high speed data transmission.

RS-423 - An unbalance electrical implementation of RS-449 for RS-232C compatibility.

RS-449 - A 37 pin physical connector specification.

RTS - See Request To Send.

Return to the top of the page.

S

SBT (Six Bit Transcode) - A 6-bit data code used with older BSC networks. Alternatives are ASCII and EBCDIC.

SCSI (Small Computer Standard Interface) - A specification of mechanical, electrical, and functional standards for connecting small computers with intelligent peripherals such as hard disks and printers.

SDLC (Synchronous Data Link Control) - Used by IBM, a subset of HDLC.

Serial Data Transmission - The most common transmission mode. Information bits are sent sequentially, in serial, along a single data channel.

Session - A connection between two stations that allows them to communicate.

Session Layer - Layer 5 of the OSI model. Provides protocols for assembling physical messages unto logical messages.

Shielding - A metallic covering that eliminates electromagnetic and radio frequency interference.

Short Haul Modem - A modem suitable for use over short distances, typically up to 30km, on leased lines.

Simulate - to test a network component by interfacing with it.

Slave - A remotely located instrument which can be controlled.

SNA (Systems Network Architecture) - The IBM network architecture for communications between IBM devices. A common architecture, used by IBM, which describes the logical structure, hmmats, protocols and operational sequences for transmitting information units through a communication network.

SNMP - Simple Network Management Protocol.

Soft Key - A key whose function changes with the application software running.

Software - A computer program or set of programs held in some storage medium and loaded into read/write memory (RAM) for execution.

SOH (Start Of Header) - A transmission control character used as the first of a heading in an information message.

Space - A binary "0". Opposite of Mark "1".

Spool (Simultaneous Peripheral Operation On-Line) - A program or piece of hardware that controls data going to an output device. Also known as a "spooler".

Star Topology - A communication system consisting of one central node with point-to-point links to several other nodes.

Start Bit - In asynchronous transmission, the first bit or element in each character, normally a space, which prepares the receiving equipment for reception and registration of the character.

Start/Stop - See Asynchronous Transmission.

Statistical Multiplexer - Multiplexer equipment that dynamically allocates trunk capacity only to active input channels, which allows more devices to be connected than with a traditional time division rnultiplexer.

Stop Bit - The last element in an asynchronous transmission, used to ensure recognition of the next start bit.

Store and Forward - A method of transmission in which messages received at intermediate points are stored and then re-transmitted to the destination.

STX (Start of Text) - A transmission control character that precedes the text and is used to terminate the header.

SUB (Substitute Character) - A control character used in the place of a character that has been found to be in error or invalid).

Switched Line - A communications link for which the physical path may vary with each usage, such as the public telephone network.

SYN (Synchronous idle) - A transmission control character used by synchronous transmission sytems in the absence of any other character, to indicate a line idle condition.

Synchronous Transmission - A data network in which the timing of all components of the network is controlled by a single timing source.

Return to the top of the page.

T

Tail Circuit - A feeder circuit or an access link to a network.

TCP/IP (Transmission Control Protocol / Intemet Protocol) - A layered set of protocols that allows sharing of applications among PCs in a high speed communications environment. Because TCP/lP's protocols are standardised across all its layers, including those that provide terminal emulation and file transfer, different vendors' computing devices can exist on the same cable and communicate with one another. Corresponds to Layers 4 and 3 of the OSI reference model. See my webpage on TCP/IP from the technical page.

Telecommunications - A term encompassing both voice and data communications in the form of coded signals over media.

Teletex - A facility enabling subscribers to communicate between terminals which are used for the preparation, editing and printing of correspondence.

Teletext - A generic term used to describe one-way broadcast information services in which users select by keypad one page at a time out of a limited number of pages for display on an adapted television set.

Terminal - Any device capable of sending and/or receiving information over a communication channel.

Terminated Line - A circuit with a resistance at the far end equal to the characteristic impedance of the line, so that no reflections or standing waves are present when a signal is entered at the near end.

Termination - The placement of a connector on a cable.

Tie-Line - A privately leased communication channel, often between two adjacent buildings.

Time-Division Multiplexer (TDM) - A device that accepts multiple channels on a single transmission line by connecting terminals, one at a time, at regular intervals by interleaving bits or characters from each terminal.

TIMS - Transmission Impairment Measuring Set. A test set for troubleshooting analog transmission lines.

Transceiver - A device that can both transmit and receive.

Transmission - The despatching of a signal, message, or other form of intelligence, by wire, radio, telegraphy, telephony, facsimile, or other means.

Transient - An abrupt change in voltage, of short duration.

Transport Layer - Layer 4 in the OSI model. Provides a logical connection between processes on two machines.

Trap - A trigger event.

Trigger - An event occuring on a link, the occurrence of which causes a specific action to be taken.

Trunk - A single circuit between two points, both of which are switching centres or individual distribution points. Usually handles many channels simultaneously.

Turnaround Time - The time taken for a half-duplex line to reverse the direction of the transmit and receive signals.

Twisted Pair - Two insulated copper conductors that are wound around each other, mainly to cancel the effects of electrical noise. Typical of standard telephone wiring.

Return to the top of the page.

U

Unbalanced Line - A transmission line on which the magnitudes of the voltages on the two conductors are not equal with respect to the ground (e.g. coaxial cable).

UNIX - Operating system originally designed by AT&T for communicating, multiuser, 32-bit computer. UNIX has come into wide commercial acceptance because of its predominance in academic institutions and its programming versatility.

Unloaded Line - A line that has no loading coils.

Return to the top of the page.

V

V.24 - An ITU recommendation that specifies the interchange circuits between DTE and DCE. Similar to EIA RS-232C.

V.35 - An ITU recommendation for high speed modem interfaces using 60 -108 kHz group band circuits.

Videotex - Applies to accessing of information stored in one or several computers via the telephone network by terminals and television sets equipped with interface units (e.g. Telstr'a Viatel service).

Virtual Call - A user communications facility in which a call setup procedure and a call clearing procedure will determine a period of communication between two DTE's, in which user data will be transferred across the network in packet mode.

Virtual Circuit - A network recognised association between two DTE's which enables a DTE to carry out the transfer or exchange of strings of messages with a second DTE according to a standard communication procedure with network preservation of sequence order. A virtual circuit requires network bandwidth only when data is actually transmitted.

Virtual Memory - A technique for using disk storage space to emulate random access memory (RAM).

Virtual Storage - Storage space that may be viewed as addressable main storage, but is actually auxilliary storage mapped into real addresses.

Voice Frequency (VF) - Any frequency within that part of the audio frequency range essential for the transmission of speech of commercial quality. Normally 300 - 3400 Hz.

Voice-Grade Line - A channel that is capable of carrying voice-frequency signals.

Return to the top of the page.

W

WACK (Wait before Acknowledgement) - A character sequence which allows a receiving device to indicate a "temporary not ready to receive" condition to the transmitting device.

Wideband - A communication channel of greater bandwidth than normal voice grade circuits.

Wiring Closet - Central location for termination and routing of on-premises wiring systems.

Workstation - Input / Output equipment at which an operator works. Also high-end desktop computers using RISC processors, often running the UNIX or XENIX operating systems.

Return to the top of the page.

X

X.21 - An ITU recommendation specifying a general purpose interface between DTE and DCE for synchronous operation.

X.25 - An ITU recommendation specifying an interface between DTE devices operating in packet mode across a packet switched network.

X.25 PAD - See PAD.

X.75 - An ITU recommendation which specifies the protocol used to communicate between packet switched networks.

Xenix - Microsoft trade name for a 16-bit microcomputer operating system derived from UNIX.

XNS - A peer-to-peer protocol developed by Xerox that has been incorporated into several local area network operating systems such as 3Com.

X-On/X-Off (Transmitter On/Transmitter Off) - Control characters used for flow control between two devices. X-On instructs a device to start transmission, and X-Off instructs it to stop transmission.

Return to the top of the page.

Cable Distance Guide

How far? How fast?

One Of the most common questions asked about a communication cable is "how long can it be?". Like everything else to do with computing, there is no simple answer to this question. In this section I would like to give you some insight into the speed and distance question and dispell a few myths.

The Parallel Interface

The Parallel interface moves data across the cable 8 bits (1 character) at a time. It does this by using two handshake signals. The first lets the sender tell the receiver that a character is on the interface and ready to be taken (read). The second lets the receiver tell the sender that it has taken the character and it is ready for another. It may sound cumbersome but this is how data is generally moved around inside a computer and it takes place at very high speeds. Typically, parallel interfaces move data at a rate of about 35,000 bits per second. Since there are 8 bits per character, this equals 4375 characters per second, or more than twice as much as a full screen Of data.

The parallel interface tends to be the standard used to drive printers because it is fast and generally less expensive to build than a serial interface. The speed however, tends to be somewhat irrelevant since most printers today operate at 200 to 300 characters per second. The real advantage of the parallel interface is its simplicity.

Distance, however, is a problem. The parallel interface is designed to drive a signal only a short distance. Exactly how far is dependent on the piece of equipment in question, although a good rule-of-thumb is not more than 25 feet. Some equipment will work successfully over longer cables, but don't rely on it. Stay With 20 or 25 feet and you should be safe.

The trend today is for a workgroup to share an expensive laser printer. Since the 25 foot limitation makes printer sharing between offices difficult, the serial interface is being used more and more to drive printers.

The Serial Interface

Serial interfaces transfer data serially one bit at a time, hence the name. Although there are several forms of the serial interface, the most common, by far, is that conforming to the RS-232C standard. This standard was first issued by the Electronics Industries Association (EIA) in 1968. It is sometimes erroneously referred to as the EIA interface. The RS-232C interface was originally intended to be used only between computers or terminals and modems. Today, however, it is used extensively for printers, plotters, digitisers and mice.

Unlike the parallel interface, the serial interface sends data out at a predetermined speed. This speed can usually be selected through software on a computer and through hard or soft swiches on a printer or terminal. One must take care to select the same speed on both devices being interconnected! The speeds commonly available range from 300 to 19,200 bits per second incremented by a factor of two.For timing reasons it is necessary to add two bits to every character for a total of 10, therefore this equates to 30 to 1920 characters per second. Typically, in a PC environment, you use a apeed of 9,600 bits per second (960 characters per second) to drive a printer. Although slower than a parallel interface this is still faster than most printers can print.

The question of how long the serial cable can be is ilidefined at best. The RS-232C standard addresses this very technically in terms of capacitance. What it essentially says is that an RS-232C interface must be able to drive a signal 15 metres at maximum speed. This is generally regarded as a minimum by designers and interfaces today are designed to do much better. Our experience is that most RS-232C interfaces are capable of operating errorfree at 9,600 bits per second up to about 50 metres using standard signalling wire. By building the cable out of low capacitance wire, that distance can be tripled to about 150 metres. This is assuming an electrically "clean" environment such as in a normal office building.

The maximum distance is inversely proportional to the speed. At 1200 bits per second the maximum distance is more like 1200 metres. To give you an idea of how this varies, figure 1 below shows the relationship of distance to speed using standard signalling cable that has a capacitance of 100 picofarads per metre. Low-capacitance wire is only 40 picofarads per metre which gives an 80% improvement.

data_rate.gif - 14.7 K

The rule-of-thumb for RS-232C distance is 35 to 50 metres using standard wire and 100 to 150 metres using low-capacitance wire. In all cases the wire should be shielded with the shield properly grounded to be effective. Although low-capacitance wire is about double the price of normal wire it triples the distance you can run. This very significant increase in distance over the parallel interface makes it practical for devices located in difference offices to be interconnected. When sharing printers or plotters it makes more sense to choose the serial interface to make the connection and use the parallel interface for a locally connected printer.

Return to the top of the page.

Code Conversion Chart

code.gif - 274.7 K

Return to the top of the page.

Data Circuit Troubleshooting Guide

trouble.gif - 126.0 K

Return to the top of the page.

Fibre Optic Technology

Introduction

Although the principle of total internal reflection was first demonstrated by Sir John Tyndall in 1870, it's only been in the last decade and a half that commercial products based on this principle have become widely available.

Why Use Fibre Optics?

Fibre optic technology is a dramatic step in the evolution of electronic communications. A constant concern in communications is ending ever-increasing amounts of information with greater efficiency over a medium requiring less space. As speeds and transmission distances increase, the difficulty of using copper cables also increases. The problems caused by electromagnetic interference (EMI), cross talk and signal distortion become more troublesome as operating frequencies increase. In both these areas - signal integrity and information carrying capacity - fibre optics offers many advantages over copper cable. It is these advantages that have sparked the intense interest in fibre optics as a next-generation transmission medium.

fibre.gif - 148.9 K

An optical fibre is a thin, flexible glass or plastic waveguide through which light can be transmitted. The most common use of fibre optics is as a transmission link connecting two electronic circuits. The fibre optic link has three main parts:

Transmitter

Receiver

Interconnection System

The transmitter converts a signal from electricity to light. It includes an electronic drive circuit and a light source such as an LED or laser diode.

The receiver converts a light signal back into an electrical signal. It includes a light detector and receiver circuit.

The interconnection system provides the transmission medium for connecting the transmitter and receiver. It includes optical fibres, connectors and associated hardware. The specific advantages of fibre optics include:

Wide Bandwidth

The bandwidth of optical fibres allows high data rates. Fibre optic telephone systems routinely carry thousands of voice channels over a single fibre. High data rates also allow efficient use of communication channels in computers and local area networks.

Low Loss

Optical fibres offer low attenuation of signals. Single mode fibres for example, have losses of less than 1 dB/km. Such low losses allow long interconnection runs without the need for regenerating the signal with a repeater.

Also important, attenuation in an optical fibre does not change with modulation frequency as it does with copper wire. Thus the length of the link or distance between repeaters decreases only slightly with increasing data rate. (This is due to dispersion rather than attenuation.)

Electromagnetic Immunity

Since the fibre is a dielectric, it is not affected by electromagnetic fields. The fibre neither picks up nor emits electromagnetic energy. It offers an ideal transmission medium for meeting EMI regulations. The elimination of cross talk, ground loops and signal distortion from EMI means error-free transmission, even in electrically hostile environments.

Small Size

An optical fibre is considerably smaller than its coaxial counterpart. Small size means the interconnection medium will not consume as much space, whether in a conduit running under a computer floor or a local area network running through an office building. In addition, the high data rates of a fibre-optic link may allow a single fibre with serial transmission to replace many wires of a parallel interconnection.

Light Weight

Since glass weighs less than copper, weight savings can be significant. Weight sensitive applications such as aircraft or motor cars may clearly benefit.

Security

Fibre optics offer secure communications. Not only is a fibre nearly impossible to tap without detection, the fact that it does not radiate electromagnetic energy makes other eavesdropping techniques equally useless.

Safety

The dielectric properties of fibre isolate it electrically. The fibre carries no current. Since fibres present no spark hazards, they can be used in flammable or explosive environments such as refineries or chemical plants.

Return to the top of the page.

Interfaces

inter1.gif - 250.1 K

Return to the top of the page.

Local Area Networks

Networks are collections of independent computers which can communicate with one another over a shared medium. Prior to the development of networking technology, individual machines were isolated and limited in their range of applications.

Local area networks (LANS) are those networks usually confined to a small geographic area, such as a single building or a college campus. LANs are not necessarily simple in design, however, as they may link hundreds of systems and service many thousands of users. The development of various standards for networking protocols and media has made possible the proliferation of LANs worldwide for business and educational applications.

Protocols

Network protocols are standards that allow computers to communicate. A typical protocol defines how computers should identify one another on a network, the form that the data should take in transit, and how this information should be processed once it reaches its final destination. Protocols also define procedures for handling lost or damaged transmissions or "packets." TCP/IP, LAT, IPX, Apple Talk, and DECnet are examples of network protocols.

Although each network protocol is different, they all use the physical cabling in the same manner. This common method of accessing the physical network allows protocols to peacefully coexist, and allows the builder of a network to use common hardware for a variety of protocols. This concept is known as "protocol independence," meaning that the physical network doesn't need to concern itself with the protocols being carried.

Ethernet

Ethemet is the most popular LAN technology in use today. Other LAN types include Token Ring, Fibre Distributed Data Interface (FDDI), and Local Talk. Each has its own advantages and disadvantages. Ethernet strikes a good balance between speed, price and ease of installation. These strong points, combined with wide acceptance into the computer marketplace and the ability to support virtually all popular network protocol, makes Ethemet the perfect networking technology for most computer users today.

The Ethernet standard is defined by the Institute for Electrical and Electronic Engineers (IEEE). IEEE Standard 802.3 defines rules for configuring an Ethernet as well as specifying how elements in a network interact with one another. Networks, equipment and network protocols that utilize and adhere to the IEEE standard will operate in the most efficient manner.

Media and Topologies

An important part of designing and installing an Ethernet is selecting the appropriate Ethernet medium for the environment at hand. There are four major types of media in use today: ThickWire, thin coax, unshielded twisted pair and fibre optic. Each type has its strong and weak points. Careful selection of the appropriate Ethernet medium can avoid recabling costs as your network grows.

Ethemet media are used in two general configurations or topologies: "bus" and "star." These two topologies define how "nodes" are connected to one another. A node is an active device connected to the network, such as a computer or a piece of networking equipment, for example, a repeater, a bridge or a router.

A bus topology consists of nodes strung together in series with each node connected to a long cable or bus. Many nodes can tap into the bus and begin communication with all other nodes on that cable segment. A break anywhere in the cable will usually cause the entire segment to be inoperable until the break is repaired.

Star media links exactly two nodes together. The primary advantage of this type of network is reliability. If a point to point segment has a break, it will only affect the two nodes on that link. Other nodes on the network continue to operate as if that segment were nonexistent. Obviously connecting only two computers together makes for a very limited network. Repeaters may be used to bind groups of point-to point segments together.

See the section on repeaters for more information on how to connect both point-topoint and/or bus segments together to make larger, more useful, networks.

ThickWire

ThickWire, or 10BASE5 Ethernet, is generally used to create large "backbones". A network backbone joins many smaller network segments into one large LAN. ThickWire makes an excellent backbone because it can support many nodes in a bus topology and the segment can be quite long. It can be run from workgroup to workgroup where smaller departmental networks can then be attached to the backbone.

A ThickWire segment can be up to 500m long and have as many as 100 nodes attached.

ThickWire is a thick, hefty, coaxial cable, and can be expensive and difficult to work with. A thick coaxial cable is used because of its immunity to common levels of electrical noise, helping to ensure the integrity of the network signals. The cable must not be cut to install new nodes, rather nodes must connect by drilling into the media with a device known appropriately as a "vampire tap".

Nodes must be spaced exactly in increments of 2.5m apart to prevent signals from interfering with one another. Due to this combinations of assets and liabilities, ThickWire is best suited for, but not limited to, backbone applications.

Thin Coax

Thin coax, or 10BASE-2 Ethernet, offers most of the advantages of ThickWire's bus topology with lower cost and easier installation. The coaxial cable is considerable thinner and more flexible than ThickWire, but it can only support 30 nodes, each at least 0.5m apart. Each segment must not be longer that 185m. Subject to these restrictions, thin coax still can be used to create backbones, with fewer nodes.

A thin coax segment is actually composed of many lengths of cables, each with a BNC connector on both ends. Each cable length connected to the next with a 'T' connector wherever a node is needed. Nodes can be connected or disconnected at the 'T' connectors as the need arises with no ill effect on the rest of the network. This coax's low cost, reconfigurability, and bus topology makes it an attractive medium for small networks, building departmental networks to connect backbones and for wiring a number of nodes together in the same room, such as a lab.

Twisted Pair

Unshielded twisted pair, or UTP, cable offers many advantages over the ThickWire and coax media. Because ThickWire and thin are coaxial cables, they are relatively expensive and require some care during installation. UTP is similar to the telephone cable that may already be installed and available for network use in your building.

Unshielded twisted pair cables come in a variety of grades, with each higher grade offering better performance. Level 5 cable is the highest and expensive grade, offering support for transmission rates of up to 100Mbps. Level 2 and Level 3 cables are far more popular for current 10BASE-T configurations; Level 4 cable can support speeds of up to 20Mbps Level 3 to 16Mbps. Level 2 and Level 1 cables are lowest grades and least expensive wire, designed primarily for voice and low-speed transmissions (less than 5Mbps); these should not be used in 10BASE-T networks.

A UTP, or 10BASE-T Ethernet, uses a star topology. Generally a computer is located one end of the segment, and the other end terminated in a central location with a repeater hub. Since UTP is often run in conjunction with telephone cabling, this central location can be a telephone closet or other area where it is convenient to connect the UTP segment to a backbone. UTP segments are limited to 100 meters, but UTP's point-to-point nature allows the rest of the network to function correctly if a break occurs in a particular segment.

Fibre Optic

Fibre Optic, or 10BASE-FL Ethernet, segments are similar to twisted pair. Fibre optic cable is more expensive, but it is invaluable for situations where electronic emissions and environmental hazards are a concem. The most common situation where these conditions threaten a network is in LAN connections between buildings. Lightning strikes can wreak havoc and easily destroy networking equipment. Fibre optic cables effectively insulate networking equipment from these conditions since they do not conduct electricity. Fibre optic cable can also be useful in areas where large amounts of electromagnetic interference are generally present, such as on a factory floor or inside a steel mill.

The Ethemet standard allows for fibre optic cable segments up to 2km long. Remote nodes and buildings that otherwise would not be reachable with LANs can be connected.

An investment in fibre optic cabling can be a wise one. As network technologies evolve and demands on the network increase, FDDI and other technologies faster than Ethernet can be run on the same cable, avoiding major rewiring.

Transceivers

Transceivers are used to connect nodes to the various Ethernet media. Transceivers, also known as Media Attachment Units (MAUs), attach to the Ethemet cable and provide an Application User Interface, or AUI, connector for the computer. The AUI connector, consists of a 15 pin D-shell type connector, female on the computer side, male on the transceiver side. Many Ethemet-compatible computers provide such an AUI connector. The transceiver is generally attached directly to the computer's AUI connector, or the transceiver may be attached to the computer with a specially shielded AUI cable. In addition to an AUI connector, many computers and interface network cards also contain a built-in 10BASE-T or 10BASE2 transceiver, allowing them to be connected directly to Ethemet without requiring an external transceiver.

Repeaters

Repeaters are used to connect two or more Ethemet segments of any media type. As segments exceed their maximum number of nodes or maximum length, signal quality begins to deteriorate. Repeaters provide the signal amplification and retiming required to connect segments.

Splitting a segment into two or more segments with a repeater allows a network to continue to grow. A repeater connection counts in the total node limit on each segment. For example, a thin coax segment may have 29 computers and 1 repeater, or a ThickWire segment can have 5 repeaters and 95 computers.

Ethemet repeaters are necessary in star topologies. As pointed out earlier, a network with only two nodes is of limited use. A twisted pair repeater allows several point to point segments to be joined into one network. One end of the point to point link is attached to the repeater and the other is attached to the computer with a transceiver. If the repeater is attached to a backbone, all computers at the end of the twisted pair segments can communicate with all the hosts on the backbone.

Repeaters also monitor all connected segments for basic characteristics necessary for Ethernet to run correctly. When these conditions are not met on a particular segment, for example when a break occurs, all segments in an Ethemet may become inoperable. Repeaters limit the effect of these problems to the faulty section of cable by "segmenting" the network, disconnecting the problem segment and allowing unaffected segments to function normally. A segment malfunction in a point to point network will generally only disable a single computer, where the same problem in a bus topology would disable all nodes attached to that segment.

Just as the various Ethemet media have segment limitations, larger Ethernets created with repeaters and multiple segments have restrictions. These restrictions generally have to do with timing constraints. Although electrical signals inside the Ethernet media travel near the speed of light, it still takes a finite time for the signal to travel from one end of a large Ethernet to another.

The Ethemet standard assumes it will not take more than a certain amount of time for a signal to propagate to the far ends of the Ethernet. If the Ethemet is too large, this assumption will not be met and the network may not perform correctly. Timing problems are not to taken lightly. When the Ethemet standard is violated, packets will be lost, network performance suffers, and applications become slow and may fail.

The IEEE 802.3 specifications describe rules for the maximum number of repeaters that can be used in a configuration. The maximum number of repeaters that can be found in the transmission path between two nodes is four; the maximum number of network segments between two nodes is five, with a further restriction that no more than three of those five segments may have other network stations attached to them (the other segments must be inter-repeater links which simply connect repeaters).

These rules are determined by calculations of maximum cable lengths and repeater delays. Networks which violate these rules may still be functional, but they are subject to sporadic failures or frequent problems of an indeterminate nature. Using repeaters simply extends a network to a larger size. As this occurs, available bandwidth on the network could become a problem. In this case, switches, bridges and routers can be used to partition one large network into several smaller segments which operate more efficiently.

Ethernet Switches

Ethernet switches increase overall network throughput by dividing a large single network into smaller segments. The switch filters unnecessary traffic from individual segments, thus making them faster and more efficient. When a packet is received by the switch, the switch determines the destination and source segments. If the segments are the same, the packet is dropped; if they are different, then the packet is forwarded. It is important to remember that all of the forwarding and filtering is done at the MAC layer, independent of any concern for network protocols.

There are two different design strategies to examining incoming packets: cut-through and store-and-forward. Cut-through switches read only a portion of the packet before forwarding or filtering it; these switches are fast but they do not filter out bad packets. Store-andforward switches read the whole packet before forwarding or filtering l; current designs are as fast as the cut-throughs but have the benefit of removing bad packets from the network.

Bridges

Bridges were originally designed to connect Ethernet networks together. As they were used more frequently, their ability to filter network traffic became more desirable and is sometimes now the only reason for installing one on a network. The first bridges were usually just two ports, connecting one Ethernet to another; they quickly became multiport devices as it became clear that perfommance would improve with the splitting of a network into multiple, smaller segments.

Bridge capabilities are defined in IEEE specification 802.1D. In addition to store-andfonward filtering and forwarding, bridges also support the Spanning Tree Algorithm for dealing with potential bridging loops between networks. This allows a configuration of redundant return paths. The Spanning Tree is a technique whereby all of the bridges communicate information on the topology of the network to insure that there are no bridg ing "loops" present (see following section on "Spanning Tree" for greater detail).

Routers

Routers work in a manner similar to switches and bridges in that they filter out network traffic. Rather than doing so by packet addresses they filter by specific protocol. Routers were born out of the necessity for dividing networks logically instead of physically. An IP router can divide a network into various subsets so that only traffic destined for particular IP addresses can pass between segments. The price paid for this type of intelligent forwarding and filtering is usually calculated in speed of the network. Such filtering takes more time than that exercised in a switch or bridge which only looks at the MAC layer. grouters is a term sometimes used to describe devices which have both bridging and routing capability, however switches and bridges frequently have some router-like features such as selective protocol filtering.

The Spanning Tree

The learning capabilities of switches and bridges make network loops deadly. As each device reams the network configuration, a loop would present conflicting information on which segment houses a specific address and force the device to forward all traffic. The Spanning Tree Algorithm is a software device for describing how switches and bridges can communicate to avoid network loops. By exchanging packets called BPDUs, the switches and bridges establish singular path for reaching each network segment. In some cases, a switch or bridge port may actually be fumed off if another pathway to that segment already exists. The process of passing the BPDU packets is continuous, so that if a switch or bridge suddenly fails, then the remaining devices will reconfigure the pathways to allow each segment to be reached. In some cases, network managers actually design loops into a bridged network so that if a switch or bridge does fail, the Spanning Tree will calculate the alternate route into the network configuration. For Spanning Tree to work correctly, all switches and bridges on the network must support the protocol.

Terminal and Printer Servers

Temminal servers and printer servers support the use of terminals and printers on networks, as well as modems and other serial devices.

A server which provides network access to serial devices only is usually called a "terminal server,. in spite of the fact that it can support serial printers as well as serial terminals. A "printer server" is expected to have at least one parallel port, but may also have one or more serial ports.

The primary difference between them is that temminal servers are bidirectional devices while printer severs are unidirectional devices.

Some printer manufacturers have introduced bidirectional data flow so printers can communicate with a host regarding which print fonts are available. This muddies the issue, but even in these cases the vast majority of the data still will flow in the direction of the printer.

Unlike transceivers and repeaters, terminal servers and print servers are intelligent devices which have their own network addresses and perform more than just a physical connection or signal-forwarding function.

The original role of terminal servers was to enable simple terminals to transmit data to and receive data from their host computers across local area networks, without requiring each terminal to have its own network connection. Sharing the local network connection was and is more convenient and more economical.

The same is true for saving on long-distance cabling. Since terminal servers usually have multiple ports, placing one near a distant workgroup can allow the whole group to be supported by a single network cable instead of requiring the stringing of one cable per terminal.

And while the terminal server's existence is still usually justified by convenience and cost considerations, its inherent intelligence provides many more advantages. Among these is enhanced remote monitoring and control. Terminal severs that support protocols like SNMP make networks easier to manage.

Another advantage comes in local communications, since a terminal server can facilitate communication between the devices which connect to it ( the devices in its "domain") without tying up network resources to do it. Being able to broadcast messages to all of these devices is another benefit.

Sharing resources and balancing workloads

Devices which are attached to a network through a terminal server or a print server can be shared between temminals and hosts at both the local site and throughout the network. A serial printer, for example, might be accessed by a local host or by a distant one.

Or several modems connected to one server on the net can act as a shared pool for computers, workstations, and terminals everywhere on the net.

Similarly, a single terminal connected t a server can access more than one computer in the network. (That's why networking's all about, after all) In fact, one terminal may be connected to several hosts at the same time (in multiple concurrent sessions), and switch back and forth between them.

Workload balancing comes almost automatically as a consequence of having multiple similar resources on the net. A terminal's server can locate the least loaded host with a given set of software on the network, or a host's server can find the next available printer at a given site.

Supporting multiple protocols

In the past it was possible to be forced to have two terminals side by side for use by the same person or group. This happened when the host computers the terminals were accessing used different communications protocols.

With the advent of multiprotocol terminal servers, this problem was alleviated. As long as the terminal server supports the protocol used by the host, the terminal attached to that server can access that host as if it were using the temminal's own native protocol. In fact it's quite possible for a terminal to be accessing a DEC host using the LAT protocol and a UNIX host using TCP/IP simultaneously.

A similar situation occurs with multiprotocol printer servers. The same printer might be accessed by both Novell and UNIX hosts. The print server can simply queue and print each job in the order in which print requests are received, regardless of protocol used.

Supporting workstations: Workstations and terminal servers can make a great combination too, even though workstations don't usually need a server to attach them to a network. The two devices work well together because workstations have plenty of computing power for supporting multiple users, but not enough connections to hook them up with, and terminal servers have the connections, but not the computing power.

In this scenario, the terminal server physically connects the temminal devices to the network. The result is the same in providing access to the workstation. An additional benefit is that the terminals can access other network resources.

Economically, it makes sense to have a single connection to the network instead of individual interface cards and transceivers.

Protocol conversion

DEC systems using the LAT protocol and UNIX systems using TCP/IP have no natural means to communicate with each omer, in spite of how common it is to have VAXes and Sun workstations in the same facility.

Given its natural translation ability, a multiprotocol terminal server can perfomm conversions between the protocols it knows, like LAT and TCP/IP, at least for those which are set up to work with terminals. While temminal server bandwidth isn't adequate for large file transfers, it can easily handle host-to-host inquiry/response applications, electronic mailbox checking, etc. And it is far more economical than the alternatives of acquiring expensive host software or special-purpose converters.

Terminal and print servers give their users great flexibility in configuring and managing their networks. Whether it is moving printers and other peripherals form one network to another, expanding the dimensions of interoperability, or preparing for growth.

Return to the top of the page.

Modem Selection Guide

What is a modem?

A modem is simply a device that enables personal computers, minicomputers, mainframes and other terminals to communicate with each other via a standard voice telephone line.

It is well known that computers process information internally in the form of digital electronic signals. But because the telephone line is 'analog' (designed originally for voice or analog sound), a modem is required to convert the computer's digital electronic signals into analog electronic signals. These can then be transmitted across the telephone network on either a switched (or dial-up) line, or a private (leased) line. These lines are usually provided in Australia by Telstra or Optus. Obviously, these signals need reconversion to digital signals by another modem acting as a receiver.

The word "modem" is an abbreviation of the words "MOdulator/DEmodulator; modulation describes the process of converting the digital signal into analog while demodulation is the process whereby the digital information is extracted from the analog signal.

Asynchronous or Synchronous?
Full or half duplex?

Most PC applications are asynchronous. The difference between asynchronous and synchronous communications can be compared with the difference between motor car traffic on a freeway and a train. Asynchronous data is like the cars, all going in the same direction and at about the same speed but with different distances in between. Each car can and must be driven on its own. On the other hand the train, synchronous data, has one engine and all the carriages are closely coupled, all travelling at around the same speed, precisely and closely synchronised.

Asynchronous is a little more error prone and less efficient but is simpler in technology terms. It was the first form of data communication and was originally designed for telex services. Synchronous communications is more sophisticated, more accurate and more efficient. It is quite a bit more complex and today is mainly used when connecting to mainframe computers.

To continue the car/train analogy, half duplex describes a one way street (or track) and the traffic can only 'turn around' if the way is clear. Full duplex is a two lane system where data can flow freely in both directions at the same time. Half duplex is only used in specialised mainframe applications (notably point-of-sale and electronic funds transfer) and PC users really need only be aware of the full duplex techniques.

Clearly, full duplex links can support higher traffic throughput and indeed the receiver and sender can be processing quite different applications.

Selecting Modems

Three basic criteria in selecting modems are the volume of data transmitted and the speed and distance it needs to travel. Volume is a function of both the characters per transaction and the number of transactions per day required to support your application. Volume of traffic then, not only determines the speed of transmission but also whether you need short-haul modems for a metallic line or long-haul modems for a telephone line. Here are some points to consider:

1. Will the modem interface be compatible with your terminals and/or computers?
2. Will the new modems be able to communicate with any existing ones?
3. Will the modem be used for leased, dial or metallic line operation?
4. Will the modem be used in a point-to-point or multipoint configuration?
5. What kind of diagnostic capabilities do you need?
6. What kind of error rate can you accept?
7. Is there a multiple speed capability?

Return to the top of the page.

Modular Jacks and Line Cords

Introduction

Modular typically are manufactured in four styles. Eight position, eight position non-keyed, six position and six position MMJ (modified modular jack). DEC introduced the MMJ and DEC hardware is its predominant application.

EIA and AT&T

EIA/TIA-568 has adopted the TIA and AT&T wiring schemes. They are similar, but pairs two and three are reversed. The TIA scheme is popular because it is compatible with one or two pair USOC systems. Both schemes will support ISDN applications.

IBM Token Ring

Token Ring usually uses an 8 position jack. The wiring scheme is compatible with TIA, AT&T and USOC wiring schemes because it uses the two centre pairs.

page1.gif - 105.5 K

"RJ" Registered Jack

Modular jacks should be referred to by their position count; 8position, 8-position keyed, 6-position, 6-position modified (or MMJ). The RJ designations actually refer to specific wiring configurations called Universal Service Ordering Codes (USOC). The RJ designation means Registered Jack. Each of the 8 position jacks and the 6 position jack can be wired for different RJ configurations. The 6 position jack can be wired as an RJ1 1C (1 pair), RJ14C (2 pair) or RJ25C (3 pair). The 8 position jacks can be wired as RJ61 C (4 pair) and RJ48C; the keyed jack can be wired as RJ45S, RJ46S and RJ47S.

Pairing

Proper pairing must be observed in all cabling systems. It is essential that the paring of the wires in the modular plugs and jacks be identical to that of the building and horizontal and vertical wiring. If proper pairing isn't observed, poor quality or non-existent data transmission may result.

Modular Cords

Modular cords are used at patch panel locations and to connect workstations to the building wiring jacks. In both cases, the cord should be wired straight through, that is pin 1 to 1, pin 2 to 2, and so forth. The cable itself should be a high grade twisted pair cable, preferably of stranded construction for maximum flexibility. Most voice systems use reversed cables for connecting the device to the building wiring. Thus, voice and data base cords or patch cables are seldom interchangeable.

For data transmission applications such as Local Area Networks and other computer systems, the quality of the patch cables and base cords is of the utmost importance. Low quality cable will provide unsatisfactory system performance and will require replacement at some time in the future. Flat, non paired cable of the type used in voice systems (often referred to as Silver Satin") is not acceptable for any type of high-speed data transmission applications.

To identify whether an individual patch cord is wired straight through or reversed, hold the plugs at either end of the cable aligned side by side with the contacts facing you (up), the clip down and compare the wire colours from left to right. The colours should appear in the same order on both plugs if the cord is wired straight through. If the colours appear reversed on the second plug from right to left, the cord is reversed.

Return to the top of the page.

PC Video Standards

pc.gif - 255.3 K

Return to the top of the page.

Protocols

Asynchronous Data Format

Start Bit

5, 6, 7 or 8 Data Bits

Parity Bit

Stop Bit

Start Bit:	Tells the receiving device to start looking for data bits.
Data Bits:	The binary value of the data bits represent a character in accordance with the code being used (eg ASCII, EBCDIC etc)
Parity Bit:	This bit is optionally used to enable the receiving device to detect any errors that have corrupted the data during transmission.
Stop Bit:	Tells the receiving device to stop looking for data bits.

Character Oriented Protocols - BISYNC

						ITB
SYN	SYN	SOH	Header	STX	Text	ETX	BCC	BCC
						ETB

SYN:	Sync characters establish and maintain character synchronisation between transmitting and receiving stations. They are also used as a time fill in the absence of data. For character synchronisation, the receving station must recognise at least two contigu
SOH:	Start of header indicates the beginning of the optional header block.
Header:	A user defined segment containing such information as: source or destination code, priority, date, message type.
STX:	Start of text terminates the heading and indicates the beginning of the text segment.
Text:	Information portion of the message.
ITB:	End of intermediate block. Indicates that text is to continue in the next block.
ETB:	End of transmission block. Indicates that the message is ended, however more blocks may follow.
ETX:	End of text. Same as ETB except that it indicates no more data blocks will follow.
BCC:	Block check character for error detection.

Bit Oriented Protocols - SDLC, HDLC, X.25

Flag

Address

Control

Data

FCS

Flag

Flags:	Flags (hex 7E) act as frame delimiters
Address Field:	Command frames contain receiving station's address. Response Field frames contain sending station's address.
Control Field:	Identifies function and purpose of frame. Contains commands, responses and sequence numbers.
Data Field:	The data, or information field contains any number of bits, typically in multiples of 8 (octets).
FCS:	Frame check sequence for error detection.

Return to the top of the page.

Please

me and tell me if you liked my webpage on terminology and the reasonably comprehensive glossary I managed to collect, or even if you have any contributing sites, additional words/terms on similar info that I can include here as additional links.

Click here to go back to my Technical Page

This page has been accessed times.

Last revised: Friday, 04 July 1997