`
`A Local Area Network
`
`Data Link Layer
`and
`Physical Layer
`Specifications
`
`11111111
`
`infel
`
`XEROX
`
`Digital Equipment Carporation
`
`Maynard, MA
`
`Intel Corporation
`
`Santa Cia ra, CA
`
`Xerox Corporation
`
`Stamford, CT
`
`Version 1.0
`
`September 30, 1980
`
`Page 1
`
` Dell Inc.
` Exhibit 1017
`
`
`
`IMPORTANT INFORMATION AND DISCLAIMERS
`
`1. This specification includes subject matter relating to a patent(s) of Xerox
`Corporation. No license under such patent(s) is granted by implication, estoppel
`or otherwise as a result of publication of this specification. Applicable licenses
`may be obtained from Xerox Corporation.
`·
`
`2. This specification is furnished for informational purposes only. Digital, Intel, and
`Xerox do not warrant or represent that this specification or any products made in
`conformance with it will work in the intended manner or be compatible with
`other products in a network system. Nor do they assume responsibility for any
`errors that the specification may contain, or have any liabilities or obligations for
`damages (including but not limited to special, indirect or consequential damages)
`arising out of or in connection with the use of this specification in any way.
`Digital, Intel and Xerox proqucts may follow or deviate from the specification
`without notice at any time.
`
`3. No representations or warranties are made that this specification or anything made
`from it is or will be free from infringements or patents of third persons.
`
`Page 2
`
`
`
`ETHERNET SPEOFICA TION: Preface
`
`i
`
`Preface
`
`-
`
`This document contains the specification of the Ethernet, a local area network
`developed jointly by Digital Equipment Corporation, Intel Corporation, and Xerox
`Corporation. The Ethernet specification arises from an extensive collaborative effort
`of the three corporations, and several years of work at Xerox on an earlier prototype
`Ethernet
`This specification is intended as a design reference document, rather than an
`introduction or tutorial. Readers seeking introductory material are directed to the
`reference list in Section 2, which cites several papers describing the intent, theory,
`and history of the Ethernet.
`This document contains 7 sections, falling into three main groups:
`Sections 1, 2, and 3 .provide an overall description of the Etht!rnet, including its
`goals, and the scope of the specification.
`Sections 4 and 5 describe the architectural structure of the Ethernet in terms of a
`functional model consisting of two layers, the Data Link Layer and the Physical
`Layer.
`Sections 6 and 7 specify th~ two layers in detail, providing the primary technical
`specification of the Ethernet.
`Readers wishing to obtain an initial grasp of the organization and content of the
`specification will be best served by reading Sections 1, 3, and 4. Readers involved in
`actual implementation of the Ethernet will find Sections 5, 6, and 7 to contain the
`central material of the specification. Section 2 provides references, and the
`appendices provide supplementary material.
`The approach taken in the specification of the Data Link Layer in Section 6 is a
`procedural one; in addition to describing the necessary algorithms in English and
`control flow charts, the specification presents these algorithms in the language Pascal.
`This approach makes clear the required behavior of Data Link Layer, while leaving
`individual implementations free to exploit any appropriate technology.
`Because the procedural approach is not suitable for specifying the details of the
`Physical Layer, Section 7 uses carefully worded English prose and numerous figures
`and tables to specify the necessary parameters of this layer.
`Some aspects of the Ethernet are necessarily discussed in more than one place in this
`specification. Whenever any doubt arises concerning the official definition in such a
`case, the reader should utilize the Pascal procedural specification of the Data Link
`Layer in Section 6.5, and the detailed prose specification of the Physical Layer in
`Sections 7.2 through 7.9.
`
`Page 3
`
`
`
`ii
`
`ETHERNET SPECIFIC<\ TION: Preface
`
`One aspect of an overall network architecture which is not addressed by this
`specification is network management The network management facility performs
`operation, maintenence, and planning functions for the network:
`- Operation functions include parameter setting, such as address selection.
`- Maintenance functions provide for fault detection, isolation, and repair.
`- Planning functions include collection of statisical and usage information, necessary
`for planned network growth.
`While network management itself is properly performed outside the Ethernet Data
`Link and Physical Layers, it requires appropriate additional interfaces to those layers,
`which will be defined in a subsequent version of this specification.
`
`Page 4
`
`
`
`ETHERNET SPECIFICATION: Contents
`
`iii
`
`Table of Contents
`
`Preface
`1. INTRODUCTION
`2. REFERENCES
`3. GOALS AND NON-GOALS
`3.1 Goals
`3.2 Non-Goals
`4. FuNCTIONAL MODEL OF THE ETHERNET ARCHITECTURE
`4.1 Layering
`4.2 Data Link Layer
`4.3 Physical Layer
`4.4 Ethernet Operation and the Functional Model
`4.4.1 Transmission Without Contention
`4.4.2 Reception Without Contention
`4.4.3 Collisions: Handling of Contention
`5.~INTER-LA YER INTERFACES
`5.1 Client Layer to Data Link Layer
`5.2 Data Link Layer to Physical Layer
`6. ETHERNET DATA LINK LAYER SPECIFICATION
`6.1 Data link Layer Overview and Model
`6.2 Frame format
`6.2.1 Address Fields
`6.2.1.1 Destination Address Field
`6.2.1.2 Source Address Field
`6.2.2 Type Field
`6.2.3 Data Field
`6.2.4 Frame Check Sequence Field
`6.2.5 Frame Size Limitations
`6.3 Frame Transmission
`6.3.1 Transmit Data Encapsulation
`6.3.1.1 Frame Assembly
`6.3.1.2 Frame Check Sequence Generation
`6.3.2 Transmit Link Management
`
`1
`1
`3
`4
`4
`5
`6
`8
`9
`10
`12
`12
`12
`13
`15
`15
`16
`19
`19
`19
`21
`21
`21
`21
`22
`22
`22
`23
`23
`23
`23
`23
`
`Page 5
`
`
`
`iv
`
`ETHERNET SPEOFICA TION: Contents
`
`23
`6.3.2.1 Carrier Deference
`24
`6.3.2.2 Interframe Spacing
`24
`6.3.2.3 Collision Handling
`6.3.2.3.1 Collision Detection and Enforcement 24
`6.3.2.3.2 Collision Backoff and Retransmission 25
`6.4 Frame Reception
`25
`6.4.1 Receive Data Decapsulation
`25
`6.4.1.1 Framing
`25
`6.4.1.1.1 Maximum Frame Size
`26
`6.4.1.1.2 Integral Number of Octets in Frame
`26
`6.4.1.2 Address Recognition
`26
`6.4.1.2.1 Physical Addresses
`26
`6.4.1.2.2 Multicast Addresses
`26
`6.4.1.3 Frame Check Sequence Validation
`26
`6.4.1.4 Frame Disassembly
`27
`6.4.2 Receive Link Management
`27
`6.4.2.1 Collision Filtering
`27
`6.5 The Data Link Layer Procedural Model
`27
`6.5.1 Overview of the Procedural Model
`27
`6.5.1.1 Ground Rules for the Procedural Model
`27
`6.5.1.2 Use of Pascal in the Procedural Model
`29
`6.5.2 The Procedural Model
`31
`6.5.2.1 Global Declarations
`34
`6.5.2.1.1 Common Constants and Types
`34
`6.5.2.1.2 Transmit State Variables
`35
`6.5.2.1.3 Receive State Variables
`35
`6.5.2.1.4 Summary oflnterlayer Interfaces
`36
`6.5.2.1.5 State Variable Initialization
`37
`6.5.2.2 Frame Transmission
`38
`6.5.2.3 Frame Reception
`42
`6.5.2.4 Common Procedures
`44
`7. ETHERNET PHYSICAL LAYER SPECIFICATION
`45
`7.1 Physical Channel Overview and Model
`45
`7.1.1 Channel Goals and Non-Goals
`45
`7.1.1.1 Goals
`45
`7.1.1.2 Non-Goals
`45
`7.1.2 Characteristics ofthe Channel
`46
`7.1.3 Functions Provided by the Channel
`46
`7.1.4 Implementation of the Channel
`46
`7.1.4.1 General Overview of Channel Hardware
`47
`7.1.4.2 Compatibility Interfaces
`48
`7.1.5 Channel Configuration Model
`49
`7.1.6 Channel Interfaces
`53
`
`Page 6
`
`
`
`ETHERNET SPECIFICATION: Contents
`
`v
`
`7.2 Transceiver Cable Compatibility Interface Specifications
`7.2.1 Transceiver Cable Signals
`7 .2.1.1 Transmit Signal
`7.2.1.2 Receive Signal
`7.2.1.3 Collision Presence Signal
`7.2.1.4 Power
`7 .2.2 Transceiver Cable Parameters
`7.2.2.1 Mechanical Configuration
`7.2.2.2 Characteristic Impedance
`7.2.2.3 Attenuation
`7 .2.2.4 Velocity of Propagation
`7 .2.2.5 Pulse Distortion
`7 .2.2.6 Resistance ~
`7 .2.2.7 Transfer Impedance
`7.2.3 Transceiver Cable Connectors
`7 .2.4 Transceiver Cable Drive
`7.2.5 Trans<..eiver Cable Receive
`7.2.5.1 Load Impedance and Termination
`7.2.5.2 Common Mode and CMRR
`7.3 Coaxial Cable Compatibility Interface Specifications
`7.3.1 Coaxial Cable Component Specifications
`7.3.1.1 Coaxial Cable Parameters
`7 .3.1.1.1 Characteristic Impedance
`7.3.1.1.2 Attenuation
`7.3.1.1.3 Velocity ofPropagation
`7.3.1.1.4 Mechanical Requirements
`7.3.1.1.5 Pulse Distortion
`7.3.1.1.6 Jacket Marking
`7.3.1.1.7 Transfer Impedance
`7.3.1.2 Coaxial Cable Connectors
`7.3.1.3 Coaxial Cable Terminators
`7.3.1.4 Transceiver-to-Coaxial Cable Connections
`7.3.2 Coaxial Cable Signaling
`7.4 Transceiver Specifications
`7.4.1 Transceiver-to-Coaxial Cable Interface
`7.4.1.1 Input Impedance
`7.4.1.2 Bias Current
`7.4.1.3 Transmit Output Levels
`7 .4.2 Transceiver-to-Transceiver Cable Interface
`7.4.2.1 Transmit Pair
`7 .4.2.2 Receive Pair
`7 .4.2.3 Collision Presence Pair
`7.4.2.4 Power Pair
`7.4.3 Electrical Isolation
`
`53
`53
`53
`54
`54
`54
`54
`54
`55
`55
`55
`55
`55
`55
`56
`56
`57
`57
`57
`58
`58
`58
`58
`58
`58
`59
`59
`59
`59
`60
`60
`61
`61
`62
`62
`62
`62
`63
`63
`63
`63
`64
`64
`64
`
`Page 7
`
`
`
`vi
`
`ETHERNET SPEOFICA TION: Contents
`
`7.4.4 Reliability
`7.5 Channel Logic
`7.5.1 Channel Encoding
`7.5.1.1 Encoder
`7.5.1.2 Decoder
`7.5.1.3 Preamble Generation
`7.5.2 Collision Detect Signal
`7.5.3 Carrier Sense Signal
`7 .5.4 Channel Framing
`7.5.4.1 Beginning-of-Frame Sequence
`7.5.4.2 End-of-Frame Sequence
`7.6 Channel Configuration Requirements
`7 .6.1 Cable Sectioning
`7.6.2 Transceiver Placement
`7 .6.3 System Grounding
`7 .6.4 Repeaters
`7 .6.4.1 Carrier Detect and Transmit Repeat
`7.6.4.2 Collision Detect and Collision Repeat
`7.6.4.3 Repeater Signal Generation
`7.6.4.3.1 Signal Amplification
`7.6.4.3.2 Signal Timing
`7.7 Environment Specifications
`7.7.1 Electromagnetic Environment
`7.7.2 Temperature and Humidity
`
`64
`64
`64
`65
`65
`66
`67
`67
`68
`68
`69
`69
`69
`70
`70
`71
`71
`71
`71
`71
`71
`72
`72
`72
`
`Page 8
`
`
`
`ETHERNET SPEOFICA TION: Contents
`
`vii
`
`Appendices
`
`APPENDIX A: GLOSSARY
`APPENDIX B: ASSIGNMENT OF ADDRESS AND TYPE VALUES
`APPENDIX C: CRC IMPLEMENTATION
`APPENDIX D: IMPLEMENTATION OF TRANSCEIVER CABLE
`DRIVER AND RECEIVER
`APPENDIX E: INTERFRAME RECOVERY
`
`Figures and Tables
`
`Figure 4-1: Ethernet Architecture and Typical Implementation
`Figure 4-2: Architectural Layering
`Figure 4-3: Data Link Layer Functions
`Figure 4-4: Physical Layer Functions
`Figure G-:;.: Data Link Layer Frame Format
`Figure 6-2: Structure of the Data Link Procedural Model
`Figure 6-3: Control Flow Summary -- Client Layer Processes
`Figure 6-4: Control Flow Summary --Data Link Layer Processes
`Figure 7-1: Physical Channel Configurations
`Table 7-1: Physical Channel Propagation Delay Budget
`Figure 7-2: Maximum Transceiver Cable Transfer Impedance
`Figure 7-3: Typical Transceiver Cable Waveform
`Figure 7-4: Maximum Coaxial Cable Transfer Impedance
`Figure 7-5: Typical Coaxial Cable Waveform
`Figure 7-6: Manchester Encoding
`Figure 7-7: Preamble Encoding
`Figure 7-8: Functional Logic of collisionDetect Signal
`Figure 7-9: Functional Logic of carrierSense Signal
`Figure C-1: CRC Implementation
`Figure D-1: Typical Transceiver Cable Driver
`Figure D-2: Typical Transceiver Cable Receiver
`
`73
`76
`77
`
`80
`82
`
`7
`8
`10
`11
`20
`31
`32
`33
`50
`52
`55
`57
`60
`62
`65
`66
`67
`68
`78
`80
`81
`
`Page 9
`
`
`
`viii
`
`ETHERNET SPEOFICA TION: Contents
`
`Page 10
`
`
`
`ETHERNET SPECIFIC<\ TION: Introduction
`
`1
`
`1. INTRODUCTION
`
`The Ethernet local area network provides a communication facility for high speed
`data exchange among computers and other digital devices located within a moderate(cid:173)
`sized geographic area.
`Its primary characteristics include:
`Physical Layer:
`Data rate: 10 Million bits/sec
`Maximum station separation: 2.5 Kilometers
`Maximum number of stations: 1024
`Medium: Shielded coaxial cable, base-band signalling
`Topology: Branching non-rooted tree
`Data Link Layer:
`Link control procedure: Fully distributed peer protocol, with statistical
`contention resolution (CSMA/CD)
`Message protocol: Variable size frames, "best-effort" delivery
`
`The Ethernet, like other local area networks, falls in a middle ground between long
`distance, low speed networks which carry data for hundreds or thousands of
`kilometers, and specialized, very high speed interconnections which are generally
`limited to tens of meters. The Ethernet is intended primarily for use in such areas as
`office automation, distributed data processing, terminal access, and other situations
`requiring economical connection to a· local communication medium carrying bursty
`traffic at high peak data rates. Use in situations demanding resistance to hostile
`environments, real-time response guarantees, and so on, while not specifically
`excluded, do not constitute the primary environment for which the Ethernet is
`designed.
`
`The precursor to the Ethernet specified in this document was the "Experimental
`Ethernet", designed and implemented by Xerox in 1975, and used continually since
`that time by thousands of stations. The Ethernet defined here builds on that
`experience, and on the larger base of the combined experience of Digital, Intel, and
`Xerox in many forms of networking and computer interconnection.
`
`In specifying the Ethernet, this document provides precise detailed definitions of the
`lowest two layers of an overall network architecture. It thus defines what is generally
`
`Page 11
`
`
`
`2
`
`ETHERNET SPEOFICA TION: Introduction
`
`referred to as a link-level facility.
`It does not specify the higher level protocols
`needed to provide a complete network architecture. Such higher level protocols
`would generally include such functions as internetwork communication, error
`·recovery, flow control, security measures (e.g. encryption}, and other higher level
`functions that increase the power of the communication facility and/or tailor it to
`In particular, it should be noted that all error recovery
`specific applications.
`functions have been relegated to higher level protocols, in keeping with the low error
`rates that characterize local networks.
`
`One of the main objectives of this specification is compatibility. As stated in Section
`3, it is intended that every implementation of the Ethernet be able to exchange data
`with every other implementation. It should be noted that higher level protocols raise
`their own issues of compatibility over and above those addressed by the Ethernet and
`other link-level facilities. This does not eliminate the importance of link-level
`compatibility, however. While the compatibility provided by the Ethernet does not
`guarantee solutions to higher level compatibility problems, it does ·provide a context
`within which such protlems ~ be addressed, by avoiding low level incompatibilities
`that would make direct communication impossible.
`
`Page 12
`
`
`
`ETHERNET SPEOFICA TION: References
`
`3
`
`2. REFERENCES
`
`The following three papers describe the Experimental Ethernet, and are rep;inted in:
`"The Ethernet Local Network: Three Reports," Xerox Palo Alto Research Center
`(February, 1980.)
`Technical Report CSL-80-2.
`[1] Metcalfe, R. M. and Boggs, D. R., "Ethernet: Distributed Packet Switching for
`Local Computer Networks," Communications of the ACM 19 7 (July 1976).
`"Practical Considerations in Ethernet Local
`[2] Crane, R. C. and Taft, E. A.
`Network Design," Presented at Hawaii International Conference on System
`Sciences (January, 1980).
`Shoch, J. F. and Hupp, J. A. "Measured Perfonnance of an Ethernet Local
`Network," Presented at Local Area Communications Network Symposium
`Boston (May 1979).
`
`[3]
`
`The following references describe the ISO Open Systems Model:
`[4] Zimmennann, H., · "OSI Reference Model -- The ISO Model of Architecture
`IEEE Transactions on Communication
`for Open Systems Interconnection,"
`COM-28 4 (April 1980).
`International Organization for Standardization (ISO), "Reference Model of
`Open Systems Interconnection," Document no. ISO/TC97/SCJ6 N227 (June
`1979).
`
`[5]
`
`[6]
`
`The following references describe the Pascal language (used in the Data Link Layer
`procedural model) and its derivative Concurrent Pascal:
`Jensen, K. and Wirth, N., Pascal User Manual and Report, 2nd Edition.
`Springer-Verlag (1974).
`[7] Brinch Hansen, P., Concurrent Pascal Report. Technical Report CIT-IS-TR
`17, California Institute of Technology (1975).
`
`The following references discuss the CRC code used for the frame check sequence:
`[8] Hammond, J. L., Brown, J. E. and Liu, S. S., "Development of a Transmission
`Error Model and an Error Control Model," Technical Report RADC-TR-75-
`138, Rome Air Development Center (1975).
`[9] Bittel, R., "On Frame Check Sequence (FCS) Generation and Checking.,"
`ANSI working paper X3-S34-77-43, (1977).
`
`Page 13
`
`
`
`4
`
`ETHERNET SPEOFICA TION: Goals and Non·Goals
`
`3. GOALS AND NON·GOALS
`This section states the assumptions underlying the design of the Ethernet
`
`3.1 Goals
`The goals of the Ethernet design are:
`Simplicity: Features which would complicate the design without substantially
`contributing to the meeting of the other goals have been excluded.
`Low cost: Since technological improvements will continue to reduce the overall
`cost of stations wishing to connect to the Ethernet, the cost of the connection
`itself should be minimized.
`Compatibility: All implementations of the Ethernet should be capable of
`exchanging data at the data link level. For this reason. the specification
`avoids optional features. to eliminate the possibility of incompatible variants
`of the Ethernet
`Addressing flexibility:
`The addressing mechanisms shouid provide the
`capability to target frames to a single node. a group of nodes. or to all nodes
`on the network.
`Fairness: All nodes should have equal access to the network when averaged
`over time.
`Progress: No single node operating in accordance with the protocol should be
`able to prevent the progress of other nodes.
`High speed: The network should operate efficiently at a data rate of 10
`Megabits per second.
`Low delay: At any given level of offered traffic. the network should introduce
`as little delay as possible in the transfer of a frame.
`Stability: The network should be stable under all load conditions. in the sense
`that the delivered traffic should be a monotonically non-decreasing function
`of the total offered traffic.
`Maintainability: The Ethernet design should allow for network maintenance.
`operation. and planning.
`Layered Architecture: The Ethernet design should be specified in layered terms
`to separate the logical aspects of the data link protocol from the physical
`details of the communication medium.
`
`Page 14
`
`
`
`ETHERNET SPEOFICA TION: Goals and Non·Goals
`
`5
`
`3.2 Non·Goals
`The following are not ·goals of the Ethernet design:
`Full duplex: At any given instant, the Ethernet can transfer data from one
`Bi-directional
`to one or more destination stations.
`source station
`communication is provided by rapid exhange of frames, rather than full
`duplex operation.
`Error control: Error handling at the data link level is limited to detection of bit
`errors in the physical channel, and the detection and recovery from collisions.
`Provision of a complete error control facility to handle detected errors is
`relegated to higher layers of the network architecture.
`Security: The data link protocol does not employ encryption or other
`mechanisms to provide security. Higher layers of the network architecture
`may provide such facilities as appropriate.
`Speed flexibility: This specification defines a physical channel operating at a
`single fixed data rate of 10 Megabits per second.
`Priority: The data link protocol provides no support of priority station
`operation.
`Hostile user. There is no attempt to protect the network from a malicious user
`at the data link level.
`
`Page 15
`
`
`
`6
`
`ETHERNET SPECIFICATION: Functional Model of the Ethernet Architecture
`
`4. FUNCTIONAL MODEL OF THE ETHERNET ARCHITECTURE
`
`There are two important ways to view the Ethernet design, corresponding to:
`Architecture, emphasizing the logical divisions of the system, and how they fit
`together.
`Implementation, emphasizing the actual components, and their packaging and
`interconnection.
`Figure 4-1 illustrates these two views as they apply to a typical implementation,
`showing how each view groups the various functions.
`This document is organized along architectural lines, emphasizing the large-scale
`separation of the Ethernet system into two parts: the Data Link Layer and the
`Physical Layer. These layers are intended to correspond closely to the lowest layers
`of the ISO Model for Open Systems Interconnection [4,5]. Architectural organization
`of the specification has two main advantages:
`Clarity: A clean overall division of the design along architectural lines IJakcs the
`specification clearer.
`Flexibility: Segregation of medium-dependent aspects in the Physical Layer allows
`the Data Link Layer to apply to transmission media other than the specified
`coaxial cable.
`As is evident in Figure 4-1, the architectural model is based on a set of interfaces
`different from those emphasized in the implementations. One crucial aspect of the
`design, however, must be addressed largely in terms of the implementation
`interfaces: compatibility. Two important compatibility interfaces are defined within
`what is architecturally the Physical Layer:
`Coaxial cable interface: To communicate via the Ethernet, all stations must adhere
`rigidly to the exact specification of coaxial cable signals defined in this document,
`and to the procedures which define correct behavior of a station. The medium(cid:173)
`independent aspects of the Data Link Layer should not be taken as detracting
`from this point: communication via the Ethernet requires complete compatibility at
`the coaxial cable interface.
`
`Page 16
`
`
`
`ETHERNET SPEOFICA TION: Functional Model of the Ethernet Architecture
`
`7
`
`ARCHITECTURE
`
`Client Layer
`
`Data Link Layer
`
`Physical Layer
`
`FUN
`CfiONS
`
`I
`
`Data Link Controller
`I
`
`I I
`
`Physical Channel
`I
`
`I
`
`Data
`
`Encapsulation - Mgmt
`
`Link
`
`Encode&
`1-- Decode
`
`Transmit&
`Receive
`
`Ethernet Controller Board
`
`I
`
`~Jciver II Tjeiver I
`
`Cable
`
`Coax
`Cable
`
`to
`station
`
`-
`
`Station
`
`Interface
`
`1-
`
`I
`
`TYPICAL
`IMPLEMENTATION
`
`to 1/0 bus, etc.
`
`r Compatibility
`
`Interfaces -----~
`
`Figure 4-1: Ethernet Architecture and Typical Implementation
`
`Page 17
`
`
`
`8
`
`ETHERNET SPEOFICA TION: Functional Model of the Ethernet Architecture
`
`Transceiver cable interface: It is anticipated that most stations· will be located some
`distance away from their connection to the coaxial cable. While it is necessary to
`place a small amount of circuitry (the transceiver) directly adjacent to the coaxial
`cable, the majority of the electronics (the controller) can and should be placed
`with the station. Since it is desirable for the same transceiver to be usable with a
`wide variety of stations, a second compatibility interface, the transceiver· cable
`interface, is defined. While conformance with this interface is not strictly
`necessary to insure communication, it is highly recommended, since it allows
`maxunum flexibility in intermixing transceivers and stations.
`
`4.1 Layering
`
`The major division in the Ethernet Architecture is between the Physical Layer and
`the Data Link Layer, corresponding to the lowest two levels in the ISO model. The
`higher levels of the overall network architecture, which use the Data Link Layer,
`will be collectively referred to in this document as the "Client Layer" since, strictly
`speaking, the identity and function of higher level facilitie'i are 0utside the scope of
`this specification. The intent, however, is that the Ethernet Physical and Data Link
`Layers support the higher layers of the ISO model (Network Layer, Transport
`Layer, etc.).
`
`The overall structure of the layered architecture is shown in Figure 4-2.
`
`OientLayer
`
`Interface
`
`Data Link Layer
`
`Interface
`
`Physical Layer.
`
`Figure 4-2: Architectural Layering
`
`Page 18
`
`
`
`ETHERNET SPECIFICATION: Functional Model ofthe Ethernet Architecture
`
`9
`
`In the architectural model used here, the layers interact via well defined interfaces.
`
`The interface between the Client Layer and the Data Link Layer includes
`facilities for transmitting and receiving frames, and provides per-operation
`status information for use by higher-level error recovery procedures.
`
`The interface between the Data Link Layer and the Physical Layer includes
`signals for framing (carrier sense, transmit initiation) and contention resolution
`(collision detect), facilities for passing a pair of serial bit streams (transmit,
`receive) between the two layers, and a wait function for timing.
`
`These interfaces are described more precisely in Section 5.
`
`As mentioned in the preface, additional interfaces are necessary to allow a higher
`level network management facility to interact with the Data Link Layer and Physical
`Layer to perform operation, maintenance and planning functions.
`
`4.2 Data Link Layer
`
`The Data Link Layer defines a medium-independent link level communication
`facility, built on the medium-dependent physical channel provided by the Physical
`Layer. It is applicable to a general class of local area broadcast media suitable for
`use with the channel access discipline known as carrier-sense multiple-access with
`collision-detection (CSMA-CD). Compatibility with non-contention media (e.g.,
`switched lines, token-passing rings, etc.), while a worthwhile topic for further
`research, is not addressed in this specification.
`
`The Data Link Layer specified here is intended to be as similar as possible to that
`described in the ISO model. In a broadcast network like the Ethernet, the notion
`of a data link between two network entities does not correspond directly to a
`distinct physical connection. Nevertheless, the two main functions generally
`associated with a data link control procedure are present:
`Data encapsulation
`-framing (frame boundary delimitation)
`-addressing (handling of source and destination addresses)
`-error detection (detection of physical channel transmission errors)
`Link management
`- channel allocation (collision avoidance)
`- contention resolution (collision handling)
`
`This split is reflected in the division of the Data Link Layer into the Data
`Encapsulation sub-layer and the Link Management sub-layer, as shown in Figure 4-
`3.
`
`Page 19
`
`
`
`10
`
`ETHERNET SPECIFIC\ TION: Functional Model of the Ethernet Architecture
`
`Client Layer
`
`Client-to-Data Link Interface
`
`)
`
`... ,.
`
`Transmit
`Data Encapsulation
`
`,,
`
`Data Link Layer
`
`Transmit
`Link Management
`
`Receive
`Data Decapsulation
`
`)'
`
`Receive
`Link Management
`
`)'
`
`Data Link-to-~hysical Interface
`
`"'
`
`Physical Layer
`
`Figure 4-3: Data Link Layer Functions
`
`In tenns of the ISO model, the Ethernet Data Link Layer provides a multi(cid:173)
`endpoint connection between higher-layer entities wishing to communicate. The
`connection provided is called a data link, and is implemented between two or more
`Data Link Layer entities called data link controllers via a Physical Layer connection
`called the physical channel.
`
`4.3 Physical Layer
`
`The Physical Layer specified in this document provides a 10 MBit/sec physical
`channel through a coaxial cable medium. Because one purpose of the layered
`architecture is to insulate the Data Link Layer from the medium-specific aspects of
`the channel,
`the Physical Layer completely specifies the essential physical
`characteristics of the Ethernet, such as data encoding, timing, voltage levels, etc.
`Implementation details are left unspecified, to retain maximum flexibility for the
`implementor. In all cases, the criterion applied in distinguishing between essential
`characteristics and implementation details is guaranteed compatibility: any two
`correct implementations of the Physical Layer specified here will be capable of
`exchanging data over the coaxial cable, enabling communication between their
`
`Page 20
`
`
`
`ETHERNET SPEOFICA TION: Functional Model of the Ethernet Architecture
`
`11
`
`respective stations at the Data Link Layer.
`
`The Physical Layer defmed in this specification performs two main functions
`generally associated with physical channel control:
`Data encoding
`• preamble generation/removal (for synchronization)
`·bit encoding/decoding (between binary and phase-encoded form)
`Channel access
`·bit transmission/reception (of encoded data)
`· carrier sense (indicating traffic on the channel)
`· collision detection (indicating contention on the channel)
`This split is reflected in the division of the Physical Layer into the Data Encoding
`sub-layer and the Channel Access sub-layer, as shown in Figure 4-4.
`
`Data Link Layer
`
`Data Link-to-Physical Interface
`
`~i'
`
`'lr
`
`Transmit
`Data Encoding
`
`,
`
`Physical Layer
`
`Transmit
`Channel Access
`
`Receive
`Data Decoding
`
`'r'
`
`Receive
`Channel Access
`
`)0.
`
`Ethernet Coaxial Cable
`
`Figure 4·4: Physical Layer Functions
`
`Page 21
`
`
`
`12
`
`ETHERNET SPEOFICA TION: Functional Model of the Ethernet Architecture
`
`4.4 Ethernet Operation and the Functional Model
`
`This section provides an overview of frame transmission and reception in terms of
`·the functional model of the architecture. This overview is descriptive, rather than
`definitional; the formal specifications of the operations described here are given in
`Sections 6 and 7.
`
`4.4.1 Transmission Without Contention
`
`When the Client Layer requests the transmission of a frame, the Transmit Data
`Encapsulation component of the Data Link Layer constructs the frame from the
`client-supplied data and appends a frame check sequence to provide for error
`detection. The frame is then handed to the Transmit Link Management component
`for transmission.
`
`Transmit Link Management attempts to avoid contention with other traffic on the
`channel by monitoring the carrier sense signal and deferring to passing traffic.
`When the channel is clear, frame transmission is initiated (after a brief interframe
`delay to provide recovery time for other data link controllers and for the physical
`channel). The Data Link Layer then provides a serial stream of bits to the Physical
`Layer for transmission.
`
`The Data Encoding component of the Physical Layer, before sending the actual bits
`of the frame, sends an encoded preamble to allow the receivers and repeaters along
`the channel to synchronize their clocks and other circuitry.
`It then begins
`translating the bits of the frame into encoded form and passes them to the Channel
`Access component for actual transmission over the medium.
`
`The Channel Access component performs the task of actually generating the
`electrical signals on the medium which represent the bits of the frame.
`Simultaneously, it monitors the medium and generates the collison detect signal,
`which, in the contention-free case under discussion, remains off for the duration of
`the frame.
`
`When transmission has completed without contention, the Data Link Layer so
`informs the Client Layer and awaits the next request for frame transmission.
`
`4.4.2 Reception Without Contention
`
`At the receiving station, the arrival of a frame is first detected by the Receive
`Channel Access component of the Physical Layer, which responds by synchronizing
`with the incoming preamble, and by turning on the carrier sense signal. As the
`encoded bits arrive from the medium, they are passed to the Receive Data
`Decoding component
`
`Page 22
`
`
`
`ETHERNET SPECIFICATION: Functional Model of the Ethernet Architecture
`
`13
`
`Receive Data Decoding translates the encoded signal back into binary data and
`discards the leading bits, up to and including the end of the preamble. It then
`passes subsequent bits up to the Data Link Layer.
`
`Meanwhile, the Receive Link Management component of the Data Link Layer,
`having seen carrier sense go on, has been waiting for the incoming bits to be
`delivered. Receive Link Management collects bits from the Physical Layer as long
`as the carrier sense signal remains on. When the carrier sense signal goes off, the
`frame is passed to Receive Data Decapsulation for processing.
`
`Receive Data Decapsulation checks the frame's destination addr