IEEE 802.3-2000标准概述：CSMA/CD与光纤测试方法

需积分: 9 12 浏览量更新于2024-07-21 收藏 8.1MB PDF 举报

IEEE 802.3标准，起源于1980年代，是一个广泛应用于局域网（LAN）技术的通信协议，其核心特点是采用了载波监听多路访问（Carrier Sense Multiple Access with Collision Detection, CSMA/CD）机制。该标准随着时间发展，历经多次修订，直至2000年发布的802.3-2000版本，涵盖了10Mbps以太网（10BASE-T）、100Mbps以太网（100BASE-TX）和1Gbps以太网（1000BASE-T）等不同速率的网络技术。这些速率的实现依赖于物理层的标准，确保了数据传输的可靠性和效率。在802.3-2000标准中，不仅规定了网络的物理层规范，还涉及到了与光纤通信相关的测试和测量方法。例如，Annex A（非强制性参考材料）中列出了多个标准，如ANSI/EIA 364A关于低频电气连接器测试、ANSI/EIA/TIA 455系列标准，这些标准详细规定了光纤连接器的插入损耗测试、多模光纤的信息传输能力测量、脉冲失真测量以及光缆和光纤的绝对光功率测试等。这些测试对于确保设备兼容性和优化光纤网络性能至关重要。此外，标准中还包含了针对过度填充（overfilled launching condition）条件下的模式混淆器要求，以及使用光学时域反射仪（Optical Time Domain Reflectometer, OTDR）进行光纤点缺陷测量的方法。同时，对激光的光谱特性进行了规范化，如ANSI/EIA/TIA 455-127标准，这对于激光源的稳定性和发射性能有着严格的规定。 802.3-2000标准不仅定义了局域网的通信协议，还在光纤通信的质量控制、性能评估和测试方法等方面提供了详尽的技术指导，是现代信息技术基础设施的重要组成部分。了解并遵循这个标准，对于设计、部署和维护高质量的以太网网络具有重要意义。

IEEE

Std 802.3, 2000 Edition LOCAL AND METROPOLITAN AREA NETWORKS:

B.2 System parameters and budgets for 1BASE5

B.2.1 Delay budget

The successful interconnection of multivendor system components mandates that the values for bits lost and

signal delays be allocated fairly and realistically among the various system elements. The following table

summarizes and indicates the derivation of some of the delays speciﬁed in 12.9. The breakdowns shown for

the parameters are illustrative only; implementors are free to make other allocations of delay within a device

so long as the speciﬁcations of 12.9 are not violated.

Component Delay (BT)

DTE Initial Transmit Delay (see 12.9.2) 3

DTE Deference Delay (see 12.9.2) 21

unsquelch 3

Carrier detect 5

MAC detects carrier and defers 10

DTE Initial Transmit Delay 3

DTE Collision Shutdown Delay (see 12.9.2) 58

detect CP and report SIGNAL_ERROR 10

detect SIGNAL_ERROR and start jamming 16

jamSize 32

Medium Transit Delay (see 12.9.3) 4

Special Link Transit Delay (see 12.9.4) 15

Hub Startup Delay (see 12.9.5) 12

unsquelch 4

half ﬁll FIFO 6

analogue of DTE Initial Transmit Delay 3

Hub Idle Collision Startup Delay (see 12.9.5)

(same as Hub Startup Delay)

Hub Transit Delay (see 12.9.5) 9

half ﬁll FIFO 6

analogue of DTE Initial Transmit Delay 3

Hub Delay Stretch/Shrink (see 12.9.5)

((preamble + <sfd> + maxFrameSize) · 0.01% · 2)

Hub Collision Detect Delay (see 12.9.5) 21

unsquelch 3

detect collision 6

Hub Transit Delay 9

First CVL or CVH may be preceded by CD0s and CD1s 3

Hub Active Collision Startup Delay (see 12.9.5) 12

Hub Transit Delay 9

First CVL or CVH may be preceded by CD0s and CD1s 3

Hub Collision Shutdown Delay (downward) (see 12.9.5)

(same as Hub Transit Delay)

Hub Collision Shutdown Delay (upward) (see 12.9.5) 25

detect loss of carrier 20

clear FIFO, if necessary 2

analogue of DTE Initial Transmit Delay 3

IEEE

CSMA/CD Std 802.3, 2000 Edition

B.2.2 Minimum frame length determination

The minimum frame length for 1BASE5 is determined using the values speciﬁed in 4.4.2.2 and 12.9, applied

to the following (worst) case:

a) DTE 1, connected to Hub 1 at a network extremity, transmits a message upward toward Hub 5.

b) There is a special link in the path between Hub 1 and Hub 5.

c) DTE 2, also connected to Hub 1, transmits, just missing deferring to the downward signal from DTE

1 that was wrapped around at Hub 5.

d) DTE 3, also connected to Hub 1, receives the transmission from DTE 1.

e) Hub 1 generates CP, which travels up and then back down the network to inform DTE 1 and DTE 2

that a collision has occurred on their messages.

f) DTE 1 and DTE 2 continue to transmit until they have received CP, reacted to it, and completed their

jams.

g) DTE 3 continues to receive until the end of CP.

The minimum frame length must allow both of the following conditions to be met:

— DTE 1 is still sending when CP is received and recognized.

— DTE 3 can discard the message fragment it receives because it is too short.

The minimum frame length must exceed both the maximum number of bits sent before recognizing CP (391

– jamsize = 359) and the maximum collision fragment size (471), as computed above. The 1BASE5 system

Event Bits Total

DTE 1 → DTE 2

DTE Initial Transmit Delay

8 · Medium Transit Delay

2 · Special Link Transit Delay

10 · Hub Startup Delay

DTE Deference Delay

120

185

206

DTE 2 → HUB 1 CP

Medium Transit Delay

Hub Collision Detect Delay

210

231

HUB 1 CP → HUB 5 CP

3 · Medium Transit Delay

Special Link Transit Delay

4 · max(Hub Startup Delay,

Hub Active Collision Startup Delay,

Hub Idle Collision Startup Delay)

243

358

306

HUB 5 CP → DTE 1 receives CP

5 · Hub Active Collision Startup Delay

4 · Medium Transit Delay

Special Link Transit Delay

366

382

397

DTE 1 receives CP → DTE 1 stops transmitting DTE Collision Shutdown Delay 58 455

COMPUTATION OF MINIMUM FRAME SIZE

original preamble + <sfd>

5 · (Hub Collision Shutdown Delay (upward) – Hub Transit Delay)

5 · (Hub Collision Shutdown Delay (downward) – Hub Transit Delay)

–64

391 =

data bits

transmitted

471

Tiny fraction of Hub Delay Stretch/Shrink 0 471 =

data bits

received

IEEE

Std 802.3, 2000 Edition LOCAL AND METROPOLITAN AREA NETWORKS:

value for minimum frame length has been set at 512 bits, which exceeds both of these values with a margin

for error.

B.2.3 Jitter budget

The total edge jitter of the signals on each link must be limited to allow proper decoding at the receiver. The

following budget has been used to allocate jitter to the indicated components that contribute to the total jitter

on each link:

The cable intersymbol interference and reﬂection allowances form the basis for the limit speciﬁed in

12.7.2.3; the reﬂection component is sufﬁcient to allow a single 20 Ω impedance mismatch anywhere along

a cable segment. The receiver-mismatch allowance is derived from the reﬂection attenuation speciﬁed in

12.5.3.2.4. The total forms the basis for the speciﬁcation in 12.5.3.2.2.

The remainder of the jitter that can be tolerated by the Manchester decoder in a receiver is reserved to allow

for distortion of the signal due to noise, receiver threshold offset, receiver skew, and receiver sampling tim-

ing error.

A simple clocked receiver/decoder with an 8 MHz sampling rate (the worst case allowed for in the design of

this standard), can achieve proper decoding with up to ± 125 ns of jitter between two edges, which is equiv-

alent to ±62.5 ns on each edge. Other receiver designs may tolerate more edge jitter. For example, a 6 MHz

sampling rate would allow up to ±83.33 ns of jitter on each edge and a 16 MHz sampling rate allows up to

±93.75 ns of jitter.

It may be necessary to use a low-pass ﬁlter as part of the receiver to reduce the noise level seen by that

receiver (see 12.7.4 for a description of the noise environment). A ﬁlter that reduces the noise may also have

an effect on the amplitude and edge rate of the received signal. The ﬁltered signal’s edge rate near the zero-

crossing is used in the critical translation from mV of noise and receiver offset into ns of jitter.

An example receiver design using an 8 MHz sampling rate and a 2 MHz Butterworth input ﬁlter might be

based on the following jitter budget:

Component

Jitter

(ns)

Transmitter skew ±10

Cable intersymbol interference 9

Cable reﬂections 8

Reﬂections due to receiver termination mismatch 5

Total

±32

Component

Jitter

(ns)

Input jitter (from above)

±32

Noise and receiver threshold offset

19.5

Receiver skew (analog)

Receiver skew (digital)

Total

±62.5

IEEE

CSMA/CD Std 802.3, 2000 Edition

The two primary contributors to noise in a 1BASE5 cable are self-crosstalk and impulse noise (see 12.7.4).

Because it is unlikely that both will be present at their 1% worst-case levels on any particular cable, the

required bit error rate attributable to each source can be set at half of the one in 10

error rate required by

12.5.3.2.6.

Crosstalk noise is speciﬁed to be no more than 105 mV (peak) through a 2 MHz ﬁlter (see 12.7.4.2).

Because crosstalk is present for the entire transmission of a packet, some crosstalk will coincide with the

most sensitive part of the received signal. Therefore, the receiver must operate without error in the presence

of this 105 mV of noise.

Impulse noise has a peak amplitude of 170 mV for ≤0.005 counts/s through the 2 MHz ﬁlter (see 12.7.4.1).

This threshold does not directly correlate to jitter, however, because the derivation of the 62.5 ns jitter toler-

ance for an 8 MHz clock assumed worst-case sampling error. Assuming a random phasing of the sampling

clock to the received signals, it can be shown that the 170 mV of noise is equivalent to a level of 85 mV with

a worst-phase clock.

Jitter due to noise should be computed using the larger of the above two levels. The 105 mV for crosstalk

noise, therefore, should be added to 50 mV for receiver threshold offset and the result should be divided by

the edge rate of the ﬁltered signal near the zero-crossing (7.9 mV/ns for the 2 MHz ﬁlter), yielding the

19.5 ns indicated above.

B.3 Example crosstalk computation for multiple disturbers, balanced-pair

cable

A method for computing multiple-disturber, near end, crosstalk attenuation (MDNEXT) into each 1BASE5

pair is speciﬁed in 12.7.3.2. This annex provides example computations of MDNEXT using that method

when only the distribution of X

is known.

The single-disturber probability distribution curve (labelled “1”) shown in Figure B–6 is based on actual

measurement of 25-pair, 24-gauge, unshielded, twisted pair cable. The remaining probability distribution

curves (labelled with the number of disturbing pairs) were computed using Monte Carlo simulation. To com-

pute each sample MDNEXT

for N disturbers, N values of crosstalk attenuation (X

) were chosen from the

single-disturber distribution and N values of crosstalk phase (

) were chosen from a uniform distribution

between 0 and 2

rad. These values were then used with the following equations to compute MDNEXT

/20–()

1iN≤≤

cosθ

∑

/20–()

1iN≤≤

sinθ

∑

MDNEXT

10log

+()=

剩余236页未读，继续阅读

zxtonl

粉丝: 0
资源: 6

IEEE 802.3-2000标准概述：CSMA/CD与光纤测试方法

IEEE 802.3-2000标准：光纤介质与MAU规范

IEEE 802.3-2018标准压缩包介绍

IEEE 802.3-2018标准：以太网技术全面解析

802.3-2000标准文档3

802.3-2000标准文档2

802.3-2000标准文档4

IEEE 802.3-2000标准

ieee802.3-2000标准.rar

802.3-2012标准高清PDF带书签

IEEE Std 802.3-2012-section5

最新资源