LIN Standard - Networked Embedded Systems - page 401

• Self-synchronization without a quartz or ceramics resonator in the slave nodes
• Deterministic signal transfer entities, with signal propagation time computable in
advance
• Signals-based API
A LIN network comprises one master and one or more slave nodes. he medium access is controlled
by a master node—no arbitration or collision management in the slaves is required. Worst-case
latency of signal transfer is guaranteed.
Today's LIN . standard is a matured specification, which has evolved through several versions
(LIN ., LIN ., and LIN .). LIN . took a major technology step by introducing new features
such as configuration of slave nodes. The LIN . standard is reusing the experiences gained from
useofLIN.,andbecometheconsensusofallmajorcarOEMs.heLIN.standardcompleted
definition of the diagnostic functionality (started in LIN .) and made the configuration process
more efficient.
LINnodesproducedtodayarebasedonLIN.,LIN.,LIN.,andLIN..hismeansthat
backwards compatibility is an issue. he LIN standard solves this by ensuring strict backwards com-
patibility in LIN masters, and by introducing new features disjoint to older features. A LIN . master
node can without much effort support heterogeneous clusters containing LIN . to LIN . slave
nodes. New features added in later versions of the standard do not conflict (disjoint) with obsolete
or older features.
17.4 Some LIN Basics
LIN is a low-cost, single wire network. The starting point of the physical layer design was the ISO
 standard. In order to meet EMC requirements the slew rates are controlled. The protocol is a
simple master slave protocol based on the common UART format. In order to enable communica-
tion between nodes clocked by low-cost RC-Oscillators, synchronization information is transmitted
by the master node on the bus. Slave nodes will synchronize with the master clock, which is regarded
tobeaccurate.hespeedoftheLINnetworkisuptokbps,andthetransmissionisprotected
by a checksum. The LIN Protocol is message-identifier based. The identifiers do not address nodes
directly, but denote the meaning of the messages. his way any message can have multiple destina-
tions (multicasting). he master sends out the message header consisting of a Synchronization Break
(serving as a unique identifier for the beginning of the frame), a synchronization field carrying the
clock information, and the message identifier, which denotes the meaning of the message.
Upon reception of the message identifier the nodes on the network will know exactly what to do
with the message. One of the nodes sends out the message response and the others either listen or
do not care. Messages from the master to the slave(s) are carried out in the same manner—in this
case the slave task incorporated into the master node sends the response.
LIN messages are scheduled in a time-triggered fashion. This provides a model for the accurate
calculation of latency times—thus supporting fully predictable behavior. Since the master sends out
the headers, it is in complete control of the scheduling and is also able to swap between a set of
predefined schedule tables, according to the specific requirements/modes of the applications running
in the subsystem.
17.5 LIN Physical Layer
The transport medium is a single line, wired-AND bus being supplied via a termination resistor
from the positive battery node (V BAT Nominally  V). he bus line transceiver is an enhanced ISO
 implementation. he bus can take two complementary logical values: the dominant value with