Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

Industrial Communication Protocols 56.3 Wired Industrial Communications 985

the device level, there are extremely limited resources

(hardware, communications), but at the plant level

powerful computers allow comfortable software and

memory consumption.

Due to the different requirements described above,

there are different types of industrial communication

systems as part of a hierarchical automation system

within an enterprise:

•

Sensor/actuator networks: at the ﬁeld (sensor/actu-

ator) level

•

Fieldbus systems: at the ﬁeld level, collect-

ing/distributing process data from/to sensors/actua-

tors, communication medium between ﬁeld devices

and controllers/PLCs/management consoles

•

Controller networks: at the controller level, trans-

mitting data between powerful ﬁeld devices and

controllers as well as between controllers

•

Wide area networks: at the enterprise level, connect-

ing networked segments of an enterprise automation

system.

Vendors of industrial communication systems offer

a set of ﬁtting solutions for these levels of the automa-

tion/communication hierarchy.

56.3.2 Sensor/Actuator Networks

At this level, several well established and widely

adopted protocols are available:

•

HART (HART Communication Foundation): high-

way addressable remote transducer, coupling analog

process devices with engineering tools [56.5]

•

ASi (ASi Club): actuator sensor interface, coupling

binary sensors in factory automation with control

devices [56.6].

Additionally, CAN-based solutions (CAN in au-

tomation (CIA)) are used for wide-spread application

ﬁelds, coupling decentralized devices with centralized

devices based on physical and MAC layers of the

controller area network [56.7]. Recently, IO Link has

been speciﬁed for bi-directional digital transmission of

parameters between simple sensor/actuator devices in

factory automation [56.8,9].

HART

HART Communication [56.5] is a protocol speciﬁ-

cation, which performs a bi-directional digital trans-

mission of parameters (used for conﬁguration and

parameterization of intelligent ﬁeld instruments by

a host system) over analog transmission lines. The host

Power supply

Resistor

(250 Ohm)

Field device

Handheld terminal

host application

HART interface

(RS 232 or USB)

Fig. 56.3 A HART system with two masters

(http://www.hartcomm.org)

system may be a distributed control system (DCS),

a programmable logic controller (PLC), an asset man-

agement system, a safety system, or a handheld device.

HART technology is easy to use and very reliable.

The HART protocol uses the Bell 202 Frequency Shift

Keying (FSK) standard to superimpose digital commu-

nication signals at a low level on top of the 4–20mA

analog signal. The HART protocol communicates at

1200bps without interrupting the 4–20 mA signal and

allows a host application (master) to get two or more

digital updates per second from a ﬁeld device. As the

digital FSK signal is phase continuous, there is no inter-

ference with the 4–20mA signal. The HART protocol

permits all digital communication with ﬁeld devices in

either point-to-point or multidrop network conﬁgura-

tions. HART provides for up to two masters (primary

and secondary). As depicted in Fig.56.3, this allows

secondary masters (such as handheld communicators)

to be used without interfering with communications

to/from theprimary master (i. e. control/monitoring sys-

tem).

ASi (IEC 62026-2)

ASi [56.6] is a network of actuators and sensors (op-

tical, inductive, capacitive) with binary input/output

signals. An unshielded twisted pair cable for data and

power (max. 2 A; max. 100m) enables the connection

of 31 slaves (max. 124 binary signals of sensors and/or

actuators). This enables a modular design using any

network topology (i. e. bus, star, tree). Each slave can

receive any available address and be connected to the

cable at any location.

AS-Interface uses the APM method (alternating

pulse modulation) for data transfer. The medium access

Part F 56.3

986 Part F Industrial Automation

is controlled by a master–slave principle with cyclic

polling of all nodes. ASi masters are embedded (ASi)

communication controllers of PLCsorPCs, as well as

gateways to other Fieldbus systems. To connect legacy

sensors and actuators to the transmission line, various

coupling modules are used. AS-Interface messages can

be classiﬁed as follows:

•

Single transactions: maximum of 4 bit information

transmitted from master to slave (output informa-

tion) and from slave to master (input information)

•

Combined transactions: more than 4bits of coherent

information are transmitted, composed of a series of

master calls and slave replies in a deﬁned context.

For more details see www.as-interface.com.

56.3.3 Fieldbus Systems

Nowadays, Fieldbus systems are standardized (though

unfortunately not uniﬁed) and widely used in in-

dustrial automation. The IEC 61158 and 61784

standards [56.11, 12] contain ten different Fieldbus

concepts. Seven of these concepts have their own com-

plete protocol suite: PROFIBUS (Siemens, PROFIBUS

International); Interbus (Phoenix Contact, Interbus

Club); Foundation Fieldbus H1 (Emerson, Fieldbus

Foundation); SwiftNet (B. Crowder); P-Net (Process

Data); and WorldFIP (Schneider, WorldFIP). Three

of them are based on Ethernet functionality: high

speed Ethernet (HSE) (Emerson, Fieldbus Foundation);

Ethernet/IP (Rockwell, ODVA); PROFINET/CBA:

(Siemens, PROFIBUS International). The world-wide

Slave 2 Slave 3Slave 3

Cyclic data

exchange

Configuration

parameteriz.

Slave 1

Cyclic access of master 1 Acyclic access of master 2

Cycle

PROFIBUS-DP

Master class 1

PROFIBUS-DP

Master class 2

DP-slave 1 DP-slave 2

Token

DP-slave 3

Fig. 56.4 Proﬁbus medium access control (from [56.10])

leading positions within the automation domain re-

garding the number of installed Fieldbus nodes hold

PROFIBUS and Interbus followed by DeviceNet (Rock-

well, ODVA), which has not been part of the IEC

61158 standard. For that reason, the basic concepts

of PROFIBUS and DeviceNet will be explained very

brieﬂy. Readers interested in a more comprehensive de-

scription are referred to the related web sites.

PROFIBUS

PROFIBUS is a universal ﬁeldbus for plantwide use

across all sectors of the manufacturing and process

industries based on the IEC 61158 and IEC 61784

standards. Different transmission technologies are sup-

ported [56.10]:

•

RS 485: Type of medium attachment unit (MAU)

corresponding to [56.13]. Suited mainly for factory

automation. Technical details see [56.10,13]. Num-

ber of stations: 32 (master stations, slave stations

or repeaters); Data rates: 9.6/19.2/45.45/93.75/

187.5/500/1500/3000/6000/12000kb/s.

•

Manchester bus powered (MBP). Type of MAU

suited for process automation: line, tree, and star

topology with two wire transmission; 31.25kBd

(preferred), high speed variants w/o bus power-

ing and intrinsic safety; synchronous transmission

(Manchester encoding); optional: bus powered de-

vices (≥ 10mA per device; low power option);

optional: intrinsic safety (Ex-i) via additional con-

straints according to the FISCO model. Intrinsic

safety means a type of protection in which a portion

of the electrical system contains only intrinsically

safe equipment (apparatus, circuits, and wiring)

that is incapable of causing ignition in the sur-

rounding atmosphere. No single device or wiring

is intrinsically safe by itself (except for battery-

operated self-contained apparatus such as portable

pagers, transceivers, gas detectors, etc., which

are speciﬁcally designed as intrinsically safe self-

contained devices) but is intrinsically safe only

when employed in properly designed intrinsically

safe system. There are couplers/link devices to cou-

ple MBP and RS485 transmission technologies.

•

Fibre optics (not explained here, see [56.10]).

There are two medium access control (MAC) mech-

anisms (Fig.56.4):

1. Master–master trafﬁc using token passing

2. Master–slave trafﬁc using polling.

Part F 56.3

Industrial Communication Protocols 56.3 Wired Industrial Communications 987

PROFIBUS differentiates between two types of

masters:

1. Master class 1, which is basically a central con-

troller that cyclically exchanges information with

the distributed stations (slaves) at a speciﬁed mes-

sage cycle.

2. Master class 2, which are engineering, conﬁg-

uration, or operating devices. The slave-to-slave

communication is based on the application model

publisher/subscriber using the same MAC mecha-

nisms.

The dominating PROFIBUS protocol is the applica-

tion protocol DP (decentralized periphery), embedded

into the protocol suite (Fig.56.5).

Depending upon the functionality of the masters,

there are different volumes of DP speciﬁcations. There

are various proﬁles, which are grouped as follows:

1. Common application proﬁles (regarding functional

safety, synchronization, redundancy etc.)

2. Application ﬁeld speciﬁc proﬁles (e.g. process

automation, semiconductor industries, motion con-

trol).

These proﬁles reﬂect the broad experience of the

PROFIBUS International organization.

DeviceNet

DeviceNetis a digital,multi-drop network that connects

and serves as a communication network between in-

dustrial controllers and I/O devices. Each device and/or

controller is a node on the network. DeviceNet uses

a trunk-line/drop-line topology that provides separate

twisted pair busses for both signal and power distribu-

tion. The possible variants of this topology are shown

in [56.14].

Thick or thin cables can be used for either trunk-

lines or droplines. The maximum end-to-end network

length varies with data rate and cable thickness. De-

viceNet allows transmission of the necessary power on

the network. This allows devices with limited power

requirements to be powered directly from the net-

work, reducing connection points and physical size.

DeviceNet systems can be conﬁgured to operate in

a master-slave or a distributed control architecture us-

ing peer-to-peer communication. At the application

layer, DeviceNet uses a producer/consumer application

model. DeviceNet systems offer a single point of con-

nection for conﬁguration and control by supporting both

I/O and explicit messaging.

DeviceNet uses CAN (controller area network

[56.7]) for its data link layer, and CIP (common indus-

PROFIBUS DP

PA devices

RIO for PA

SEMI

...

PROFIdrive

Ident systems

Weighing

& dosing

Encoder

IEC 61158/61784

DP-V0...V2

RS 485: NRZ

RS 485-IS

Intrinsic safety

Fiber optics: Glass

Single/Multi mode;

PCF/Plastic fiber

MBP: Manchester Bus

Powered (LP: Low power,

IS: Intrinsic safety

Application

profiles II

Application

profiles I

Communic.

technologies

Transmission

technologies

Descriptions (GSD, EDD)

tools (DTM, Configurators)

Master conform. classes

interfaces, Constraints

Integration

technologies

System

profiles 1... , x

Common application profiles (optional)

PROFIsafe, Time stamp, Redundancy, etc.

Fig. 56.5 PROFIBUS protocol suite (from [56.10])

trial protocol) for the upper network layers. As with all

CIP networks, DeviceNet implements CIP at the ses-

sion (i.e. data management services) layer and above,

and adapts CIP to the speciﬁc DeviceNet technology at

the network and transport layer, and below. Figure 56.6

depicts the DeviceNet protocol suite.

The data link layer is deﬁned by the CAN speci-

ﬁcation and by the implementation of CAN controller

chips. The CAN speciﬁcation [56.7]deﬁnestwobus

states called dominant (logic 0) and recessive (logic

1). Any transmitter can drive the bus to a domi-

nant state. The bus can only be in the recessive state

when no transmitter is in the dominant state. A con-

nection with a device must ﬁrst be established in

order to exchange information with that device. To

establish a connection, each DeviceNet nodewill imple-

ment either an unconnected message manager (UCMM)

CIP common industrial

protocol (IEC 61158)

DeviceNet specification

(IEC 62026)

Device profiles

& application objects

Connection manager

DeviceNet specification

(IEC 62026)

CAN specification

(ISO 11898)

Peer-to-peer, Master-slave,

Multi-master, 64 nodes/network

Controller area network (CAN)

PHY

Explicit

messaging

Implicit

messaging

Fig. 56.6 DeviceNet protocol suite (from [56.14])

Part F 56.3

988 Part F Industrial Automation

or a Group 2 unconnected port. Both perform their

function by reserving some of the available CAN

identiﬁers. When either the UCMM or the Group 2

unconnected port is selected to establish an explicit

messaging connection, that connection is then used to

move information from one node to the other (using

a publisher/subscriber application model), or to estab-

lish additional I/O connections. Once I/O connections

have been established, I/O data may be moved among

devices on the network.

At this point, all the protocol variants of the De-

viceNet I/O message are contained within the 11bit

CAN identiﬁer. CIP is strictly object oriented. Each

object has attributes (data), services (commands), and

behavior (reaction to events). Two different types of

objects are deﬁned in the CIP speciﬁcation: com-

munication objects and application-speciﬁc objects.

Vendor-speciﬁc objects can also be deﬁned by vendors

for situations where a product requires functionality

that is not in the speciﬁcation. For a given device type,

a minimumset of common objectswill be implemented.

An important advantage of using CIP is that for

other CIP-based networks the application data remains

the same regardless of which network hosts the de-

vice. The application programmer does not even need

to know to which network a device is connected.

CIP also deﬁnes device proﬁles, which identiﬁes the

minimum set of objects, conﬁguration options, and the

I/O data formats for different types of devices. Devices

that follow one of the standard proﬁles will have the

same I/O data and conﬁguration options, will respond

to all the same commands, and will have the same be-

havior as other devices that follow that same proﬁle. For

more information on DeviceNet readers are referred to

www.odva.org.

56.3.4 Controller Networks

This network class requires powerful communication

technology. Considering controller networks based on

Ethernet technology, one can distinguish between (re-

lated to the real-time classes, see Sect.56.1):

1. Local soft real-time approaches (real-time class 1)

2. Deterministic real-time approaches (real-time

class 2)

3. Isochronous real-time approaches (real-time

class 3).

The standardization process started in 2004. There

were many candidates to become part of the ex-

tended Fieldbus standard IEC 61158 (edition 4): high

speed Ethernet HSE (Emerson, Fieldbus Foundation);

Ethernet/IP (Rockwell, ODVA); and PROFINET/CBA

(Siemens, PROFIBUS International). Nine Ethernet-

based solutions have been added. In this section a short

survey of the previously mentioned real-time classes

will be given, and two practical examples will be ex-

amined.

Local Soft Real-Time Approaches

(Real-Time Class 1)

These approaches use TCP (UDP)/IP mechanisms over

shared and/or switched Ethernet networks. They can

be distinguished by different functionalities on top of

TCP (UDP)/IP, as well as by their object models and

application process mechanisms. Protocols based on

Ethernet-TCP/IP offer response times in the lower mil-

lisecond range but are not deterministic, since data

transmission is based on the best effort principle. Some

examples are given below.

MODBUS TCP/IP (Schneider) [56.15]. MODBUS is an

application layer messaging protocol for client/server

communication between devices connected via differ-

ent types of buses or networks. Using Ethernet as the

transmission technology, the application layer proto-

col data unit (A-PDU) of MODBUS (function code

and data) is encapsulated into an Ethernet frame. The

connection management on top of TCP/IP controls the

access to TCP.

Ethernet/IP (Rockwell, ControlNet International, Open

DeviceNet Vendor Association) uses a common in-

dustrial protocol CIP [56.16].

In this context, IP stands

for industrial protocol (not for Internet protocol). CIP

represents a common application layer for all physi-

cal networks of Ethernet/IP, ControlNet and DeviceNet.

Data packets are transmitted via a CIP router between

the networks. For the real-time I/O data transfer, CIP

works on top of UDP/IP. For the explicit messaging,

CIP works on top of TCP/IP. The application process is

based on a producer/consumer model.

High Speed Ethernet HSE (Fieldbus Foundation)

[56.17].

A ﬁeld device agent represents a speciﬁc Field-

bus Foundation application layer function (including

Fieldbus message speciﬁcation). Additionally, there are

HSE communication proﬁles to support the different

device categories: host device, linking device, I/O gate-

way, and ﬁeld device. These devices share the tasks

of the system using distributed function block applica-

tions.

Part F 56.3

Industrial Communication Protocols 56.3 Wired Industrial Communications 989

PROFINET (PNO PROFIBUS User Organization, Siemens)

[56.19].

uses object model CBA (component based ar-

chitecture) and DCOM wire protocol with the remote

procedure call mechanisms (DCE RPC) (OSF C 706)

to transmit the soft real-time data. An open source code

and various exemplary implementations/portations for

different operating systems are available on the PNO

web site.

P-Net on IP (Process Data) [56.20]. Based on P-Net

Fieldbus standard IEC 61158 Type 4 [56.11], P-Net on

IP contains the mechanism to use P-Net in an IP en-

vironment. Therefore, P-Net PDUs are wrapped into

UDP/IP packages, which can be routed through IP net-

works. Nodes on the IP network are addressed with two

P-Net route elements. P-Net clients (master) can ac-

cess servers onan IPnetwork without knowing anything

about IP addresses.

All of the above mentioned approaches are able

to support widely used ofﬁce domain protocols, such

as SMTP,SNMP,andHTTP. Some of the approaches

support BOOTP and DHCP for web access and/or for

engineering data exchange. But the object models of the

approaches differ.

Deterministic Real-Time Approaches

(Real-Time Class 2)

These approaches use a middleware on top of the

MAC layer to implement scheduling and smoothing

functions. The middleware is normally represented by

a software implementation. Industrial examples include

the following.

PROFINET (PROFIBUS International, Siemens) [56.19].

This variant of the Ethernet-based PROFINET IO sys-

tem (using the main application model background of

the Fieldbus PROFIBUS DP) uses the object model

IO (input/output). Figure 56.7 roughly depicts the

PROFINET protocol suite, containing the connection

establishment for PROFINET/CBA via connection-

oriented RPC on the left side, as well as for the

PROFINET IO via connectionless RPC on the right

side. The exchange of (mostly cyclic) productive data

uses the real-time functions in the center.

The PROFINET IO service deﬁnition and pro-

tocol speciﬁcation [56.21] covers the communication

between programmable logical controllers (PLCs), su-

pervisory systems, and ﬁeld devices or remote input

and output devices. The PROFINET IO speciﬁcation

complies with IEC 61158, Parts 5 and 6, specially the

Fieldbus application layer (FAL). The PROFINET pro-

Component context

management (ACCO)

DCOM

CO-RPC

TCP

CL-RPC

UDP

Component object model

IO context

management

IEEE 802.3

Real-time

Cyclic

(producer/consumer)

and acyclic services

Connection

establishment

Connection

establishment

IO object model

Fig. 56.7 PROFINET protocol suite of PROFINET (from [56.18])

(active control connection object (ACCO), connection-oriented

(CO), connectionless (CL), remote procedure call (RPC))

tocol is deﬁned by a set of protocol machines. For more

details see [56.22].

Time-Critical Control Network (Tcnet, Toshiba) [56.23].

Tcnet speciﬁes in the application layer a so-called com-

mon memory for time-critical applications, and uses

the same mechanisms as mentioned for PROFINET IO

for TCP(UDP)/IP-based non real-time applications. An

extended data link layer contains the scheduling func-

tionality. The common memory is a virtual memory

globally shared by participating nodes as well as ap-

plication processes running on each node. It provides

a temporal and spatial coherence of data distribution.

The common memory is divided into blocks with

several memory lengths. Each block is transmitted

to member nodes using multicast services, supported

by a publisher node. A cyclic broadcast transmission

mechanism isresponsible for refreshing the data blocks.

Therefore, the common memory consists of dedicated

areas for the transmitting data to be refreshed in each

node. Thus, the application program of a node has quick

access to all (distributed) data.

The application layer protocol (FAL) consists of

three protocol machines: the FAL service protocol ma-

chine (FSPM), the application relationship protocol

machine (ARPM), and the data link mapping protocol

machine (DMPM). The scheduling mechanism in the

data link layer follows a token passing mechanism.

Vnet (Yokogawa) [56.24]. Vnet supports up to 254

subnetworks with up to 254 nodes each. In its applica-

Part F 56.3

990 Part F Industrial Automation

tion layer, three kinds of application data transfers are

supported:

•

A one-way communication path used by an end-

point for inputs or outputs (conveyance paths)

•

A trigger policy

•

Data transfer using a buffer model or a queue model

(conveyance policy).

The applicationlayer FAL contains three types of proto-

col machines: the FSPM FAL service protocol machine,

ARPMs applicationrelationship protocolmachines, and

the DMPM data link layer mapping protocol machine.

For real-time data transfer, the data link layer offers

three services:

1. Connection-less DL service

2. DL-SAP management service

3. DL management service.

Real-time and non real-time trafﬁc scheduling is lo-

cated on top of the MAC layer. Therefore, one or more

time-slots can be used within a macro-cycle (depend-

ing on the service subtype). The data can be ordered

by four priorities: urgent, high, normal, time-available.

Each node has its own synchronized macro-cycle. The

data link layer is responsible for clock synchronization.

Isochronous Real-Time Approaches

(Real-Time Class 3)

The main examples are as follows.

Powerlink (Ethernet PowerLink Standardization

Group (EPSG), Bernecker and Rainer), developed

for motion control [56.25].

Powerlink offers two

modes: protected mode and open mode. The protected

mode uses a proprietary (B&R) real-time protocol

on top of the shared Ethernet for protected subnet-

works. These subnetworks can be connected to an open

standard network via a router. Within the protected

subnetwork the nodes cyclically exchange real-time

data avoiding collisions. The scheduling mechanism

is a time-division scheme. Every node uses its own

time slot [slot communication network management

(SCNM)] to send its data. The mechanism uses a man-

ager node, which acts comparably with a bus master,

and managed nodes act similar to a slave. This mecha-

nism avoidsEthernet collisions. The Powerlink protocol

transfers the real-time data isochronously. The open

mode can be used for TCP(UDP)/IP based applications.

The network normally uses switches. The trafﬁc has

to be transmitted within an asynchronous period of the

cycle.

EtherCAT [EtherCAT Technology Group (ETG), Beckhoff]

developed as a fast backplane communication sys-

tem [56.26].

EtherCAT distinguishes two modes: direct

mode and open mode. Using the direct mode,amas-

ter device uses a standard Ethernet port between the

Ethernet master and an EtherCAT segment. EtherCAT

uses a ring topology within the segment. The medium

access control adopts the master/slave principle, where

the master node (typically the control system) sends the

Ethernet frame to the slavenodes (Ethernet device). One

single Ethernet device is the head node of an Ether-

CAT segment consisting of a large number of EtherCAT

slaves with their own transmissiontechnology. The Eth-

ernet MAC address of the ﬁrst node of a segment is

used for addressing the EtherCAT segment. For the seg-

ment, special hardware can be used. The Ethernet frame

passes each node. Each node identiﬁes its subframe and

receives/sends the suitable information using that sub-

frame. Within the EtherCAT segment, the EtherCAT

slave devices extract data from and insert data into these

frames. Using the open mode, one or several Ether-

CAT segments can be connected via switches with one

or more master devices and Ethernet-based basic slave

devices.

PROFINET IO/Isochronous Technology (PROFIBUS User

Organization, Siemens) developed for any indus-

trial application [56.27].

PROFINET IO/Isochronous

Technology uses a middleware on top of the Ethernet

MAC layer to enable high-performance transfers, cyclic

data exchange and event-controlled signal transmission.

The layer 7 functionality is directly linked to the mid-

dleware. The middleware itself contains the scheduling

and smoothing functions. This means that TCP/IP does

not inﬂuence the PDU structure. A special Ethertype is

used to identify real-time PDUs (only one PDU type for

real-time communication). This enables easy hardware

support for the real-time PDUs. The technical back-

ground is a 100Mbps full duplex Ethernet (switched

Ethernet). PROFINET IO addsan isochronousreal-time

channel to the RT channels of real-time class 2 op-

tion channels. This channel enables a high-performance

transfer of cyclic data in an isochronous mode [56.28].

Time synchronization and node schedulingmechanisms

are located within and on top of the Ethernet MAC

layer. The offered bandwidth isseparated forcyclic hard

real-time and soft/non real-time trafﬁc. This means that

within a cycle there are separatetime domains for cyclic

hard real-time, for soft/non real-time over TCP/IP traf-

ﬁc, and for the synchronization mechanism, see also

Fig.56.8.

Part F 56.3

Industrial Communication Protocols 56.4 Wireless Industrial Communications 991

cRT

aRTcRT non RT aRT non RT

31.25µs ≤ T

sendclock

≤ 4 ms

60%

sendclock+1

sendclock

Fig. 56.8 LMPM MAC access used in PROFINET

IO [56.22] (cyclic real-time (cRT), acyclic real-time (aRT),

nonreal-time (non RT), medium access control (MAC),

link layer mapping protocol machine (LMPM))

The cycle time should be in the range of 250μs

(35 nodes) up to 1 ms (150 nodes) when simultane-

ously TCP/IP trafﬁc of about 6Mbps istransmitted. The

jitter will be less than 1μs. PROFINET IO/IRT uses

switched Ethernet (full duplex). Special four-port and

two-port switch ASICs have been developed and allow

the integration of the switches into the devices (nodes)

substituting the legacy communication controllers of

Fieldbus systems. Distances of 100m per segment

(electrical) and 3km per segment (ﬁber-optical) can be

bridged.

Ethernet/IP with Time Synchronization (ODVA, Rock-

well Automation).

Ethernet/IP with time synchroniza-

tion [56.29], an extension of Ethernet/IP,usestheCIP

Synch protocol to enable the isochronous data trans-

fer. Since the CIP Synch protocol is fully compatible

with standard Ethernet, additional devices without CIP

Synch features can be used in the same Ethernet sys-

tem. The CIP Synch protocol uses the precision clock

synchronization protocol [56.30] to synchronize the

node clocks using an additional hardware function.

CIP Synch can deliver a time-synchronization accuracy

of less than 500ns between devices, which meets the

requirements of the most demanding real-time applica-

tions. The jitter between master and slave clocks can be

less than 200ns.

SERCOS III, (IG SERCOS Interface e.V.). A SERCOS

network [56.31], developed for motion control, con-

sists of masters and slaves. Slaves contain integrated

repeaters, which have a constant delay time T

rep

(in-

put/output). The nodes are connected via point-to-point

transmission lines. Each node (participant) has two

communication ports, which are interchangeable. The

topology can be either a ring or a line structure. The

ring structure consists of a primary and a secondary

channel. All slaves work in forwarding mode. The re-

dundancy provided by the ring structure prevents any

downtime caused by a broken cable. The line structure

consists of either a primary or a secondary channel.

The last physical slave performs the loopback func-

tion. All other slaves work in forwarding mode. No

redundancy against cable breakage is achieved. It is

also possible to insert and remove slaves during oper-

ation (hot plug). This is restricted to the last physical

slave.

56.4 Wireless Industrial Communications

56.4.1 Basic Standards

Wireless communication networks are increasingly

penetrating the application area of wired communica-

tion systems. Therefore, they have been faced with the

requirements of industrial automation. Wireless tech-

nology has been introduced in automation as wireless

local area networks (WLAN) and wireless personal

area networks (WPAN). Currently, the wireless sensor

networks (WSN) are under discussion especially for

process automation. For speciﬁc application aspects of

wireless communications see Chap. 13 on Communica-

tion in Automation, Including Networking and Wireless,

and Chap.20 on Sensors and Sensor Networks. The ba-

sic standards are the following:

•

Mobile communications standards: GSM, GPRS,

and UMTS wireless telephones (DECT)

•

Lower layer standards (IEEE 802.11: Wireless

LAN [56.32], and 802.15 [56.33]: personal area

networks) as a basis of radio-based local networks

(WLANs, Pico networks and sensor/actuator net-

works)

•

Higher layer standards (application layers on top

of IEEE 802.11 and 802.15.4, e.g. WiFi, blue-

tooth [56.34], wireless HART, and ZigBee [56.35])

•

Proprietary protocols for radio technologies (e.g.

wireless interface for sensors and actuators (WISA)

[56.36])

•

Upcoming radio technologies such as ultra wide

band (UWB) and WiMedia

Part F 56.4

992 Part F Industrial Automation

For more detailed information and survey see [56.37–

41].

56.4.2 Wireless Local Area Networks (WLAN)

The term WLAN refers to a wireless version of the Eth-

ernet used to build computer networks for ofﬁce and

home applications. The original standard (IEEE802.11)

speciﬁed an infrared, a direct sequence spread spec-

trum (DSSS) and a frequency hopping spread spectrum

(FHSS) physical layer. There is an approval for WLAN

to use special frequency bands, however it has to

share the medium with other users. The WiFi Alliance

was founded to assure interoperability between WLAN

clients and access points of different vendors. There-

fore, a certiﬁcation procedure and a WiFi logo are

provided. WLANs use a licence-free frequency band

and no service provider is necessary.

WLAN is a mature technology and it is imple-

mented in PCs, laptops, and PDAs. Modules for em-

bedded systems development are also available. WLAN

can be used almost world wide. Embedded WLAN de-

vices need a powerful microcontroller. WLAN enables

wireless access to Ethernet based LANs and is helpful

for the vertical integration in an automated manufactur-

ing environment. It offers high speed data transmission

that can be used to transmit productive data and man-

agement data in parallel. The WLAN propagation

characteristics ﬁt into a number of possible automation

applications. WLAN enables more ﬂexibility and a cost

effective installation in automation associated with mo-

bility and localization. The transition to Ethernet is

simple and other gateways are possible. The largest part

of the implementation is achieved in hardware; however

improvements can be made above the MAC layer.

56.4.3 Wireless Sensor/Actuator Networks

Various wireless sensor network (WSN) concepts are

under discussion, especially in the area of industrial

automation. Features such as time synchronized opera-

tion, frequency hopping, self-organization (with respect

to star, tree, and mesh network topologies), redundant

routing, and secure data transmission are desired. Inter-

esting surveys on this topic are available in [56.41–44].

Process automation requirements can be generally ful-

ﬁlled by two mesh network technologies:

•

ZigBee (ZigBee Alliance) [56.35]

•

Wireless HART [56.45,46].

Both technologies use the standard IEEE 802.15.4

(2003) low-rate wireless personal area network

(WPAN) [56.33], specifying the physical layerand parts

of the data link layer (medium access control).

ZigBee

ZigBee distinguishes between three device types:

•

Coordinator ZC: root ofthe network tree,storing the

network information and security keys. It is respon-

sible for connection the ZigBee network to other

networks.

•

Router ZR: transmits data of other devices.

•

End device ZED: automation device (e.g. sensor),

which can communicate with ZR and ZC, but is

unable to transmit data of other devices.

An enhanced version allows one to group devices

and to store data for neighboring devices. Addition-

ally, to save energy, there are full-function devices and

reduced-function devices. The ZigBee application layer

(APL) consists of three sublayers: application support

layer (APS) (containing the connection lists of the con-

nected devices), an application framework (AF), and

Zigbee device objects (ZDO) (deﬁnition of devices

roles, handling of connection requests, and establish-

ment of communication relations between devices).

For process automation, the ZigBee application

model and the ZigBee proﬁles are very interesting. The

application functions are represented by application ob-

jects (AO), and the generic device functions by device

objects (DO). Each object of a ZigBee proﬁle can con-

tain one or more clusters and attributes, transferred to

the target AO (in the target device) directly or to a co-

ordinator, which transfers them to one or more target

objects.

WirelessHART

Revision 7 of HART protocol includes the speci-

ﬁcation of WirelessHART [56.46]. The mesh type

network allows the use of redundant communication

paths between the radio-based nodes. The temporal be-

havior is determined by the time synchronized mesh

protocol (TSMP) [56.47, 48]. TSMP enables a syn-

chronous operation of the network nodes (called motes)

based on a time slot mechanism. It uses various radio

channels (supported by the MAC layer) for an end-

to-end communication between distributed devices. It

works comparably with a frequency hopping mech-

anism missed in the basic standard IEEE 802.15.4.

Part F 56.4

Industrial Communication Protocols 56.5 Wide Area Communications 993

TSMP supports star, tree, as well as mesh topolo-

gies. All nodes have the complete routing function

(contrary to ZigBee). A self-organization mechanism

enables devices to acquire information of neighbor-

ing nodes and to establish connections between them.

The messages have their own network identiﬁer. Thus,

different networks can work together in the same ra-

dio area. Each node has its own list of neighbors,

which can be actualized when failures have been

recognized.

To support security, TSMP uses mechanisms for

encryption (128bit symmetric key), authentiﬁcation

(32bit MIC for source address), and integrity (32bit

MIC for message content). Additionally, the frequency

hopping mechanism improves the security features. For

detailed information see [56.46].

56.5 Wide Area Communications

With the application of remote automation mechanisms

(remote supervisory, operation, service) using wide

area networks, the stock of existing communication

technology becomes broader and includes the follow-

ing [56.18]:

•

All appearances of the Internet (mostly supporting

best effort quality of services)

•

Public digital wired telecommunication systems: ei-

ther line-switched [e.g. integrated services digital

network (ISDN)] or packet-switched [such as asym-

metric/symmetrical digital subscriber line (ADSL,

SDSL)]

•

Public digital wireless telecommunication systems

(GSM-based, GPRS-based, UMTSbased)

•

Private wireless telecommunication systems, e.g.

trunk radio systems.

The transition between different network technolo-

gies can be made easier by using multiprotocol label

switching (MPLS) and synchronous digital hierarchy

(SDH). There are several private protocols (over leased

lines, tunneling mechanisms, etc.) that have been used

in the automation domain using these technologies.

Most of the wireless radio networks can be used in non

real-time applications, some of them in soft real-time

applications; however industrial environments and in-

dustrial, scientiﬁc, and medical (ISM) band limit the

applications. Figure 56.9 depicts the necessary remote

channels.

The end-to-end connection behavior via these

telecommunication systems depends on the recently of-

fered quality of service (QoS). It strongly limits the

use of these systems within the automation domains.

Therefore, the following application areas have to be

distinguished:

Non-Real-Time Communication in Automation

•

Non real-time communication (Standard IT: up-

load/download, SNMP) with lower priority to real-

time communication: for conﬁguration, diagnostics,

automation-speciﬁc up/download.

•

Manufacturing-speciﬁc functions, context manage-

ment, establishment of application relationships and

connection relationships to conﬁgure IO devices,

application monitoring to read status data (diagnos-

tics, I&M), read/write services (HMI, application

program), open loop control.

The automation domain has the following im-

pact on non real-time WA N connections: addressing

between multiple distributed address spaces, and redun-

dant transmission for minimized downtime to ensure its

availability for a running distributed application.

Real-Time Communication in Automation

•

Cyclic real-time communications (i.e. PROFINET

IO data) for closed loop control and acyclic

alarms (i.e. PROFINET IO alarms) as major

manufacturing-speciﬁc services

•

Transfer (and addressing) methods for RT data

across WA N can be distinguished as follows:

Communication channel CRs

IO deviceIO controller

Configuration data

record data

Real-timeNon real-time

devdev

AlarmsIO data

Fig. 56.9 Remote communication channels over WA N (input/

output (IO); communication relation (CR))

Part F 56.5

994 Part F Industrial Automation

– MAC based: Tunnel (real-time class 1, partially

real-time class 2 for longer send intervals, e.g.

512ms), clock across WA N and reserved send

phase for real-time transmission

– IP based: Real-time over UDP (routed); web

services based [56.49,50].

The automation domain has the following impact on

real-time WA N connections: a constant delay-sensitive

and jitter-sensitive real-time base load (e.g. in LAN:up

to 50% bandwidth reservation for real-time transmis-

sion).

To use a wide area network for geographically

distributed automation functions, the following basic

design decisions were made following the deﬁnitions in

Sect.56.2:

•

A virtual automation network (VAN)isanin-

frastructure for standard LAN-based distributed

industrial automation concepts (e.g. PROFINET or

other) in an extended environment. The productive

automation functions (applications) are described

by their object models used in existing industrial

communications. The application service elements

(ASEs), as they are speciﬁed in the IEC 61158 stan-

dard, can additionally be used.

•

The establishment of the end-to-endconnections be-

tween distributed objects within a heterogeneous

network is based on web services. Once this con-

nection has been established, the runtime channel

between these objects is equivalent to the runtime

channel within the local area by using PROFINET

(or other) runtime mechanisms.

•

The VAN addressing scheme is based on names

to avoid the use of IP and MAC addresses during

establishment of the end-to-end path between logi-

cally connected applications within a VAN domain.

Therefore, the IP and MAC addresses remain trans-

parent to the connected application objects.

•

Since there is no new Fieldbus or real-time Eth-

ernet protocol, no new speciﬁed application layer

is necessary. Thus, the well-tried models of indus-

trial communications (as they are speciﬁed in the

IEC 61158 standard) can be used. Only the ad-

ditional requirements caused by the inﬂuence of

wide area networks have to be considered and they

lead to additional functionality following the above-

mentioned design guidelines.

Most of the WA N systems that offer quality-of-

service (QoS) support cannot provide real guarantees,

and this strongly limits the use of these systems within

the automation domain. To guarantee a deﬁned QoS

for data transmission between two application access

points via a wide area network, written agreements

between customer and service provider [service level

agreements (SLA)] must be contracted. In cases where

the provider cannot deliver the promised QoS, an alter-

native line must be established to hold the connection

for the distributed automation function (this operation is

fully transparent to the application function). This line

should be available from another provider, independent

from the currently used provider. The automation de-

vices (so called VA N access points (VA N-APs)] should

support functions toswitch (eithermanually or automat-

ically) to an alternative line [56.51].

There are different mechanisms to realize a connec-

tion between remote entities:

•

The VAN switching connection:thelogical connec-

tion between two VA N-APs over a WA N or a public

network. One VAN switching connection owns one

or more communication paths. VA N switching line

is deﬁned as one physical communication path be-

tween two VA N-APs over a WA N or a public

network. The endpoints of a switching connection

are VA N-APs.

•

The VAN switching line:thephysical communica-

tion path between two VA N-APs over a WA N or

a public network. A VAN switching line has its

provider and own QoS parameter. If a provider of-

fers connections with different warranted QoS each

of these shall be a new VA N switching line.

•

VAN switching endpoint access: the physical com-

munication path between one VAN-AP and a WA N

or a public network.This is a newly introducedclass

for using the switching application service elements

of virtual automation networks for communication

via WA N or public networks.

These mechanisms are very important for the concept

of VA Ns using heterogeneous networks for automa-

tion. Depending on the priority and importance of the

data transmitted between distributed communications

partners, the kind of transportation service and com-

munication technology is selected based on economical

aspects. The VA N provider switching considers the fol-

lowing alternatives:

•

Use case 1: For packet-oriented data transmission

via public networks a connection from a corre-

sponding VAN-AP to the public network has to be

established. The crossover from/to the public net-

work is represented by the VA N switching endpoint

Part F 56.5