1:1 wire cable
1:1 wire cables or straight through cables are required for connecting ETHERNET components over copper cable. In general 1:1 wire cables are required for connections between terminal devices such as SPS, HMI, etc. and network components such as hubs, switches, etc.
10Base2
Standard for data transmission of 10 Mbit/s ETHERNET on thin coaxial cables (thin wire, cheapernet). Segment length max. 185 m.
10Base5
Standard for data transmission of 10 Mbit/s ETHERNET on coaxial cables (thick wire, yellow cable). Segment length max. 500 m.
10Base-FL
Standard for data transmission of 10 Mbit/s ETHERNET on fiber optic cables. Each connection is created with 2 fibers, in each case, one fiber for "Transmit Data" and another one for "Receive Data".
10Base-T
Standard for data transmission of 10 Mbit/s ETHERNET on unshielded twisted pair cables (category 3, 4 or 5). Each connection is created with 2 wire pairs, in each case with one wire pair for "Transmit Data" and another one for "Receive Data".
100Base-FX
Standard for data transmission of 100 Mbit/s ETHERNET on fiber optic cables. Each connection is created with 2 fibers, in each case, one fiber for "Transmit Data" and another one for "Receive Data".
100Base-TX
Standard for data transmission of 100 Mbit/s ETHERNET on twisted pair cables (category 5). Each connection is created with 2 wire pairs, in each case with one wire pair for "Transmit Data" and another one for "Receive Data".
1000Base-LX
Standard for data transmission of 1000 Mbit/s ETHERNET on fiber optic cables for a wavelength of 1300 nm. Each nnection is created with 2 fibers, in each case, one fiber for "Transmit Data" and another one for "Receive Data".
1000Base-SX
Standard for data transmission of 1000 Mbit/s ETHERNET on fiber optic cables for a wavelength of 850 nm. Each onnection is created with 2 fibers, in each case, one fiber for "Transmit Data" and another one for "Receive Data".
1000Base-TX
Standard for data transmission of 1000 Mbit/s ETHERNET on twisted pair cables (category 5e). Each connection is created with 4 wire pairs, in each case with all 4 pairs being used for "Transmit Data" and "Receive Data" simultaneously.
A
Aging
Process for the updating of data, especially of address tables. An address is marked as "old" after the expiry of the certain period of time and the it is deleted at the time of the next pass if it is not detected at a port once again.
ARP
A ddress R esolution P rotocol A protocol that asks for the relevant MAC address on the basis of an IP address. Each device manages its own dynamic ARP table. If the MAC address of a participant to whom a message is to be sent is not present in the table, the device first sends an ARP request. This message is read by all stations. The device whose IP address is contained in the request sends an ARP reply with its MAC address. The participant making the request completes his ARP table with this MAC address and is then able to transmit the message.
AUI
A ttachment U nit I nterface Designation of an ETHERNET interface with a 15-pole Sub-D plug connector.
Autocrossing
A function that enables automatic crossing of transmission and reception lines on twisted pair interfaces. Switches that support this function can be connected to each other over a 1:1 wire cable instead of a crossover cable.
Autonegotiation
A protocol in Fast ETHERNET with which the participant devices agree a common transmission mode before the actual data transmission (100 Mbit/s or 10 Mbit/s, full duplex or half duplex).
Autopolarity
A function of devices with 10Base-T or 100Base TX interface for automatic correction of wiring errors in twisted pair cables that lead to a polarity reversal of the data signals.
Autosensing
A function that enables a device to automatically detect the data rate (10 Mbit/s or 100 Mbit/s) and to transmit and receive at this data rate.
Auto-MDI/MDI-X
See autocrossing.
B
Backpressure
A function that simulates a collision in half duplex operation by generating a jam signal.
Bandwidth
A characteristic size for fiber optic ables.
Bandwith Length Product
The bandwidth length product is a factor that decides the maximum length of multimode fibers.
BFOC
B ayonet F iber O ptic C onnector A widely used plug connector for fiber optic cables with bayonet locking. It is also called ST plug. The only plug connector that is standardized in ETHERNET with a transmission speed of 10 Mbit/s. (ST is a registered trademark of AT&T.)
BNC
B ayonet N eill C oncelman A widely used plug connector for connecting coaxial cables and transceivers as per 10Base2; named after the developers.
BootP
B ootstrap P rotocol A protocol that delivers a statically allocated IP address to a device connected to the ETHERNET on the basis of its MAC address.
Bridge
A device that works on Layer 2 of the OSI reference model and connects 2 similar networks to each other. In this connection, data packets are transferred from one subnetwork to another subnetwork through the analysis of the MAC address.
Broadcast
Term for transmitting a message to a groupof unspecified receivers.
Browser
Term for software that enables the viewing and processing of data in the Internet. The most well-known browsers are Microsoft Internet Explorer and Netscape.
Burst
Term for a short-term increase in load that occurs suddenly.
C
CSA<
www.csa.ca C anadian S tandard A ssociation
CSA
Canadian standard for the security of C22.2 No. 950 IT devices, including electrical office machines based on the IEC 950.
CSA
Canadian standard for the safety of C22.2 No. 142 industrial control equipment, based on UL 508.
CSA
Canadian standard for electrical operating C22.2 No. 213 equipment for explosion-endangered rooms of the Class I and II, Department 2, and Class III; based on UL 1604.
Cheapernet
See 10Base2.
Collision
The CSMA/CD access process limits the Domain runtime of a data package from one participant to another. Depending on the data rate, what results is a spatially limited network, the so-termed collision domain. The maximum diameter of collision domain is 5120 m at 10 Mbit/s (ETHERNET) and 512 m at 100 Mbit/s (Fast ETHERNET). The full duplex operation of a connection enables expansion over this limit value since it precludes collisions. The precondition for this is the use of bridges or switches.
Concentrator
See Hub.
CoS
C lass o f S ervice
Connection
A function that enables the copying of data
Crossover Cable
For connecting ETHERNET components over copper cable, what are required are either 1:1 wire cables, or crossover cables. Crossover cables are required for direct cabling of terminal devices such as SPS, HMI, etc. or network components such as hubs, switches, etc. to each other. If the devices support autocrossing, one can also use 1:1 wire cables. Pin allocation of RJ45 plugs in a crossover cable:
CRC
C yclic R edundancy C heck Term for algorithms that are used for error detection and correction of bit-oriented protocols.
CSMA/CD
C arrier S ense M ultiple A ccess/ C ollision D etection Access process in ETHERNET as per IEEE 802.3. A station ready to transmit checks whether the transmission medium is free (carrier sense). It then starts transmitting while simultaneously checking whether other stations (multiple access) have also started to transmit data. If 2 or more stations transmit simultaneously, there is a collision. The stations stop transmission accordingly (collision detection) and> attempt transmission later on. In the CSMA/CD process, the network expansion is determined by a maximum permissible runtime of data signals on the network that depend on the data rate.
cUL 508
US standard for the security of industrial control equipment.
cUL 1604
US standard for electrical operating equipment for explosion-endangered rooms of the Class I and II, Department 2, and Class III; based on UL 1604.
cUL 60950
US standard for the safety of IT equipments including electrical office machines; based on IEC 950.
Cut Through
Method of working of switches in which a data packet is immediately relayed further after detecting the target address. The delay time (latency time) is thereby small, but wrong packets are also transmitted onward. In this process, it is not possible to adjust the speed between the individual segments. The phenomenon is also called "On-the- Fly-Switching".
D
DA
D estination A ddress Target address within a data telegram.
DCE
D ata C ircuit-terminating E quipment Term for devices that are used for network termination and to which terminal equipments such as computers, control systems and printers are connected.
DHCP
D ynamic H ost C onfiguration P rotocol A protocol that temporarily allocates an IP address to ETHERNET participants from an established range of IP addresses.
Dispersion
Runtime differences in a LWL (fiber-optic cable). Through dispersion, a pulse transmitted in a fiber optic cable is extended. A distinction is made between mode, material and wave dispersion. Mode dispersion arises due to runtime difference between the individual modes. For this reason, this type of dispersion occurs only in multimode fiber optic cables. The material dispersion arises due to the wave length dependency of the refractive index. The fiber optic cable dispersion arises due to differing extension speeds in the energy transmitted in the core and in the jacket. This type of dispersion is of practical importance only for single mode fiber optic cables.The chromatic dispersion is a characteristic quantity for single mode fiber optic cables. It is the total of material and wave dispersion.
DSC
D uplex S traight C onnector A widely used plug connector for fiber optic cables. Also see SC.
DNS
D omain N ame S ystem Term for a system which maps host names in plain text to IP addresses. The data source for the conversion are for example DNS servers or files with the designation "Hosts ".
DTE
D ata T erminal E quipment Term for terminal equipment such as computers, control systems and printers that are connected to a network. In German, they are also referred to as Datenendeinrichtung (DEE).
Dual Homing
A term that was coined in connection with FDDI networks. Dual homing is a technology in which a device is connected to a network through 2 independent connecting points. One connecting point is for the primary connection, the other is a standby connection. If the primary connection fails, the standby connection is automatically activated. With this technology, it is also possible to connect network segments redundantly.
E
EGP
E xterior G ateway P rotocol Classification of routing protocols for exchanging information between routers of independent networks.
EIA
www.eia.org E lectronic I ndustries A ssociation American industrial association of electrical industry active in the field of standardization. Standards of the EIA are designated with RS (related EIA standard). The wellknown standards include the serial interfaces RS 232 C, RS 422 and RS 485.
EMV
Electromagnetic compatibility. In EMV, both the aspects of interference immunity as well as interference emission must be kept in mind. Electrical devices, installations and systems must have a specific immunity against normal interference effects that normally occur in the planned environment. In addition, devices should not emit any interference variables that may possibly disturb other devices in their environment.
EN
E uropean N orm European standards relate to standards developed by CENELEC and CEN.
EN 61000-4-2
EMV Part 4: Measurement and Testing Processes, Main chapter 2: Testing interference immunity to the discharge of static electricity. Details in the catalog: x kV Contact discharge x kV Air discharge
EN 61000-4-3
EMV Part 4: Measurement and Testing Processes, Main chapter 3: Testing interference immunity to high-frequency electromagnetic fields.
EN 61000-4-4
EMV Part 4: Measurement and Testing Processes, Main chapter 4: Testing the Interference immunity to fast, short disturbance variables (Burst). Details in the catalog: x kV DC Power lines x kV Data lines
EN 61000-4-5
EMV Part 4: Measurement and Testing Processes, Main chapter 5: Testing interference immunity to surges. Details in the catalog: x kV Power supply asymmetrical (power supply) x kV Data lines
EN 61000-6-2
Generic standard Part 6-2: Interference immunity in industry.
EN 50081-1
Generic Standard Interference Emission, Part 1: Residential, business and trade sectors as well as small businesses.
EN 50081-2
Generic Standard Interference Emission, Part 2: Industry.
EN 50082-1
Generic Standard Interference Immunity, Part 1: Residential, business and trade sectors as well as small businesses.
EN 50082-2
Generic Standard Interference Immunity, Part 2: Industry no longer valid since the 01.04.2002.
EN 55022
Product Group Standard Interference Emission for IT installations.
EN 55024
Product Group Standard Interference Immunity for IT installations.
EN 60950
Safety of IT installations including electrical office machines. European standard, based on the IEC 950.
EN 60825-1
Safety Of Laser Devices,Part 1: Classification of Installations, Requirements and User Guidelines.
EN 61131-2
Product Group Standard Stored-Program Control Systems, Part 2: Requirements and Tests for Operating Materials.
Encapsulation
See Tunneling.
ETHERNET
Term for a data network that has been standardized since 1985 by the IEEE 802.3. The standard specifies the functions and the construction of the Levels 1 and 2 in accordance with the OSI reference model. ETHERNET is based on the access process CSMA/CD with a variable packet length of between 64 and 1518 bytes and transmission speeds of 10 Mbit/s ( 4 bytes TAG field optional). The concept of ETHERNET is often used as a general designation without making any distinction between the different variations (ETHERNET, Fast ETHERNET, etc.). In addition, the protocols of the Levels 3 and 4 are often included.
EtherNet/IP
www.ab.com/networks ETHERNET /Industrial P rotocol. A standard for Industrial ETHERNET applications, based on TCP and UDP.
ETHERNET Packet
Term for an ETHERNET data packet. The packet size varies between 64 and 1522 bytes. It contains the destination and source address field (DA or SA) apart from the actual payload data, the TAG field as well as the length/type field.
ESD
E lectro S tatic D ischarge Term for electrostatic discharges. Electrostatic discharges can cause short and irregular disturbances in electronic devices or they may destroy electronic components.
Ex
Independent designation of devices under DIN EN 50020 that can be used in accordance with the specifications even inside explosion-endangered areas.
F
Fast ETHERNET
Term for a fast data network that was standardized in 1995 by the IEEE 802.3. Based on a transmission speed of 100 Mbit/s with a variable packet length ranging from 64 to 1518 bytes (4 bytes TAG field optional).
FCC
www.fcc.gov F ederal C ommunications C ommission A US authority established in 1934, responsible for telecommunications. It administers the frequency spectrum and allocates it over local, regional and national levels.
FCC CFR47 Part 15
F ederal C ommunications C ommission C ode of F ederal R egulations Standard for interference emission for IT installations. t - min. 64, max. 1518 Octets - Preamble Field - Data Field - Start Frame Delimiter Field - Destination Address Field Source Address Field - Length/Type Field - Data Field - Pad Field - Frame Check - Sequence Field
FCS
F rame C heck S equence Term for a bit field for data security of payload data in bit-oriented protocols. The sender of a message determines a checksum according to an established algorithm, and this checksum is affixed to the end of the packet. In the receiver a checksum is also created according to the same algorithm, and this checksum is compared to the checksum received. With this process, errors in the data transmission can be detected.
FDX
See Full Duplex.
FDDI
F iber D istributed D ata I nterface A standard for data networks, that covers the Layers 1 and 2 of the OSI reference model. FDDI is originally based on a double ring topology with fiber optic cables as the transmission medium.
Fiber optic
In contrast to electrical transmission cable technology in which twisted pair cables are used for data transmission, glass or plastic is used as a transmission medium for optical transmission technology. Fiber optic cables come in the form of multimode and single- mode fibers (monomode fibers).
Firewall
Term for protective measures that partitions off a LAN from another network, for example the Internet.
Flow Control
A function that in case of overload at an output port, dumps packets at the input port or signals connected devices to stop transmission. The signal to stop transmission is sent in half duplex operation by simulating a collision or, in full duplex, by sending special "Pause" packets.
Flow Controls
See Flow Control.
FM 3611
US standard for electrical operating equipment for explosion-endangered rooms of the Classes I and II, Department 2, and Class III.
FM 3810
US standard for the Safety of Process Control Equipment
FTP
F ile T ransfer P rotocol A protocol on Layer 5 of the OSI reference model for the transportation of files.
Full Duplex
A mode of operation in which a device can simultaneously transmit and receive data. If a transmission path is operated in full duplex in ETHERNET, the CSMA/CD bus access process does not apply and network diameter then depends solely on the performance limits of the transmission and reception components used.
F/O
F iber O ptics See Fiber Optic Cable.
G
GARP
G eneric A ttribute R egistration P rotocol Term for a protocol family that is used for exchanging parameters between switches and Layer 2 of the OSI reference model. At present there are the protocols GMRP and GVRP.
Gateway
A device that operates above the Layer 2 of the OSI reference model and converts protocols. At Layer 3, these devices are generally designated as routers.
Gbps
G iga b it p er s econd.
Gigabit ETHERNET
Term for an extremely fast data network that has been standardized by the IEEE 802.3 in 1999. Based on a transmission speed of 1000 Mbit/s with a variable packet length of 64 to 1518 bytes (4 bytes TAG field optional). LAN 1 LAN 2 RS2-FX/FX RS2-FX/FX F/O cable 2 x 10 Mbit/s = 20 Mbit/s
GL
G ermanischer L loyd A company for the classification of seagoing ships, established 1867 in Hamburg.
GMRP
G ARP M ulticast R egistration P rotocol A protocol standardized as per IEEE 802.1p that enables participants to log-on and logoff to/from multicast groups dynamically. Switches that support GMRP only switch multicasts to those ports at which participants of the respective multicast groups are registered.
GVRP
G ARP V LAN R egistration P rotocol A protocol that can use switches to exchange information on VLAN’s. If a VLAN is installed on a switch, the switch sends this information to all the other switches in the network. In addition, the port at which the information was received can also be made a participant of this VLAN.
H
Half Duplex
A mode of operation in which a device can either send or receive data at any given point in time. In half duplex, collision detection is active in ETHERNET. Network expansion is limited by the runtime delay of the devices and transmission media.
HCS
A name for a fiber optic cable, the optical core of which is made of silica glass and whose optical jacket is made of a special patented plastic layer (HCS is a registered trademark of Spectran Specialty Optics).
HDX
See Half Duplex.
Header
Term for that part of a data packet that is located before the payload data and contains data such as addresses, packet numbers, etc.
HIPER-Ring
Term for a redundancy process based on the construction of ring-shaped network structures. In rings of these types, network components that support the HIPER-Ring are connected to each other over their backbone or ring ports. A redundancy manager carries out monitoring of the ring and prevents circulating telegrams.
HMI
H uman M achine I nterface Devices for operating and observing machines and equipment.
Hops
Term for the routers that a data packet may pass through on its way through a network. The number of hops within a connection does not indicate anything about the quality of the connection. Thus for example a connection with eight hops may be faster than a connection with five or six hops.
HSRP
H ot S tandby R outing P rotocol A protocol for controlling redundant routers.
HTML
H yper T ext M arkup L anguage A format for displaying websites.
HTTP
H yper T ext T ransfer P rotocol A protocol used by browsers and webservers for transmitting websites.
Hub
A device that works on Layer 1 of the OSI reference model and that regenerates incoming signals before distributing the same to all the other ports. Synonym: star coupler or repeater.
I
IAONA-EU
www.iaona-eu.com I ndustrial A utomation O pen N etworking A lliance EU rope An independent umbrella organization set up in 1999 in Europe, the objective of which is to standardize and harmonize Industrial ETHERNET technologies.
ICMP
I nternet C ontrol M essage P rotocol A protocol that is used to signal failures and errors during transmission of IP packets. An extremely well-known command of this protocol is the "ping" command.
ID
ID entifier
IDA
www.ida-group.org I nterface for D istributed A utomation A standard in the field of Industrial ETHERNET developed by a group of companies using TCP and UDP.
IE
I ndustrial E THERNET Term for ETHERNET in automation technology. The enhanced requirements concern the accessibility and the security of the network and the environmental conditions to which ETHERNET components are exposed.
IEC
I nternational E lectrotechnical C ommission A commission set up in 1906 for the standardization of electrical components and modules.
IEC 60068-2-6
Environmental tests Part 2: Fc test, sine-shaped vibrations.
IEC 60068-2-27
Environmental tests Part 2: Ea test, shock.
IEC 60068-2-32
Environmental tests Part 2: Ed test, free fall.
IEEE
www.ieee.org I nstitute of E lectrical and E lectronics E ngineers An association of technicians and engineers having their headquarters in the USA that develops de facto standards, particularly in the field of data communication.
IEEE 802.3
A committee of the Institute of Electrical and Electronics Engineers, that lays down standards for LANs.
IETF
www.ietf.org I nternet E ngineering T ask F orce A group that consists of several technical persons interested in the Internet, responsible for technical questions.
IFG
I nter F rame G ap A measure for the minimum distance between 2 data packets.
IGMP
I nternet G roup M anagement P rotocol Term for a Layer 3 protocol that communicates the association of participants and routers to multicast groups to the adjacent routers.
IGMP Snooping
I nternet G roup M anagement P rotocol Snooping A function in which switches investigate IGMP packets and allocate membership of a participant to a multicast group to the respective port. Thereby muliticasts can also be switched specifically to those segments in which the participants of a group are located.
IGP
I nterior G ateway P rotocol Classification of routing protocols for exchanging information between routers within an independent network. The protocols used include IGRP, RIP and OSPF.
IGRP
I nterior G ateway R outing P rotocol Routing protocol developed by Cisco.
IP
I nternet P rotocol A protocol on Layer 3 of the OSI reference model. It is used for the connectionless transportation of data over several networks. Each telegram is allocated a clear IP address. The telegrams may arrive at the receiving end in a sequence different to the one in which they were sent. TCP is responsible for assembly in the correct sequence.
IP Address
The address of a participant on Layer 3 of the OSI reference model. In Version 4, an IP address consists of 4 bytes separated from each other by decimal points. These 4 bytes indicate the address for the network (Net ID) and the address area of the terminal devices (Host ID). The entire address range is classified into classes from A to E in accordance with the number of network addresses and host addresses, the number of host addresses becoming increasingly smaller from A to E. Since IP addresses must be unique on the Internet, the network addresses are managed by a central organization. The allocation of host addresses is done by the administrator of the respective local network. In order to split-up local networks into smaller subnetworks that are easier to manage, part of the host addresses is used. The network address is thereby increased with a subnetwork component. This extension is done using a subnetwork mask. The subnetwork mask marks all the bits of an IP address that identify the network and subnetwork. A device that wants to transmit, compares its IP address with the IP address of the receiver. If the addresses do not match within the framework of the network mask, it means that the receiver is in a different network. In such case the message is sent to a gateway or a router.
IPv4
I nternet P rotocol v ersion 4 The IPv4 has an address length of 4 bytes. Also see IP.
IPv6
I nternet P rotocol v ersion 6 The IPv6 has an address length of 16 bytes. In addition, it is also differentiated by the structure of the header and the division of the networks into address types rather than classes.
IPX
I nternetwork P acket E xchange Term for a protocol by Novell that creates connections to Internet protocols.
ISO
www.iso.org I nternational S tandards O rganization An umbrella organization of national standardization committees that is also a member of the Deutsches Institut für Normung (DIN) (German Standards Institute). More than 200 technical committees (TC) make up the various bodies of the ISO. The TCs may be subdivided if so required into subcommittees (SC). The SCs in turn may be split up into working groups (WG) and special task groups.
IT
I nformation T echnology
ITU-T
www.itu.int I nternational T elecommunications U nion- T elecommunication Standardization committee with its head office in Geneva.
J
Jabber
Term for an ETHERNET packet with more than 1522 bytes.
Jitter
Term for the oscillation of signal edge in time.
K
Kbps
K ilo b it p er s econd (kbit/s)
L
LAN
L ocal A rea N etwork Term for local network which is typically no bigger than 10 km in diameter.
Latency Time
Term for the time difference between the receipt and the relaying of data. As a rule, the latency time is measured between the last bit received and first bit sent out.
LED
L ight E mitting D iode An electronic component that emits light.
Link Aggregation
Term for a function that combines up to 4 ports with the same transmission speed to one virtual port. The result is redundancy in the case of failure of a connection. Also called trunking.>
Long Haul
Term for optical transceivers with a very high link budget that is used in connection with single-mode fibers.
LSB
L east S ignificant B it Low-value bit within a bit sequence.
LWL
See Fiber Optic Cable.
M
MAC
M edia A ccess C ontrol Term for a sublayer of Layer 2 of the OSI reference model. It controls access to the transmission medium. In this sublayer, processes may be used in which either several equally authorized participants are competing for access (for example CSMA/CA or CSMA/CD) or in which no collisions occurs, for example such as token ring.
MAC-Address
The address of a participant on Layer 2 of the OSI reference model.
MAN
M etropolitan A rea N etwork Term for a network within a city that connects various LANs to each other.
MAP
M anufacturing A utomotion P rotocol A protocol developed in the early 1980s on the initiative of General Motors. However in view of its complexity, it was hardly used commercially.
MAU
M edium A ttachment U nit A coupling module between an ETHERNET terminal device and the transmission medium. As a rule the terminal device is connected to an AUI interface. Also see Transceiver.
Mbps
M ega b it p er s econd (Mbit/s)
MDI
M edium D ependent I nterface Term for the physical (electrical, optical) and mechanical interface of a device for connection to the transmission medium.
MDI-X
MDI-Crossover Term for a MDI interface with crossed connected signal lines.
MDI/MDI-X
See Autocrossing.
MIB
M anagement I nformation B ase A database for objects and functions which help network management systems manage individual objects using Simple Network Management Protocol (SNMP).
MII
M edia I ndependent I nterface Term for an interface as per the OSI reference model between the Physical Layer (1) and the Data Link Layer (2).
Media Converter
A device that operates on Layer 1 of the OSI reference model and converts signals between various media. For example optical signals into electrical signals.
Mirroring
Transmission between 2 ports of a switch to other ports, in order to have the data analyzed by an analyzer.
Monomode Fiber
See Single-mode Fiber and Fiber Optic Cable.
MSB
M ost S ignificant B it The most significant bit within a bit sequence.
MTBF
M ean T ime B etween F ailures Probability factor that indicates after how much time an error may be expected.
MTRJ
A widely used small sized plug connector for fiber optic cables.
Multicast
Term for transmission of a message to a group of specific receivers. It is possible to contact this group using only one address.
Multicast Filtering
Term for processes that enable a switch to relay multicasts in a targeted manner.
Multimode Fiber
Multimode fibers are fiber optic cables that are distinguished through core diameters of comparable size. The typical core diameter for step-index fiber optic cables is 100 m for glass fibers, 200 m for PCS/HCS fibers and 980 m for POF fibers. The graded index fibers on the other hand have a typical core diameter 50 or 62.5 µm. Because of this relatively large core diameter, the light in multimode fibers spreads over several paths and modes. The distance that can be covered by a multimode fiber depends on several factors: the characteristics of the fiber, the link budgets and the attenuation due to plug connectors, splices and other components. For example: A 62.5/125 µm fiber with an attenuation of 1 dB/km and a bandwidth of 500 MHz * km should transmit data packets over Fast ETHERNET using light with a wavelength of 1300 nm. The link budget is 11 dB. A reserve of 3 dB should be taken into account. The attenuation of the plug connectors should be ignored. Attenuation: Lmax = (Link Budget Reserve)/ fiber attenuation Lmax = (11 dB-3 dB)/1 dB/km Lmax = 8 km Bandwidth length product Lmax = Bandwidth/Signal bandwidth Lmax = (500 km* MHz)/125 MHz Lmax = 4 km In this example, the maximum distance to be covered is 4 km. Signal bandwidths: ETHERNET = 10 MHz, Fast ETHERNET = 125 MHz and Gigabit ETHERNET = 1.25 GHz
Multiplexer
Term for devices or function units that combine several channels of low capacity into one channel of high capacity.
Multiport Bridge
A bridge that connects not only 2 but several LANs together. In ETHERNET LANs, multiport bridges are also designated as switches.
N
NAT
N etwork A ddress T ranslation Term for a protocol that is defined in RFC 1631 and RFC 1918.
NEXT
N ear- E nd Xr oss T alk A form of crosstalk in which signals of participants that are located on the same side of a twisted pair cable get superimposed.
NIC
N etwork I nterface C ard Term for PC insertion cards that enable connection to an ETHERNET network.
Network Nodes
Term for network elements such as hubs, switches and routers on which different data transmission paths run together. Input puls Output puls
Network Management
A general concept for the management, configuration and monitoring of network nodes and the devices connected the same. The tasks of a network management system may be subdivided into error management, configuration management, safety management and performance management. To do this, the network management agent communicates with the network management station using the network management protocol SNMP.
Network Mask
The network mask marks all bits in an IP address for identifying the network and the subnetwork. Also see IP address.
NMS
N etwork M anagement S tation See Network Management.
Node
Term for a participant in a network.
NRZ
N on- R eturn to Z ero Term for a coding process in which the electrical signals do not go back to zero even when there is a sequence of several logical ones.
NRZI
N on- R eturn-to- Z ero, I nvert on ones Term for a coding process with inverted NRZ signals.
O
OID
O bject ID entification
OLE
O bject L ink and E mbedding Term for a central architecture principle in Windows.
On-the-Fly Switching
Working method of switches see Cut Through. Binary notation IP address 10010101.11011010.00010011.01011010 Network mask 11111111.11111111.11111111.00000000 ->Subnetwork 10010101.11011010.00010011.00000000 Decimal notation IP address 149.218.19.90 Network mask 255.255.255.0 - >Subnetwork 149.218.19.0 Available address range Host addresses 149.218.19.1 bis 149.218.19.254 Broadcast address 149.218.19.255
OPC
www.opcfoundation.org O LE for P rocess C ontrol Standard interface for Windows applications for data exchange concerning process data and status information.
OSI
O pen S ystems I nterconnection An international standardization program that has been instituted by the ISO and theITU. The objective is to lay down standards for data networks that ensure the compatibility of devices made by various manufacturers.
OSI Reference Model
Also termed ISO/OSI reference model. This model is divided into 7 Layers that describe the communication of open, distributed systems. The individual layers form a group, that are independent of each other, but each describes an area that is relevant for data transmission and processing.The layers are termed the Physical Layer (1) the Data Link Layer (2) Network Layer (3) the Transport Layer (4) the Session Layer (5) the Presentation Layer (6) and the Application Layer (7).
OSPF
O pen S hortest P ath F irst Term for a routing protocol. OSPF uses information given by the routers over the topology of the network in order to find the shortest path between the routers. The precondition for this is that each router creates a routing table in which the current topology of the network is fully displayed. Since each router immediately communicates changes in the topology to the adjacent routers, the routing tables in the routers get constantly updated. The advantage of OSPF over RIP consists in the speed and the better distribution of load.
OUI
O rganizationally U nique I dentifier Term for the first 3 bytes of the MAC address.
P
Packet Size
See ETHERNET Packet.
Patch Cable
Term for cables that are used for connecting ETHERNET component within a room (19" rack, control cabinet, etc.). Patch cables are mostly used in connection with patch panels.
Patch Field
Term for a patching distribution frame.
PCF
Term for a fiber optic cable, the optical core of which is made of silicon glass with an optical jacket consisting of a polymer layer.
PDU
P rotocol D ata U nit Term for a data packet assembled on a layer of the OSI reference model that is relayed to the layer below it over a Service Access Point (SAP).
Ping
P acket In ternet G roper A program for testing connections between 2 IP addresses.
PLC
P rogrammable L ogic C ontrol Stored-program control systems.
POF
P lastic O ptical F iber Term for a fiber optic cable, the optical core and jacket of which is made of plastic. POF fibers have a typical core diameter of 0.98 mm.
Port
General term for an interface to devices for transmission of data and control information in the transmission and reception direction. Port Mirroring A function that enables the copying of incoming and outgoing data at one port of a switch to another port, in order to be analyzed there with an analyzer for example.
Port Security
A function that offers protection against unauthorized access to the network. Switches that support this function offer the possibility of setting, for each port, the terminal device from which data can be transmitted or received. The checks are carried out on the basis of the MAC addresses of the devices connected. If the device is connected to a port, the MAC address of which is not registered, this port can be automatically switched off.
Port Trunking
See Link Aggregation.
PPP
P oint-to- P oint- P rotocol A protocol of the TCP/IP family for serial data transfer over dial-up connections such as the telephone. This is used for connecting computers that are not permanently connected over LAN’s to the Internet.
pps
P ackets p er S econd Measurement unit for the switching speed.
Prioritization
In a prioritized data transmission, data packets are switched on the basis of the defined criteria. The tagging of such packets is done at Layer 2 of the OSI reference models in the TAG field and at Layer 3 in the TOS field.
PROFInet
www.profibus.org A concept that defines the communication from the field level to the conducting level with the integration of profibus and ETHERNET as well as a model for companywide engineering.
Q
QoS
Q uality o f S ervice Term for a range of factors that have an effect on the quality of a network. These factors include network breakdown times, delay times, stability of connections and many more. For QoS, there is a series of different definitions.
R
RAM
R andom A ccess M emory Term for a volatile memory.
RARP
R everse A ddress R esolution P rotocol A protocol that delivers statically allocated IP addresses to a MAC address.
Redundancy Manager
Term for a switch or hub in a HIPER-Ring that monitors the ring and in case of an interruption in the ring structure, activates the connection that has been switched off upto that point. After the interruption has been removed, the redundancy manager again switches this connection off. The ring is thereby physically switched off, but from the point of view of communication, it is interrupted.
RFC xxx
R equest F or C omments An abbreviation that was coined within the context of the Internet. It is closely linked to the publication of Internet standard.
RIP
R outing I nformation P rotocol A protocol for the cyclic exchange of routing tables between routers within independent networks per broadcast. RIP is one of the oldest, easiest and most widely used routing protocols. The successor of RIP is the more complex OSPF.
RJ45
A widely used plug connector in telephone technology and in LANs. It is also known as the Western plug with 8 poles.
RMON
R emote network MON itoring A protocol for network management. RMON defines new classes of data that relate to and can be recorded on the lower layers of the OSI reference model. The data are then transmitted to a network management station using Simple Network Management Protocol (SNMP).
RMON 2
R emote network MON itoring A protocol for network management. RMON 2 is an extension of RMON and extends to higher layers of the OSI reference model.
Router
A device that works at Layer 3 of the OSI reference model and connects different segments of the network to each other, or splits-up networks into subnetworks. A router transmits only data packets to other segments that are sent to its own MAC addresses. The router then sends the data packets onward on the basis of routing tables. In other words, the transmitting participant must know that the receiver is not located in the same network segment. The transmitting station obtains this information from the IP address of the recipient. Routing tables are either given as fixed tables or are given by the router itself using routing protocols.
Routing
A function of Layer 3 of the OSI reference model. A distinction is made between dynamic and static routing. In dynamic routing, routers calculate rules and parameters for path selection through the network. This information is written to routing tables and exchanged using routing protocols between routers. This ensures that the path selection is adapted to the current topology and load distribution of network. In dynamic routing, each telegram is individually routed. As a result, telegrams may arrive at the receiving end in a sequence different to the one in which they were sent. In static routing, the paths for data transmission between the transmitters and receivers is fixed and a specific bandwidth is reserved for each connection. As a result, data packets take the same path between two terminal devices. It is therefore not possible to respond automatically to changes in the topology or in the case of overloads of connections. Since all changes in the network structure are entered into the routers by hand, routers do not have to support any routing protocols in this process. While dynamic routing supports the transmission of data in an optimized manner, in static routing, the transmission of data, speech and video are equally supported.
Routing Protocol
Term for protocols that routers use during dynamic routing in order to exchange information over connected networks amongst each other. This information is stored in routing tables in the routers.
RS 232 C
R ecommended S tandard 232 C A widely used serial interface for data transmission with data rates of up to 20 kbit/s and over distances up to 15 m. This interface was standardized by the EIA in 1969 as standard no. 232 in Version C. It is also often referred to as RS 232.
RS 422
R ecommended S tandard 422 A serial interface for data transmission in full duplex operation. This interface was standardized in the 70’s by the EIA as standard no. 422.
RS 485
R ecommended S tandard 485 A serial interface for data transmission that enables a bus structure with several participants. This interface was standardized by the EIA in the 70s by the EIA as standard no. 485.
RSTP
Rapid Spanning Tree Protocol
RSVP
R esource Re S er V ation Setup P rotocol A protocol that reserves resources for applications over the Internet. After a path has been established from the sender to the receiver, all the routers participating in this path are notified via RSVP that they should reserve specific resources for this connection.
RTP
R eal- T ime P rotocol A protocol that supports real-time applications such as video conferencing on the Internet. In this protocol, additional information such as the nature of the payload data transmitted (speech, video, etc.) or the time of generation of the payload data is transmitted.
Rx
Abbreviation for Receiver. Term for the connection to a port at which data is received.
S
SA
S ource A ddress Source address within a data telegram.
SAP
S ervice A ccess P oint Term for the interface between two layers of the OSI reference model where a layer that is placed at a higher level makes use of services in the layer below.
SC
S traight C onnector A widely used plug connector for fiber optic cables. Also see DSC.
SCADA
S upervisory C ontrol A nd D ata A cquisition Term for systems for control and visualization of processes. SCADA systems are based on Windows operating systems as a rule.
SDH
S ynchronous D igital H ierarchy A European standard that defines several standards of transmission rates and transmission forms for optical fibers (fiber optic cable).
SFD
S tart of F rame D elimiter Part of an ETHERNET telegram.
Shared Network
Term for an ETHERNET network in which participants share the available bandwidth. In these networks, the CSMA/CD process controls the access of the participants to the transmission medium.
Single-mode
A single-mode fiber is a fiber optic cable fiber that is characterized by its extremely small core diameter (max. 10 m). As a result, in this fiber, the light after the cutoff waveline can only get extended along one path one mode. The distance that is to be covered by a single-mode fiber depends on several factors: the characteristic data of the fiber, the link budget as well as the attenuation to plug connectors, splices and other components. Example: A 9/125 m fiber with an attenuation (A) of 0.3 dB/km should transmit a wavelength of 1550 nm of Fast ETHERNET data packets. The link budget is 29 dB. A reserve of 3 dB is taken into account. The attenuation of the plug connector is to be ignored. Attenuation: Lmax = (Link Budget-Reserve)/Fiber attenuation Lmax = (29 dB-3 dB) /0.3 dB/km Lmax = 86.6 km In this example, the maximum distance to be covered is 86.6 km. Signal bandwidths: ETHERNET = 10 MHz, Fast ETHERNET = 125 MHz and Gigabit ETHERNET = 1.25 GHz
SLIP
S erial L ine I nternet P rotocol A protocol for serial data transfer over dialup connections such as the telephone. It is used for connecting computers that are not networked permanently over LANs to the Internet. In comparison to the more recent PPP, SLIP has the disadvantage that erroneous data is not recognized.
SMTP
S imple M ail T ransfer P rotocol Term for a protocol for sending e-mail messages.
SNMP
S imple N etwork M anagement P rotocol A protocol for network management. SNMP defines commands for the reading and writing of information, status and error messages as well as providing a structured model. This model consists of agents with their respective Management Information Base (MIB) and a Manager. The Manager is a program that runs on a network management station. Agents are mostly located within devices such as switches, routers and terminal devices that support the SNMP. The task of the agents consists in collecting and preparing data in the MIB. These data is requested at regular intervals by the Manager and displayed on the network management station. The devices are configured, for example, with the data that the Manager writes to the MIBs in question. In urgent cases, the agent can also send messages (traps) directly to the Manager.
SOHO
S mall O ffice H ome O ffice Network solutions and access technologies to the Internet for small offices and offices at home that are not directly connected to large company networks.
Spanning Tree
Term for a protocol that is used in ETHERNET networks for path determination. It is specified as standard IEEE 802.1 D. The spanning tree algorithm prevents the circulation of data packets in a LAN with several possible paths by switching-off individual connections or ports. In addition it determines the optimum path if there are several alternatives. If a path fails due to the fault or interruption, an alternative connection is searched for using the spanning tree protocol. The reconfiguration of a network of this type may takes 30-90 seconds.
SPS
Stored memory controlled system.
Star Coupler
See Hub.
ST
A widely used plug connector for fiber optic cable with bayonet locking. It is also known as BFOC plug. It is standardized as the only plug connector for ETHERNET (10 Mbit/s). (ST is a registered trademark of AT&T).
Store-and-Forward
A method of working for switches in which a data packets is first read-in completely and checked for errors before the switch relays the same. This process enables the connection of segments with differing transmission rates.
STP
S hielded T wisted P air See Twisted Pair Cable.
Subnetwork Mask
Network mask or subnet mask. The network mask marks all the bits of an IP address for the identification of the network and the subnetwork. Also see IP address.
Switch
A device that works on Layer 2 of the OSI reference model. In contrast to hubs, switches analyze the incoming data packets and only relay them to ports at which the receiver is registered. Exceptions from such targeted switching are multicasts and broadcasts that are sent to all ports. The transmission of data packages can be done at several ports simultaneously and in full duplex operation. Thus switches optimize the available bandwidth of the LAN. Recently, Layer 3- and Layer 4-switches have been brought out, that have additionally implemented the partial function of these layers.
Switched Network
Term for an ETHERNET network that is made up of switches.
Switching Hub
See Switch.
T
TAG Field
An optional field in the ETHERNET telegram that contains information about the priority and associated VLAN of the payload data.
TCP
T ransmission C ontrol P rotocol A connection-oriented protocol at Layer 4 of the OSI reference model. It enables a full duplex point-to-point connection and extends the Internet protocol below it by functions for data security and connection control.
Telnet
An emulation program based on TCP/IP that executes processes or uses programs on a different device. The system resources of the other device are used. This distinguishes Telnet from FTP for example, which only searches for file systems.
TFTP
T rivial F ile T ransfer P rotocol A protocol based on Layer 5 of the OSI reference model and uses UDP for fast and uncomplicated transmission of files. TFTP is considerably quicker than FTP.
Thick Wire
See 10Base5.
Thin Wire
See 10Base2.
Transceiver
1. General term for a transmission/reception component. 2. Term for media converter within the Rail family. In addition there are plug-on transceivers for fiber optic cables, twisted pair and coax cables. These transceivers are provided with power supply over the 15-pole AUI interface by the terminal device connected.
Topology
A description of the type of line routing. The essential basic forms are line topology, tree topology, ring topology and star topology.
TOS
T ype O f S ervice A field in the Internet protocol for prioritizing data.
TP
T wisted P air See Twisted Pair Cables.
Trap
Term for the signaling of error signals to a network management station.
Trunking
See Link Aggregation.
TTL
T ime T o L ive A field in the header of the Internet protocol that indicates for how long the packet is valid.
Tunnelling
Term for the packaging of data in data packets of another protocol that operates on the same Layer of the OSI reference model. This process is also termed encapsulation.
Twisted Pair Cable
Term for 2 wires that are isolated from each other but are twisted together. A distinction is made in this connection between Unscreened (UTP) and Screened Twisted Pair cables (STP).
Tx
Abbreviation for transmitter. Term for the connection to a port to which data is sent.
Transmission Rate
Term for the speed at which data is transmitted. For ETHERNET: 10, 100, 1000 and 10000 Mbit/s.
U
UDP
U ser D atagram P rotocol A connectionless protocol at Layer 4 of the OSI reference model. In contrast to the Transport Control Protocol (TCP), UDP does not have any functions for data security and connection control. As a result it is considerably faster and more suitable for real-time applications such as speech and video transmissions as well as for the transmission of short messages that can be repeated in case of error.
UL
www.ul.com U nderwriters L aboratories Independent institution in the USA that lays down and executes safety tests for products.
Unicast
Term for sending a message to a specific receiver.
UPS
U ninterruptible P ower S upply
URL
U niform R esource L ocator A standardized scheme for access to hypertext documents and other services through a browser.
USB
U niversal S erial B us Term for a serial bus for connection of modems, mice, keyboards, printers and other peripheral devices. A maximum of 127 devices can be connected to the bus. The cable length between two devices must not exceed 5 m.
UTP
U nshielded T wisted P air See Twisted Pair Cable.
V
VLAN
V irtual LAN Term for LAN’s that are logically configured independently of their real physical topology. A distinction is made between static and dynamic VLANs. In static VLANs, the ports of a switch are permanently allocated to one or more VLANs. A subnetwork is therefore made up of a list of port numbers. In the case of dynamic VLANs, the subnetworks are made up of MAC or IP addresses that are maintained in a database. The ports of the switches are automatically configured on the basis of this database. VLAN’s are intended for making groups of participants who can only communicate with each other in accordance with predefined rules. A further application of VLAN’s is the delimitation of broadcasts.
VPN
V irtual P rivate N etwork Virtual private networks are used in connection with public networks for secure data transmission, consequently the entire data traffic is transmitted in encoded form.
VRRP
V irtual R edundant R outer P rotocol A protocol for the control of redundant routers.
W
WAN
W ide A rea N etwork Term for private or public networks that frequently connect several LANs or MANs together.
Web Interface
Term for the interface of a device that enables access to device data over browsers.
WFQ
W eighted F air Q ueuing A process with which queues in a switch are processed when the data is prioritized. This process ensures that all the queues are serviced on the basis of the bandwidths that are allocated to the queues.
Wire Speed
Term for the relaying of data with line speed.
WLAN
W ireless LAN Wireless data transmission in local networks.
WWW
W orld W ide W eb Term for an application in the Internet that enables access to database information through hyperlinks. There are software programs called browsers to view and further process data.
XYZ
Yellow Cable
See 10Base5.
Tuesday, November 27, 2007
IEEE 1588 primer
This comes from Symmetricom www.symmttm.com:
see http://www.symmttm.com/info_center_white_papers.asp
A BRIEF PRIMER ON IEEE 1588 PRECISE TIME PROTOCOL
The recently developed IEEE 1588 Precise Time Protocol (PTP) promises to revolutionize time synchronization by improving accuracy and reducing cost. While certain other precise sync protocols require significant investment in hardware and cabling, PTP makes highly precise timekeeping possible using the most widely deployed medium for network connectivity – Ethernet.
While Ethernet has proven to be a ubiquitous and inexpensive medium for connectivity, it has not been well-suited for applications requiring precise synchronization. By nature Ethernet is nondeterministic, which creates difficulty for real-time or time sensitive applications that require synchronization. On an NTP-based LAN, network devices and computer operating systems add latency and jitter that reduces synchronization accuracy to 1 to 2 milliseconds. Applications requiring greater accuracy have often had to deploy separate cabling systems and dedicated clocks – the IRIG B protocol, for instance, requires a dedicated system of coaxial cables to carry timing signals directly between IRIG B clocks separate from any data network.
PTP overcomes the Ethernet latency and jitter issues through hardware time stamping at the physical layer of the network. The result can be an unprecedented accuracy in the 10 to 100 nanosecond range that is achieved using an Ethernet network to carry the timing packets, allowing for remarkable cost savings.
NTP vs. IRIG vs. PTP
NTP has been the most common and arguably the most popular synchronization solution because it performs well over LANs and WANs and is relatively inexpensive to implement, requiring little in the way of hardware. And while NTP should be able to deliver accuracy of 1 – 2 milliseconds on a LAN or 1 – 20 milliseconds on a WAN, this is far from guaranteed network-wide. This is largely due to the use of switches and routers on LANs and WANs and the fact that many NTP clients run on non-real-time operating systems such as Windows or Linux, that were not designed for time keeping accuracy and therefore rarely deliver it. On Windows, for instance, one can often witness clock corrections of 10 – 50 milliseconds because the system was busy performing tasks it deemed more important than timekeeping.
IRIG time code provides increased accuracy – up to 1–10 microseconds – and is often used where precision timing is mission critical: in military, aerospace and power utility instrumentation. But improved accuracy comes at a cost. IRIG systems eschew Ethernet in favor of dedicated coaxial timing cabling between dedicated hardware clocks, which are not only an added expense but which create an extra burden on the physical infrastructure of the facility.
PTP, on the other hand, offers the cost effectiveness of NTP by using existing Ethernet LANs, and it exceeds the accuracy of the IRIG clocks. PTP can coexist with normal network traffic on a standard Ethernet LAN using regular hubs and switches, and yet provides synchronization accuracy to the sub microsecond level. With the addition of IEEE 1588 boundary clocks or transparent switches, 20-100 nanosecond synchronization accuracy is achievable. The key to this caliber of performance is hardware assisted time stamping.
PTP Usage
Because of its precision, cost-effectiveness and ease-of-use, PTP can be expected to be widely used in military and aerospace applications. Systems that are deployed in military theater situations to identify enemy threats can be constructed from ad hoc Ethernet networks that connect a variety of sensors. Sonar systems in submarines can improve acoustic intercept and ranging accuracy by deploying PTP time stamp technology close to the sensors. And PTP is replacing IRIG systems in aircraft flight test because it provides tighter sync and eliminates the need for additional cabling.
The Future of PTP
PTP has justifiably received considerable attention since its introduction in 2002, and its influence is growing. A variety of vendors are producing hardware that supports PTP, including Intel, which has recently embedded an IEEE 1588 Time Stamping Unit into one of its networking microprocessors. The next version of the 1588 protocol is currently being defined and is expected to increase accuracy even more. And improved fault tolerance and compatibility with SNMP are expected to enhance PTP interoperability with existing network infrastructure. With its nanosecond accuracy, ease-of-deployment, and cost-effectiveness, PTP is poised to transform the landscape of time-synchronized applications in any number of fields.
see http://www.symmttm.com/info_center_white_papers.asp
A BRIEF PRIMER ON IEEE 1588 PRECISE TIME PROTOCOL
The recently developed IEEE 1588 Precise Time Protocol (PTP) promises to revolutionize time synchronization by improving accuracy and reducing cost. While certain other precise sync protocols require significant investment in hardware and cabling, PTP makes highly precise timekeeping possible using the most widely deployed medium for network connectivity – Ethernet.
While Ethernet has proven to be a ubiquitous and inexpensive medium for connectivity, it has not been well-suited for applications requiring precise synchronization. By nature Ethernet is nondeterministic, which creates difficulty for real-time or time sensitive applications that require synchronization. On an NTP-based LAN, network devices and computer operating systems add latency and jitter that reduces synchronization accuracy to 1 to 2 milliseconds. Applications requiring greater accuracy have often had to deploy separate cabling systems and dedicated clocks – the IRIG B protocol, for instance, requires a dedicated system of coaxial cables to carry timing signals directly between IRIG B clocks separate from any data network.
PTP overcomes the Ethernet latency and jitter issues through hardware time stamping at the physical layer of the network. The result can be an unprecedented accuracy in the 10 to 100 nanosecond range that is achieved using an Ethernet network to carry the timing packets, allowing for remarkable cost savings.
NTP vs. IRIG vs. PTP
NTP has been the most common and arguably the most popular synchronization solution because it performs well over LANs and WANs and is relatively inexpensive to implement, requiring little in the way of hardware. And while NTP should be able to deliver accuracy of 1 – 2 milliseconds on a LAN or 1 – 20 milliseconds on a WAN, this is far from guaranteed network-wide. This is largely due to the use of switches and routers on LANs and WANs and the fact that many NTP clients run on non-real-time operating systems such as Windows or Linux, that were not designed for time keeping accuracy and therefore rarely deliver it. On Windows, for instance, one can often witness clock corrections of 10 – 50 milliseconds because the system was busy performing tasks it deemed more important than timekeeping.
IRIG time code provides increased accuracy – up to 1–10 microseconds – and is often used where precision timing is mission critical: in military, aerospace and power utility instrumentation. But improved accuracy comes at a cost. IRIG systems eschew Ethernet in favor of dedicated coaxial timing cabling between dedicated hardware clocks, which are not only an added expense but which create an extra burden on the physical infrastructure of the facility.
PTP, on the other hand, offers the cost effectiveness of NTP by using existing Ethernet LANs, and it exceeds the accuracy of the IRIG clocks. PTP can coexist with normal network traffic on a standard Ethernet LAN using regular hubs and switches, and yet provides synchronization accuracy to the sub microsecond level. With the addition of IEEE 1588 boundary clocks or transparent switches, 20-100 nanosecond synchronization accuracy is achievable. The key to this caliber of performance is hardware assisted time stamping.
PTP Usage
Because of its precision, cost-effectiveness and ease-of-use, PTP can be expected to be widely used in military and aerospace applications. Systems that are deployed in military theater situations to identify enemy threats can be constructed from ad hoc Ethernet networks that connect a variety of sensors. Sonar systems in submarines can improve acoustic intercept and ranging accuracy by deploying PTP time stamp technology close to the sensors. And PTP is replacing IRIG systems in aircraft flight test because it provides tighter sync and eliminates the need for additional cabling.
The Future of PTP
PTP has justifiably received considerable attention since its introduction in 2002, and its influence is growing. A variety of vendors are producing hardware that supports PTP, including Intel, which has recently embedded an IEEE 1588 Time Stamping Unit into one of its networking microprocessors. The next version of the 1588 protocol is currently being defined and is expected to increase accuracy even more. And improved fault tolerance and compatibility with SNMP are expected to enhance PTP interoperability with existing network infrastructure. With its nanosecond accuracy, ease-of-deployment, and cost-effectiveness, PTP is poised to transform the landscape of time-synchronized applications in any number of fields.
************************************
During this offset correction, the master cyclically transmits a unique synchronization (SYNC) message to the related slave clocks at defined intervals (by default every 2 seconds). This sync message contains an estimated value for the exact time the message was transmitted.
For highly accurate synchronization a mechanism is now provided that determines the time of transmission and reception of PTP messages as precisely and as closely as possible to the hardware, best of all directly on the medium.
The master clock measures the exact time of transmission TM1 and the slave clocks measure the exact times of reception TS1. The master then sends in a second message, the follow-up message, the exact time of transmission TM1 of the corresponding sync message to the slave clocks.
On reception of the sync message and, for increased accuracy, on reception of the corresponding follow-up message, the slave clock calculates the correction (offset) in relation to the master clock taking into account the reception time stamp of the sync message. The slave clock Ts must then be corrected by this offset. If there were to be no delay over the transmission path, both clocks would now be synchronous.
The second phase of the synchronization process, the delay measurement, determines the delay or latency between slave and master. For this purpose the slave clock sends a so-called "delay request" packet to the master and during this process determines the exact time of transmission of the message TS3. The master generates a time stamp on reception of the packet and sends the time of reception TM3 back to the slave in a "delay response" packet.
From the local time stamp for transmission TS3 and the time stamp for reception provided by the master TM3, the slave calculates the delay time between slave and master.
The delay measurement is performed irregularly and at larger time intervals (random value between 4 and 60 seconds by default) than the offset measurement. In this way, the network, and particularly the terminal devices, are not too heavily loaded. However, a symmetrical delay between master and slave is crucial for the delay measurement and its precision, i.e. same value for both directions.
Using this synchronization process, timing fluctuations in the PTP elements especially the protocol stack and the latency time between the master and slave are eliminated.
This is from http://www.ieee1588.com/, written by our friends at Hirschmann:
1. IEEE 1588 - Precise Time Synchronization as the Basis for Real Time Applications in Automation
More and more data can be transferred and processed in shorter and shorter time. This trend is observed in automation technology.
Because of this, Ethernet has been chosen as the transport technology for the foreseeable future. Beside speed of transmission and ease of setup, the words “real time” have a special meaning in this field. A substantial facet is precise time synchronization of different end devices.
With the Precision Time Protocol IEEE 1588 there is now, for the first time, a standard available which makes it possible to synchronize the clocks of different end devices over a network at speeds faster than one microsecond.
This article is written to supply answers to several questions: how this protocol works, how to use and implement it, and what results you can expect.
Because of this, Ethernet has been chosen as the transport technology for the foreseeable future. Beside speed of transmission and ease of setup, the words “real time” have a special meaning in this field. A substantial facet is precise time synchronization of different end devices.
With the Precision Time Protocol IEEE 1588 there is now, for the first time, a standard available which makes it possible to synchronize the clocks of different end devices over a network at speeds faster than one microsecond.
This article is written to supply answers to several questions: how this protocol works, how to use and implement it, and what results you can expect.
2. Real-time
If hard real-time requirements are to be met, then the communication system must be able to guarantee deterministic behavior. This means always being able to exchange the required amount of data within a predefined time and the ability to provide mechanisms to synchronize all participants very precisely.
Today only a few special field buses or other proprietary solutions are suitable for achieving cycle times of less than one millisecond or jitter values in the range of a microsecond. To be able to continue the trend towards using Ethernet for networking in automation systems, special measures were created whereby Ethernet can guarantee the required determinism.
There is a degree of variance in Ethernet propagation times during transmission. A solution to guarantee deterministic system behavior is to have a precise clock in all terminal devices synchronized with all other systems. If actions are referred to such a high precision clock, then the process can be decoupled from the propagation times of the communication.
This issue applies particularly to co-operating systems that must start specific actions simultaneously. An example is several robots that work together on one task, e.g. the transport of heavy objects, something that can only function if the movements of the robots are very precisely matched to each other.
If hard real-time requirements are to be met, then the communication system must be able to guarantee deterministic behavior. This means always being able to exchange the required amount of data within a predefined time and the ability to provide mechanisms to synchronize all participants very precisely.
Today only a few special field buses or other proprietary solutions are suitable for achieving cycle times of less than one millisecond or jitter values in the range of a microsecond. To be able to continue the trend towards using Ethernet for networking in automation systems, special measures were created whereby Ethernet can guarantee the required determinism.
There is a degree of variance in Ethernet propagation times during transmission. A solution to guarantee deterministic system behavior is to have a precise clock in all terminal devices synchronized with all other systems. If actions are referred to such a high precision clock, then the process can be decoupled from the propagation times of the communication.
This issue applies particularly to co-operating systems that must start specific actions simultaneously. An example is several robots that work together on one task, e.g. the transport of heavy objects, something that can only function if the movements of the robots are very precisely matched to each other.
3. IEEE1588
The new IEEE Standard Precision Time Protocol (PTP) IEEE1588 is now a very comprehensive solution to do very precise time synchronization in an Ethernet network.
The protocol was originally developed by Agilent for distributed instrumentation and control tasks. The technique is based on the work of John Eidson, who, as chairman of the standardization committee, is largely responsible for the approval of the standard in November 2002.
Using IEEE1588, it is possible for the first time to synchronize, in the sub-microsecond range, the local clocks in sensors, actuators, and other terminal devices using the same Ethernet network that also transports the process data.
Without such a standardized synchronization protocol, which is defined to be used with any protocol, not just Ethernet, it would probably not be possible to synchronize local clocks in terminal devices from different manufacturers with this precision.
Existing time synchronization protocols such as NTP and SNTP do not achieve the required synchronization accuracy or the convergence speed. Others, such as SynUTC from the Technical University in Vienna, were not accepted on the market.
Like other protocols, PTP is based on the most precise matching of times when synchronization packets are transmitted and received possible. Unlike SNTP, the transmission time stamp does not need to be transmitted in the synchronization packet itself, as described in detail below, but is transmitted in a following packet. In this way measurement of transmission and reception, and transmission of measured time stamps can be decoupled.
The protocol was designed for small homogeneous and heterogeneous local networks. The designers paid particular attention to low resource usage so that the protocol can also be used in low end and low cost terminal devices. No special requirements are placed on memory or CPU performance, and only minimal network bandwidth is needed. The low administration effort for this protocol is also significant. As redundant masters are also supported, a PTP domain automatically configures itself using the best master clock algorithm and is also fault-tolerant.
The most important characteristic of the protocol is the synchronism in the microsecond and sub-microsecond range.
The new IEEE Standard Precision Time Protocol (PTP) IEEE1588 is now a very comprehensive solution to do very precise time synchronization in an Ethernet network.
The protocol was originally developed by Agilent for distributed instrumentation and control tasks. The technique is based on the work of John Eidson, who, as chairman of the standardization committee, is largely responsible for the approval of the standard in November 2002.
Using IEEE1588, it is possible for the first time to synchronize, in the sub-microsecond range, the local clocks in sensors, actuators, and other terminal devices using the same Ethernet network that also transports the process data.
Without such a standardized synchronization protocol, which is defined to be used with any protocol, not just Ethernet, it would probably not be possible to synchronize local clocks in terminal devices from different manufacturers with this precision.
Existing time synchronization protocols such as NTP and SNTP do not achieve the required synchronization accuracy or the convergence speed. Others, such as SynUTC from the Technical University in Vienna, were not accepted on the market.
Like other protocols, PTP is based on the most precise matching of times when synchronization packets are transmitted and received possible. Unlike SNTP, the transmission time stamp does not need to be transmitted in the synchronization packet itself, as described in detail below, but is transmitted in a following packet. In this way measurement of transmission and reception, and transmission of measured time stamps can be decoupled.
The protocol was designed for small homogeneous and heterogeneous local networks. The designers paid particular attention to low resource usage so that the protocol can also be used in low end and low cost terminal devices. No special requirements are placed on memory or CPU performance, and only minimal network bandwidth is needed. The low administration effort for this protocol is also significant. As redundant masters are also supported, a PTP domain automatically configures itself using the best master clock algorithm and is also fault-tolerant.
The most important characteristic of the protocol is the synchronism in the microsecond and sub-microsecond range.
4. Many groups are interested in IEEE1588
The biggest interest in determinist Ethernet currently is found in automation; especially motion control applications. Many drive manufactures are equipping their devices with Ethernet interfaces, but now have problems synchronizing all connected drives as precisely as possible over the network.
Several groups of this industrial sector have decided to use this protocol in their Ethernet based field busses. The ODVA has decided to use IEEE1588 for CIPSync, the real-time extension for Ethernet/IP - CIP. Siemens is working on a modification of IEEE1588 for Profinet V3. Also solutions by Beckhoff and Jetter are being developed to ensure time synchronization with this protocol or a similar approach. Likewise the EPSG (Ethernet Powerlink Standardization Group) has planned this protocol as a firm component of the version 3 of their specification.
But interest is not only coming from the automation industry. Increasing demand is growing out of test and measurement; the origin of this protocol. Also initial projects have been started to use IEEE1588 for military applications. Other groups which show interest are coming from telecommunications and electrical power distribution (IEC61850 -Communication networks and systems in substations).
Before we focus on initial results, a short overview of the function of this protocol is given.
Several groups of this industrial sector have decided to use this protocol in their Ethernet based field busses. The ODVA has decided to use IEEE1588 for CIPSync, the real-time extension for Ethernet/IP - CIP. Siemens is working on a modification of IEEE1588 for Profinet V3. Also solutions by Beckhoff and Jetter are being developed to ensure time synchronization with this protocol or a similar approach. Likewise the EPSG (Ethernet Powerlink Standardization Group) has planned this protocol as a firm component of the version 3 of their specification.
But interest is not only coming from the automation industry. Increasing demand is growing out of test and measurement; the origin of this protocol. Also initial projects have been started to use IEEE1588 for military applications. Other groups which show interest are coming from telecommunications and electrical power distribution (IEC61850 -Communication networks and systems in substations).
Before we focus on initial results, a short overview of the function of this protocol is given.
5. How does the IEEE1588 protocol work?
The basic function is that the most precise clock on the network synchronizes all other users. A clock with only one network port is termed an ordinary clock. There are two clocks, Master and Slave. In principle any clock can perform both the master and slave function.
The precision of a clock, more exactly stated of their time sources, is categorized by the protocol in classes (stratum). Here the highest class is an atomic clock which has the stratum value 1. The selection of the best clock in the network is performed automatically using the best master clock algorithm.
The precision of the synchronization depends very heavily on the network and the components used in the network. For this reason the transition over less deterministic components, e.g. routers and switches, is also made possible by the protocol using the boundary clock.
For the administration and configuration of clocks in the network, there is also a management protocol available.
PTP is based on IP multicast communication and is not restricted to Ethernet, but can be used on any bus system that supports multicasting. Multicast communication offers the advantage of simplicity; IP address administration does not need to be implemented on the PTP nodes. Furthermore, PTP can thus be scaled for a large number of PTP nodes.
The basic function is that the most precise clock on the network synchronizes all other users. A clock with only one network port is termed an ordinary clock. There are two clocks, Master and Slave. In principle any clock can perform both the master and slave function.
The precision of a clock, more exactly stated of their time sources, is categorized by the protocol in classes (stratum). Here the highest class is an atomic clock which has the stratum value 1. The selection of the best clock in the network is performed automatically using the best master clock algorithm.
The precision of the synchronization depends very heavily on the network and the components used in the network. For this reason the transition over less deterministic components, e.g. routers and switches, is also made possible by the protocol using the boundary clock.
For the administration and configuration of clocks in the network, there is also a management protocol available.
PTP is based on IP multicast communication and is not restricted to Ethernet, but can be used on any bus system that supports multicasting. Multicast communication offers the advantage of simplicity; IP address administration does not need to be implemented on the PTP nodes. Furthermore, PTP can thus be scaled for a large number of PTP nodes.
6. Time synchronization
Every slave synchronizes to its master's clock by exchanging synchronization messages with the master clock.
The synchronization process is divided into two phases. First the time difference between master and slave is corrected; this is the offset measurement.
Every slave synchronizes to its master's clock by exchanging synchronization messages with the master clock.
The synchronization process is divided into two phases. First the time difference between master and slave is corrected; this is the offset measurement.
During this offset correction, the master cyclically transmits a unique synchronization (SYNC) message to the related slave clocks at defined intervals (by default every 2 seconds). This sync message contains an estimated value for the exact time the message was transmitted.
For highly accurate synchronization a mechanism is now provided that determines the time of transmission and reception of PTP messages as precisely and as closely as possible to the hardware, best of all directly on the medium.
The master clock measures the exact time of transmission TM1 and the slave clocks measure the exact times of reception TS1. The master then sends in a second message, the follow-up message, the exact time of transmission TM1 of the corresponding sync message to the slave clocks.
On reception of the sync message and, for increased accuracy, on reception of the corresponding follow-up message, the slave clock calculates the correction (offset) in relation to the master clock taking into account the reception time stamp of the sync message. The slave clock Ts must then be corrected by this offset. If there were to be no delay over the transmission path, both clocks would now be synchronous.
The second phase of the synchronization process, the delay measurement, determines the delay or latency between slave and master. For this purpose the slave clock sends a so-called "delay request" packet to the master and during this process determines the exact time of transmission of the message TS3. The master generates a time stamp on reception of the packet and sends the time of reception TM3 back to the slave in a "delay response" packet.
From the local time stamp for transmission TS3 and the time stamp for reception provided by the master TM3, the slave calculates the delay time between slave and master.
The delay measurement is performed irregularly and at larger time intervals (random value between 4 and 60 seconds by default) than the offset measurement. In this way, the network, and particularly the terminal devices, are not too heavily loaded. However, a symmetrical delay between master and slave is crucial for the delay measurement and its precision, i.e. same value for both directions.
Using this synchronization process, timing fluctuations in the PTP elements especially the protocol stack and the latency time between the master and slave are eliminated.
7. PTP reference architecture, or how to build a PTP synchronization element
As previously mentioned, what is so special about the architecture is the separation of the time-critical part which is implemented in hardware and the protocol itself and is decoupled from hard real-time conditions - the software part. Thus the protocol is running in a low priority process and/or on a processor with low performance requirements.
The hardware unit consists of a highly precise real-time clock and a time stamp unit (TSU) to generate the time stamp. The software part implements the actual IEEE1588 protocol with the binding to the real-time clock and the HW time stamp unit. Figure 3 illustrates a co-operating of the hard and software component of a IEEE1588 synchronization element.
As previously mentioned, what is so special about the architecture is the separation of the time-critical part which is implemented in hardware and the protocol itself and is decoupled from hard real-time conditions - the software part. Thus the protocol is running in a low priority process and/or on a processor with low performance requirements.
The hardware unit consists of a highly precise real-time clock and a time stamp unit (TSU) to generate the time stamp. The software part implements the actual IEEE1588 protocol with the binding to the real-time clock and the HW time stamp unit. Figure 3 illustrates a co-operating of the hard and software component of a IEEE1588 synchronization element.
The intention of the presented architecture supports an almost OS independent modeling of the software component.
In order to reach this, we introduced three layers with different abstraction level. The Protocol Layer implements the operating system independent Precision Time Protocol. The OS Abstraction Layer forms the interface between PTP and the selected operating system. The functions made available by the operating system - tasks/processes, semaphores, timers, sockets, etc. - are merged over the OS Layer. The following figure shows the interaction of the individual layers.
In order to reach this, we introduced three layers with different abstraction level. The Protocol Layer implements the operating system independent Precision Time Protocol. The OS Abstraction Layer forms the interface between PTP and the selected operating system. The functions made available by the operating system - tasks/processes, semaphores, timers, sockets, etc. - are merged over the OS Layer. The following figure shows the interaction of the individual layers.
The highest layer implements PTP for the synchronization of clocks in a network and can be used on different communication elements (PC, switch, router, etc..). Here is the actual intelligence located for synchronizing the individual communication elements. Within the Protocol Layer we used only ANSI/ISO C conformal functions, thus you can easily transfer the protocol without deep interference into the functionality on different platforms. The protocol dispatchers ensures the atomic execution of functions during an individual process. Communication between the protocol and the OS Abstraction Layer has been realized by a queue and three well defined interfaces.
The middle layer encases operating system dependent functions, which one must adapt, if necessary.
The Timestamp Interface provides the Precision Time Protocol with the seized time stamps of the Sync and Delay-Request messages. However depending upon the stage of development (precision requirement) either a HW unit (TSU) or the software generates the time stamps. The best way to generate "software time stamps” is in the operating system dependent NIC drivers (RX- ISR, sends process) - as near as possible at the transportation medium.
Over the Clock Interface you read and modify the local clock. You have also to adapt these functions depending upon platform. Realizations, which do not have a hard real-time clock, use the system clock of the operating system or optimized solutions as for example the nano-kernel under UNIX derivatives. Apart from setting the local clock, this interface contains the control algorithms which are responsible for the quality of time synchronization [accuracy, stability, transient behavior, etc.].
The Port Interface is used to dispatch and/or receive PTP messages. The IEEE1588 telegrams use excluding UDP/IP multicast packets and thus make it possible to send and receive them over the socket interface of the IP protocol stack. One may neglect temporal requirements, since the time stamps are generated directly at transport medium. The inputs to the protocol (configuration, diagnosis, PTP packets) run over the PTP API.
This modular software platform made it possible for us to build implementations of this protocol for Linux, Windows and VxWorks. The implementations in Windows and Linux use time stamping in software. But even a pure software implementation reaches a precision of about 100 µs, and it seems to be possible to increase this up to a precision of better than 10 µs.
The middle layer encases operating system dependent functions, which one must adapt, if necessary.
The Timestamp Interface provides the Precision Time Protocol with the seized time stamps of the Sync and Delay-Request messages. However depending upon the stage of development (precision requirement) either a HW unit (TSU) or the software generates the time stamps. The best way to generate "software time stamps” is in the operating system dependent NIC drivers (RX- ISR, sends process) - as near as possible at the transportation medium.
Over the Clock Interface you read and modify the local clock. You have also to adapt these functions depending upon platform. Realizations, which do not have a hard real-time clock, use the system clock of the operating system or optimized solutions as for example the nano-kernel under UNIX derivatives. Apart from setting the local clock, this interface contains the control algorithms which are responsible for the quality of time synchronization [accuracy, stability, transient behavior, etc.].
The Port Interface is used to dispatch and/or receive PTP messages. The IEEE1588 telegrams use excluding UDP/IP multicast packets and thus make it possible to send and receive them over the socket interface of the IP protocol stack. One may neglect temporal requirements, since the time stamps are generated directly at transport medium. The inputs to the protocol (configuration, diagnosis, PTP packets) run over the PTP API.
This modular software platform made it possible for us to build implementations of this protocol for Linux, Windows and VxWorks. The implementations in Windows and Linux use time stamping in software. But even a pure software implementation reaches a precision of about 100 µs, and it seems to be possible to increase this up to a precision of better than 10 µs.
8. IEEE1588 and Switches
The precision of the protocol also depends on the latency jitter of the underlying network topology. Point to point connections provide the highest precision, with hubs imposing very little network jitter. Under very low or no network load, Layer 2 switches have a very low processing time, typically 2 to 10µs plus packet reception time. In this case, new switch designs also have low latency jitter e.g. the Hirschmann Switch RS2-FX/FX with about 0.4 µs latency jitter.
But switches are working with queues and store and forward, so only one queued maximum length packet imposes a delay for the following packet of about 122µs, and under high load conditions, more than one packet will be in the queue. The next issue for the precision of the protocol is that latency is completely symmetric for both directions: from the master to the slave and visa verse. This can nearly never be guaranteed under higher network loads.
Prioritization of packets e.g. IEEE802.1D/p does not really solve the problem, because at least one long packet can be in front of a synchronization packet and so will impose up to 122µs to the jitter of transmission. Currently, available switches show that after the priority scheduler, there is another queue for 2 up to 8 packets, and not only one as expected. This means a jitter from 360µs up to 1ms under heavy load conditions.
The solution for all these problems is the usage of IEEE 1588 Boundary clocks in switches. In this case you only have point to point connections and there is nearly no delay jitter between master clock and slave clock and internal queuing delay/jitter of switches is not relevant any more.
The precision of the protocol also depends on the latency jitter of the underlying network topology. Point to point connections provide the highest precision, with hubs imposing very little network jitter. Under very low or no network load, Layer 2 switches have a very low processing time, typically 2 to 10µs plus packet reception time. In this case, new switch designs also have low latency jitter e.g. the Hirschmann Switch RS2-FX/FX with about 0.4 µs latency jitter.
But switches are working with queues and store and forward, so only one queued maximum length packet imposes a delay for the following packet of about 122µs, and under high load conditions, more than one packet will be in the queue. The next issue for the precision of the protocol is that latency is completely symmetric for both directions: from the master to the slave and visa verse. This can nearly never be guaranteed under higher network loads.
Prioritization of packets e.g. IEEE802.1D/p does not really solve the problem, because at least one long packet can be in front of a synchronization packet and so will impose up to 122µs to the jitter of transmission. Currently, available switches show that after the priority scheduler, there is another queue for 2 up to 8 packets, and not only one as expected. This means a jitter from 360µs up to 1ms under heavy load conditions.
The solution for all these problems is the usage of IEEE 1588 Boundary clocks in switches. In this case you only have point to point connections and there is nearly no delay jitter between master clock and slave clock and internal queuing delay/jitter of switches is not relevant any more.
9. Test results
In the beginning, we directly connected in our test setup two Ordinary Clocks. We used IEEE1588 enhanced plug-in modules on our modular Ethernet switches (MICE series). To put the protocol through its paces, we added a high network load with an Ethernet packet generator.
So that we could examine the difference between local clock and reference clock as close-to-applications as possible, we enhanced both units with a Pulses Per Second (PPS) signal output and connected them to an oscilloscope. Thus we could seize now very elegantly the deviation of the two signals and also represent the frequency distribution of the deviation. The synchronization accuracy which we reached lay within ± 100 ns (max. jitter). The measurement ran for a period of 84 hours.
In the beginning, we directly connected in our test setup two Ordinary Clocks. We used IEEE1588 enhanced plug-in modules on our modular Ethernet switches (MICE series). To put the protocol through its paces, we added a high network load with an Ethernet packet generator.
So that we could examine the difference between local clock and reference clock as close-to-applications as possible, we enhanced both units with a Pulses Per Second (PPS) signal output and connected them to an oscilloscope. Thus we could seize now very elegantly the deviation of the two signals and also represent the frequency distribution of the deviation. The synchronization accuracy which we reached lay within ± 100 ns (max. jitter). The measurement ran for a period of 84 hours.
The following figure shows the frequency distribution of the offset values between master clock and slave clock in nanoseconds. The standard deviation amounts to 23.95 ns and the average value -4.248 ns.
The drift values of the oscillators limits the synchronization accuracy with the available prototypes. Quartz frequency of 50 MHz (± 50 ppm) results in a dissolution of 20 ns. Thus the system can adjust the drift within the range of ± 20 ns per second. If you now regard the relative drift between the local clock and the master clock during two other following synchronization telegrams, then it becomes clear that the short-term stability of the oscillators substantially determines the synchronization accuracy in the in-swing condition.
10. Conclusion
The Precision Time Protocol standardized in IEEE1588 reaches synchronization accuracy within the sub-microsecond range and has further potential for higher precision. It is suited for applications which need a time synchronization of distributed clocks of highest accuracy in a limited network domain.
Many manufacturers have already begun the development of appropriate components and have already started to evaluate their first prototypes.
For high precision in a switched Ethernet network, it is recommended to equip switches with IEEE1588 technology.
The PTP reference architecture introduced in this article tries to unite the advantages of equipment modularity and scalability.
The Precision Time Protocol standardized in IEEE1588 reaches synchronization accuracy within the sub-microsecond range and has further potential for higher precision. It is suited for applications which need a time synchronization of distributed clocks of highest accuracy in a limited network domain.
Many manufacturers have already begun the development of appropriate components and have already started to evaluate their first prototypes.
For high precision in a switched Ethernet network, it is recommended to equip switches with IEEE1588 technology.
The PTP reference architecture introduced in this article tries to unite the advantages of equipment modularity and scalability.
About the author:
Dirk S. Mohl is currently head of development for Industrial Ethernet products at Hirschmann Automation and Control, Germany. He received a Diploma of Engineering degree in Electrical Engineering from the Technical University of Stuttgart, Germany, in 1991. He started work in 1992 at Hirschmann in the Networking Department as a hardware and software development engineer. His main interest is Ethernet and Switching (Layer 2 and Layer 3) with focus on industrial applications.
Dirk S. Mohl is currently head of development for Industrial Ethernet products at Hirschmann Automation and Control, Germany. He received a Diploma of Engineering degree in Electrical Engineering from the Technical University of Stuttgart, Germany, in 1991. He started work in 1992 at Hirschmann in the Networking Department as a hardware and software development engineer. His main interest is Ethernet and Switching (Layer 2 and Layer 3) with focus on industrial applications.
Subscribe to:
Posts (Atom)