Automotive Ethernet Flashcards
Ethernet PHY
Physical Layer
A couple element is needed to have a point-to-point connection, which is known as an Ethernet PHY. It allows the connection of the physical medium to the Ethernet Controller in the Micro Controller. The MDI (Medium Dependent Interface) depends of the connection to the physical layer. It is the physical connection (e.g. a twisted-pair connection 100Base-T1). On the other end of the PHY is the MII (Medium Independent Interface) which connects the PHY to the Micro Controller (the Ethernet controller in the micro controller) thinks in a bit-stream. The MDI thinks in a symbol stream In Ethernet, the PHY converts a bit-stream into a symbol stream. So the PHY is not only an amplifier like a transceiver for CAN, it's a real component with a coding and decoding mechanism. This converts the bit-stream to symbol streams and back again.
IEEE 100Base-T1 (formerly OABR)
Physical Layer
IEEE 100Base-T1 (formerly OABR):
Automotive Ethernet, formerly known as Open Alliance Broader Reach. It uses one channel (T1 = twisted pair channel).
Different voltages represent symbols and the symbols represent the bit-stream.
Coding and decoding mechanisms are needed and T1 uses a combination of 4B/3B (four bit to 3 bit), 3B2T (3 bit to 2 Turnery) and PAM3 (Pulse-Amplitude-Modulation with three states).
On an Ethernet, the network is never silent, there is a continuous communication on the two lines (duplex) the Master constantly transmits sychronisation symbol streams when the Master doesn’t have any data to transmit.
IEEE 1000Base-T1
Physical Layer
IEEE 1000Base-T1:
A gigabit Ethernet can transmit 1GB/s in full duplex. Coding uses an 80B/81B conversion - the difference is that this uses a parity bit for the FEC (Forward Error Correction), the 81st bit is created as a parity bit which then allows the receiver to correct any errors in the symbol stream. This allows the receiver to recreate the original bit-stream if any value is disturbed during transmission.
As with 100Base-T1, a continuous synchronisation is in use.
IEEE 100Base-TX
Physical Layer
IEEE 100Base-TX:
Often used in the automotive industry to connect the car to the outside world (preferred for outside communication). A use case example is the diagnostic over IP - it’s possible to connect a tester via traditional Ethernet connection to the car, the tester can be any kind of PC or laptop or any integrated tester in the lab.
Makes use of two channels, one to transmit and one to receive - a node has a transmission (Tx) and a reception (Rx) path. This means you need four pins.
Coding of the bit-stream uses a combination of NRZI (Non-Return-to-Zero-Inverted: the original 4-bit-stream packet is inverted so that if you have a logical zero as a bit value, it becomes a 1 and visa-versa) with 4B/5B (the inverted bit-stream receives a fifth bit - in CAN it’s a stuffed bit - and MLT-3 (…and this bit-packet is transmitted with 3 different states - -1, 0, +1).
It’s not the differential voltage itself that you want to transmit but the rising and falling edges tells you if you’re transmitting a logical 1 (not rising or falling edge? then you’re transmitting a logical 0).
The transmitting path is responsible for creating the sychronisation symbol stream.
IEEE 1000Base-T
Physical Layer
IEEE 1000Base-T:
This is a technology that requires 4 channels, with each channel supporting a transmission rate of 250Mb/s. As it’s using a PAM5 (Pulse-Amplitude-Modulation with three states with 5 states - -2, -1, 0, +1, +2), it’s available in full-duplex.
1000Base-T needs clock synchronisation for the PAM5, a master-slave is in use for gigabit Ethernet (the ECU’s decide between themselves which one has the better precision and that ECU becomes the master - this can change, depending on the temperature of the oscillator)
IEEE Ethernet MAC + VLAN
Layer 2, Ethernet Controller
Ethernet Medium Access Control + Virtual LAN:
For addressing, It’s important to note that Ethernet thinks in Node addressing (each node gets it’s own address - known as a MAC-address) - it’s important to find out from the OEM which MAC variant is used by them (either a supplier MAC address management or whether the MAC addresses should be defined by the OEM).
The Virtual LAN concept is very well established in the automotive industry. Many manufactuers like to implement a domain thinking in their Ethernet networks - meaning for different use cases it’s possible to define own virtual networks where communication between nodes is limited to a certain use case. Diagnostics is a typical example for making use of Virtual LAN. It also offers a little bit more security, limiting access to vehicle information from outside the car.
IEEE
IEEE:
Institute of Electrical and Electronics Engineering
Internet Protocol - two versions IPv4/IPv6
Layer 3
Internet Protocol:
Is a software layer that allows the use of addresses allowing routing of data packets across network boundaries (outside of the network). So this makes it possible to access nodes on different networks via IP addresses. The important couple-element for IP communication is a router, which allows the connection of a number of networks to converge on the router.
It’s possible to implement this network in a connected gateway which in turn is able to get a connection to the internet via, for example, a mobile connection.
In a vehicle today (2017), IP addresses are used in addition to PCP and UDP for addressing (see Layer 4) The two versions IPv4 (four byte addresses, 32 bit) /IPv6 (sixteen byte addresses, 128 bit - the main difference is a longer address), both are available and it’s an OEM decision which variant is applied.
TCP/UDP
Layer 4
TCP/UDP:
Both protocols are used as transport protocols on IP. These two transport protocols are available on Layer 4 because they support two transmission philosophies. What is the difference? TCP allows a connection oriented connection, meaning a communication channel exists between the two communicating nodes. Communication between these two nodes needs to be established so data can be transmitted through this channel.
UDP, on the other hand, is connectionless, meaning a packet can be transmitted without establishing a connection.
PRO’s of UDP:
It’s connectionless, there’s no channel allowing transmission of information that’s multicast and broadcast
PRO’s of TPC:
It allows the acknowledgement of information, providing the transmitter with feedback on the correct reception of the transmission.
It’s more reliable because of this acknowledgement
CON’s of TPC:
Is capable only of unicast communication, requiring a channel between two nodes and this channel needs to be established.
So it depends on the use-case which of these transport protocols are used; if you need multicast, UDP will be used, if you need reliability TPC is the one to use.
Both protocols use additional port addressing on Layer 4 (remember, Layer 2 = MAC addresss, Layer 3 = IP addresses of nodes). Ports, on the other hand, are used for addressing a functionality of a node
BroadR-Reach a.k.a. UTSP Ethernet a.k.a. OPEN Alliance BroadR Reach
BroadR-Reach technology is an Ethernet physical layer standard designed for use in automotive connectivity applications. BroadR-Reach technology allows multiple in-vehicle systems to simultaneously access information over unshielded single twisted pair cable (UTSP). Benefits for automotive manufacturers integrating the BroadR-Reach Ethernet standard include reduced connectivity costs and cabling weight, according to Broadcom Corporation, now Boradcom LImited, inventor of the BroadR-Reach standard.
Also its use is not limited, so it doesn’t matter whether the application is diagnostics, driver assistance, infotainment, or something else. This means that Ethernet-based communication is flexible interms of applications, speed grades, and in terms of requirements that are brought into the car from the outside world.
Modern Ethernet-based networks function via switches. This means that the available bandwidth is not necessarily shared, especially as the topology is not predefined by the technology but can be chosen to suit the specific situation best.
Deterministic Ethernet
Deterministic Ethernet refers to a networked communication technology that used time scheduling to bring deterministic real-time communication to standard Ethernet. Deterministic Ethernet can be used in a wide range of applications where guaranteed latency is vital, either for reasons of operational efficiency or functional safety. These applications include autonomous driving, machine-to-machine communication,and aerospace flight control. Parts of this technology has been standardized as TSN by the Institute of Electrical and Electronics Engineers (IEEE), and as Time-Triggered Ethernet by the Society of Automotive Engineers (SAE).
TSN
Time-Senstive Networking is a set of Ethernet standards that enable deterministic real-time communication over Ethernet. In a TSN network, the latency of critical scheduled communication is guaranteed. In combination with other TTTech technologies, TSN enables open data exchange between industrial controllers from different vendors, making the vision of open, real-time machine-to-machine communication a reality for all applications, including those with critical safety requirements.
Time-Triggered Ethernet
Time-Triggered Ethernet is a scalable networking technology that uses time scheduling to deliver deterministic real-time communication other Ethernet. It was specifically designed for safe and highly available real-time and fault-tolerant applications that require certifications, such as in aerospace applications.
PHY
PHY is an abbreviation for the physical layer of the OSI model and refers to the circuitry required to implement physical layer functions.
A PHY connects a link layer device (often called MAC as an abbreviation for medium access control) to a phzysical medium such as an optical fiber or copper cable. A PHY device typically includes both Physical Coding Sublayer (PCS) and Physical Medium Dependent (PMD) layer functionality.
