What is Ethernet and how does it work?

Ethernet is the typical method used for creating wired networks, either LAN (local area network) or WAN (wide area network). We have all used Ethernet at one point or another to connect devices together and enable communication, but how does this work?

Networks have 7 layers, which are defined as the following:

1. Physical Layer
2. Data Link Layer
3. Network Layer
4. Transport Layer
5. Session Layer
6. Presentation Layer
7. Application Layer

This article will focus on layers 1 and 2, which make up ‘Ethernet’, as defined by IEEE 802.3.

Physical Layer

There are a number of different variants of the physical layer, but the most commonly used is a category 5 or 6 cable, with an 8P8C (RJ45) connector. These can be made up of coax cable, twisted pair cabling, or optical fibre. Usually, we see twisted pair cabling for the ‘standard’ style home network which is 10BASE-T or 100BASE-T.

Data Link Layer

The Ethernet Data Link Layer is split into two parts, the Logical Link Control (LLC) and the Media Access Control (MAC). LLC is responsible for routing the data between devices or nodes, and the MAC address is done using hardware addresses assigned via Network Interface Cards (NIC) to identify individual devices on a network.

Ethernet has two units of data transmission and receiving; packets and frames. Data sent over an Ethernet network is split into frames, which contain both a header and a payload.

The header is 14 bytes, compiled with 3 parts:

• The first 6 bytes are the destination MAC address
• The second set of 6 bytes is the source MAC address
• And the final two bytes are the length.

The data or payload section of a frame is where real data is contained. This can be 46 – 1500 bytes long.

Each frame is wrapped in a packet which contains several pieces of information to correctly establish a connection with another device, and also to indicate where each frame starts and stops.

There is a crucial algorithm running underneath the data link layer, called Carrier Sense Multiple Access with Collision Detection (CSMA/CD). CSMA/CD is standard in Ethernet networks, and is used to reduce data collisions and increase success rates… but how exactly does this work?

The algorithm will send a bit of information onto the network to see if any collisions occur. If this first bit does not collide, then it will send the rest of the bits out, but will also continue to monitor for collisions whilst doing so.

If there is any data collision, then the algorithm will stop sending data, wait for a calculated period of time, and then make a second attempt at sending the data from the first bit again until they are all sent with no collisions.

Why have Actisense implemented Ethernet usage?

Our PRO-Range buffers and multiplexers have long been able to be configured via ethernet. Over the past few years, boaters and technicians have requested the ability to stream NMEA 0183 data over ethernet. We have now applied this function to the PRO-BUF-2, PRO-NDC-1E and the PRO-MUX-2.

Ethernet has long existed as a standard onboard many vessels, with the ability to transfer large amounts of data in a fast, precise way, allowing onboard devices and instruments to communicate with each other in an efficient way.