Next we want to address two common ways that nodes or computers use to access the network directly. In other words, to be able to get onto the network not necessarily to get resources but to actually access the physical network itself. The two ways we’re going to look at are both c methods which were briefly mentioned in a previous video. Basically, in a Contention-Based Network Access method the computers compete for network access. There are two types of contention-based access that are commonly used in modern networks. One is CSMA/CA and CSMA/CD and we will discuss there operations.
One of those is the Carrier Sensing Media Access/Collision Detection Access Method. Also called CSMA/CD. This method is most commonly used by wired Ethernet insulations. The other contention method that’s commonly used is Carrier Sensing Media Access/Collision Avoidance. Also called CSMA/CA. This access method is most commonly going to be seen in Wi-Fi networks.
A fun fact about Wi-Fi is that the term Wi-Fi actually doesn’t mean anything specifically. The person invented Wi-Fi, when asked why he called it Wi-Fi simply said, “I thought it sounded cool “so that’s what I called it.” Let’s go ahead and look at the CSMA/CD contention method first. In a CSMA/CD contention method, basically what happens is each node or computer on the network listens for traffic going by on the network. If it doesn’t hear any traffic then the node will release a packet onto the network.
However, if two nodes were listening to the network at the same time and neither one of them hear any traffic on the network they will both release their packet at the same time. When this happens the packets end up hitting each other and we have what is called a collision. The result of a collision is basically when the packets hit each other, they cause a power spike that is then heard by all the nodes on the network. Also, when two packets collide with each other they destroy the data those contain in both the packets.
CSMA/CD collision Detection
This diagram here illustrates what a collision would look like. Basically what happened is Computer 1 and Computer 5 both listen to the network and decided to send the packet at the same time.
Well, when they did their packets ended up hitting each other, somewhere in the middle of the network, causing a collision, and therefore, the data was lost and the collision was heard by all the computers on the network. If a collision doesn’t take place then CSMA/CD considers that the transmission was successful and now the network is freed up for another node to transmit.
When a collision doesn’t take place this is generally how a CSMA/CD Operation would take place. Computer 1 needs to send a packet to Computer 3. Therefore, Computer 1 will listen to the network, if it doesn’t hear any traffic on the network it will then release its packet onto the network and the pack will then travel down the network to Computer 3. Once the packet gets to Computer 3 the other computers will have heard the packet go by and they’ll know that that was now freed up for them to send their own packets.
CSMA/CD collision Response
And so, the contention starts all over again. If a collision does occur, then all the nodes on the network will hear it and they’ll start an internal clock that’s set to a random number of milliseconds. Once a clock on a specific node or computer runs down to zero then that particular node or computer will attempt to send a packet again. This diagram, here, illustrates what happened. A collision took place and so a timer or a clock on each one of the computers on the network randomly generated a number of milliseconds and set the clock to that number of milliseconds.
Now each node has a different number of milliseconds on their clock. And so as each nodes clock runs down they then attempt to send their packet and this way is hope that they will avoid any further collisions. The other type of contention method I want to look at is CSMA/CA. The CSMA/CA works very similarly to the CSMA/CD. The main difference is, instead of releasing the packet on the network when it hears the network is clear it releases a warning packet on the network.
That warning packet basically runs down the network and basically warns all the other nodes that a packet is about to be transmitted. When the other nodes on the network hear the warning packet they won’t transmit until after they hear the data packet go by. Once they hear the data packet go by they’ll then attempt to transmit their own warning packet before they transmit their data packet. However, if two warning packets are transmitted at the same time, just like in the previous example where two data packets were transmitted at the same time, again, it will result in a collision.
And the collision is handled exactly the same way as with the CSMA/CD method. This diagram here, illustrates how the CSMA/CA method works. Computer 1 again, wants to send a packet to Computer 3 and so Computer 1 releases a warning on the network. All the computers on the network hear the warning go by and so they refrain from trying to transmit a warning packet of their own.
Once the warning goes by and everybody has a chance to hear it Computer 1 will then release its data packet onto the network. The data packet will move down the network to a computer seeing and as each node hears the data packet go by they will then try to release their own warning packet first so they have the opportunity to send the next data packet.
Latest posts by Hamza Arif (see all)
- Wide area networks (WAN) & Metropolitan area networks (MAN) - August 18, 2018
- Understanding the Internet, intranets, and extranets - August 17, 2018
- Exploring network collisions, CSMA/CD, and CSMA/CA - August 16, 2018