CSMA Full Form

What is the full form of CSMA?

CSMA stands for Carrier-Sense Multiple Access. The medium access control (MAC) protocol known as carrier-sense multiple access (CSMA), a node first confirms that there is no other traffic present before broadcasting on a shared transmission medium like an electrical bus or a band of the electromagnetic spectrum. In accordance with CSMA, a transmitter first checks to see if another transmission is already in progress using a carrier-sense method. In other words, it checks to see if a carrier signal from another node is there before attempting to broadcast. If a carrier is detected, the node waits for the current transmission to finish before starting a new one. Multiple nodes may send and receive on the same media turn wise using CSMA.

This Story also Contains

1. What is the full form of CSMA?
2. Types of CSMA Access
3. Signal Jam
4. Late Collision
5. Local Collision
6. Application of CSMA

All other nodes linked to the medium typically receive transmissions sent by a single node. The addition of collision-avoidance (CSMA/CA), collision-detection (CSMA/CD), and collision-resolution techniques are variations on the fundamental CSMA.

Types of CSMA Access

Different CSMA variants employ various methods to choose when to begin transmission onto the shared medium. These algorithms' aggressiveness or persistence in commencing transmission is a crucial differentiator. A more aggressive algorithm can start transmitting more rapidly and use a larger portion of the medium's available bandwidth. Usually, this comes with a higher chance of colliding with other transmitters.

• 1-persistent: An aggressive transmission algorithm is 1-persistent CSMA. The transmitting node detects whether the transmission medium is idle or busy when it is prepared to send. If idle, it communicates right away. If busy, it constantly senses the transmission medium until it is idle, at which point it sends the frame of the message unconditionally (i.e., with a probability equal to 1). If there is a collision, the sender waits for an arbitrary amount of time before trying the same method again. In CSMA/CD systems, such as Ethernet, 1-persistent CSMA is employed.

• Non-persistent: A non-aggressive transmission algorithm is non persistent CSMA. The transmitting node detects whether the transmission medium is idle or busy when it is prepared to transfer data. If idle, it communicates right away. If the channel is busy, it skips straight to the 1-persistent CSMA's final random waiting step before restarting the entire logic cycle once more. This avoids monitoring the busy channel repeatedly while trying to send a message, hence the name. This method, compared to 1-persistent, results in a greater initial delay but lowers the likelihood of collision and increases medium throughput overall.

• P-persistent: The transmitting node detects whether the transmission channel is idle or busy when it is ready to send data, falling between the 1-persistent and non-persistent CSMA access modes. If idle, it communicates right away. If it's busy, it continuously senses the transmission channel until it is idle, at which point it transmits with probability p. The node waits until the following time slot is open if it decides not to transmit (the likelihood of this happening is 1-p). It transmits again with the same probability p if the transmission medium is not busy. This probabilistic hold-off continues until the frame is eventually transferred or the medium is discovered to be busy once more (i.e. some other node has already started transmitting).

• O-persistent: A supervisory node assigns a transmission order to each node. Nodes wait for their time slot according to the transmission order they were given when the transmission medium went idle. The node designated to transmit initially does so right away. A time slot is waited by the node designated to send the second (but by that time the first node has already started transmitting). Nodes keep an eye out for broadcasts from other nodes and adjust their allocated order whenever one is found (i.e. they move one position closer to the front of the queue). The controller area network, LonWorks, and CobraNet all employ O-persistent CSMA.

Signal Jam

The jam signal, also known as a jamming signal, is a signal that carries a 32-bit binary pattern that is broadcast by a data station to alert other transmitting stations to a collision and to instruct them to stop transmitting. This is how the maximum jam time is determined: An Ethernet system can only have a maximum diameter of 232 bits. This results in a 464-bit round-trip time. Due to Ethernet's 512-bit slot time, the maximum jam-time is 48 bits (6 bytes), which is the difference between the slot time and round-trip time. This thus implies: A station delivering a 4 to 6-byte long pattern with 16 1-0 bit combinations after detecting a collision does so. Note: This jam signal is obviously larger than what is permitted.

In order to prevent any other node from now receiving a frame from receiving the jam signal instead of the correct 32-bit MAC CRC, this is done. As a result, any other receivers will discard the frame because of a CRC problem.

Late Collision

A late collision is a collision type that occurs later in the packet than the protocol standard in question permits. A late collision is stated to have happened in 10-megabit shared-medium Ethernet if a collision error occurs after the transmitting station has transferred the first 512 bits of data. Importantly, unlike collisions that happen before the first 64 octets, late collisions are not re-sent by the NIC; instead, it is up to the higher layers of the protocol stack to assess whether there was data loss.

Since a properly configured CSMA/CD network link shouldn't experience late collisions, full-duplex/half-duplex mismatch, exceeding Ethernet cable length restrictions, or damaged hardware, such as improper cabling, an unallowable number of network hubs, or a bad NIC, are the most common causes.

Local Collision

A local collision is one that takes place at the NIC rather than on the wire. Without making an informational transmission, a NIC cannot identify local collisions. Only when a station detects a signal on the RX pair while it is simultaneously sending on the TX pair will a local collision on UTP cable be identified on the local segment. There is no difference in the signal's characteristics because the two signals are on distinct pairs. On UTP, collisions can only be detected when the station is running in half-duplex. In this aspect, the only functional distinction between half-duplex and full-duplex operation is whether or not the transmit and receive pairs are allowed to be used concurrently.

Application of CSMA

CSMA/CD was utilised in the early iterations of twisted-pair Ethernet, which made use of repeater hubs, as well as in the now-outdated shared media Ethernet variations (10BASE5, 10BASE2). Due to the isolation of each Ethernet segment, or collision domain, modern Ethernet networks constructed with switches and full-duplex connections no longer require the usage of CSMA/CD. For connections that are half-duplex and backward compatible, CSMA/CD is still supported. Due to historical reasons, the name "Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications" continued to be used for the IEEE 802.3 standard, which defines all Ethernet variants, until 802.3-2008, which uses the new name "IEEE Standard for Ethernet."