# Channel Access Method **A method that allows for several device to share and transmit over the same transmission link.** A **channel access method** or **multiple access method** is a method which allows *multiple devices* connected by the same transmission link to transmit over it and *share its capacity*. It is similar in concept to [[multiplexing]] but is provided by the [[Internet Protocol Suite#Link layer|link layer]] instead of the physical layer of network infrastructure. Channel access methods are used to avoid *collisions*, instances where two or more devices are attempting to transmit over the same transmission link unregulated thus causing interference. Channel access methods can be combined with channel access protocols to form the *[[medium access control]] sublayer* of the link layer, which controls the access of hardware to connected transmission links. Channel access methods are categorised into two main types, channel-based and packet-based, which can be seen as analogous to [[Circuit Switching|circuit switched]] and [[Packet Switching|packet switched]] multiplexing. ## Types ### Channel-based In channel-based multiple access, multiple devices are simultaneously given access to a shared transmission link by *partitioning the bandwidth* of the link. These methods are based largely on circuit switching multiplexing methods. #### Frequency-division multiple access **Frequency-division multiple access** or **FDMA** is based on [[frequency-division multiplexing]] and involves dividing the bandwidth of the transmission into *distinct frequency ranges*, allowing each device full access to the bandwidth of its frequency range for the duration of its connection. #### Time-division multiple access **Time-division multiple access** or **TDMA** is based on [[time-division multiplexing]] and involves assigning the entire transmission link to each device *periodically within a time slot*. Each device can thus utilise the full bandwidth of the link but only during its time slot. ### Packet-based In packet-based multiple access, each device is given full access to a shared transmission link *for the duration of a packet transmission*. These channel access methods may be further characterised by whether they are collision-free or rely on collision avoidance or recovery. - *Collision-free* - only one device at any given time is allowed to transmit. - *Collision avoidance* - transmissions are initiated such that collisions are avoided. - *Collision recovery* - collisions are detected and handled if they occur. #### Carrier-sense multiple access **[[Carrier-sense multiple access]]** or **CSMA** is a *collision avoidance* channel access method in which a device checks if other devices are transmitting on the link before starting its own transmission. Carrier-sense multiple access has also been modified with additional collision avoidance and collision detection techniques in CSMA/CA and CSMA/CD. #### ALOHA **[[ALOHA]]** or **pure ALOHA** is a *collision recovery* channel access method in which a device attempts to transmit whenever they need to. If a collision occurs, then all devices reattempt transmission later. ALOHA has also modified into slotted and reservation ALOHA. #### Polling In **polling**, a master device checks every device connected to the shared transmission link for whether or not they are ready to transmit. If a device when checked is ready to transmit the master device then allows only that device to transmit. #### Token passing **Token passing** is a *collision-free* channel access method in which a *token* which allows a device to transmit is passed sequentially between the devices connected to the transmission link. Unlike in polling, token passing does not require a master device to dictate who is allowed to transmit. However, it may require a protocol to pass the token between devices, although it is possible for this to be implicit and thus not requiring a token.