Media Access Control (MAC) - Multiple Access Protocols
Multiple Access Protocols
The Data-Link Layer is divided into two sublayers:
Data-Link Layer │ ├── Data Link Control (DLC) │ └── Media Access Control (MAC)
The previous section discussed DLC. This section deals with MAC, particularly Multiple Access Protocols.
1. Why Do We Need Multiple Access Protocols?
When a link is dedicated, only two devices use the link. For example, a dedicated telephone line connects two endpoints.
In this situation, we mainly need a data-link-control protocol, such as PPP, to manage data transfer.
But consider a shared link:
Shared Medium ─────────────────── │ │ │ Node A Node B Node C
Several nodes use the same communication medium.
If Node A and Node B transmit at the same time, their signals may interfere.
This is called a:
Collision
Therefore, when several nodes share a common link, we need a multiple-access protocol to coordinate access to the medium.
2. What is a Multiple-Access Protocol?
A multiple-access protocol is a protocol used to coordinate the access of multiple nodes to a shared or broadcast link.
We can compares this situation with people speaking in an assembly.
The rules should:
- Give each node an opportunity to access the link.
- Prevent two nodes from transmitting at the same time.
- Handle collisions if they occur.
- Prevent one node from monopolizing the medium.
Thus, there are two important goals:
Goal 1: Prevent collisions
Node A ──┐ ├── Shared Link Node B ──┘ A and B should not transmit simultaneously.
Goal 2: Handle collisions
If two nodes transmit simultaneously and a collision occurs, the protocol must provide a method to deal with it.
These are the fundamental problems addressed by MAC protocols.
3. Dedicated Link vs Shared Link
| Dedicated Link | Shared Link |
|---|---|
| Link is used by two endpoints | Link is shared by multiple nodes |
| No competition for the medium | Nodes compete for access |
| DLC protocol is sufficient | Multiple-access protocol is required |
| Example: dedicated telephone connection | Example: shared wireless/cellular medium |
Consider the example of a cellular phone system where the communication band is shared by different users.
4. Types of Multiple-Access Protocols
The multiple-access protocols are divided into three groups.
Multiple-Access Protocols │ ┌────────────┼────────────┐ │ │ │ ▼ ▼ ▼ Group 1 Group 2 Group 3
The three groups are:
- Random Access Protocols
- Controlled Access Protocols
- Channelization Protocols
These groups provide different ways of coordinating access to a shared medium.
5. Basic Idea of the Three Groups
1. Random Access
In random-access methods, nodes compete for access to the shared medium.
A node can attempt transmission when it has data to send.
Because multiple nodes may attempt to transmit, collisions can occur. The protocol therefore also provides a way to handle collisions.
Shared Medium / | \ A B C │ │ │ └── compete ────┘
2. Controlled Access
In controlled-access methods, nodes do not simply compete randomly.
Access to the medium is controlled or coordinated so that nodes take turns using the shared medium.
The objective is to avoid collisions by controlling who can transmit.
Conceptually:
Node A → transmit Node B → wait Node C → wait Node A finishes Node B → transmit Node A → wait Node C → wait
3. Channelization
In channelization, the shared medium is divided so that different users can be given separate portions of the available communication resource.
Conceptually:
Shared Medium │ ┌──────────┼──────────┐ │ │ │ Part 1 Part 2 Part 3 │ │ │ Node A Node B Node C
This allows multiple nodes to use the medium without directly competing for exactly the same resource at the same time.
6. Overall Concept
The main idea of the entire MAC section can be summarized as:
Shared Link │ ▼ Multiple nodes want to transmit data │ ▼ Multiple Access Protocol │ ┌───────────┼───────────┐ ▼ ▼ ▼ Random Controlled Channelization Access Access │ │ │ Compete Take turns Divide the for medium / controlled medium │ access ▼ Collisions possible
Key point
Multiple-access protocols are MAC protocols used when several nodes share a common or broadcast link. Their purpose is to coordinate access to the shared medium, prevent collisions, and handle collisions when they occur.
Comments
Post a Comment