Channel Allocation Problem in Computer Network

In this tutorial, you will learn about the channel allocation problem. After reading this tutorial, you will understand the advantages, disadvantages, assumptions, and working of channel allocation methods.

Contents:

1. What is Channel Allocation?
2. Channel Allocation Schemes
3. Static Channel Allocation
4. Disadvantages of Static Channel Allocation
5. Dynamic Channel Allocation
6. Independent Traffic, Single Channel, and Observable Collisions
7. Continuous/Slotted Time and Carrier Sense/No Carrier Sense
8. Concept of Dynamic Channel Assumptions

What is Channel Allocation?

On a network, multiple devices are communicating with each other. It is the responsibility of the data link layer to provide reliable communication by allocating a channel to the device for communication. Allocating channels to specific devices for communication is known as channel allocation.

• The data link layer allocates a single broadcast channel between competing devices.
• Depending on the network and geographic region, the channel can be guided media or unguided media. On the channel, several nodes are connected.
• The purpose of a channel is to connect one device to another device on a network for communication.

Channel Allocation Schemes

Channel allocation problem plays a major role in the network. There are two types of channel allocation schemes used on the network. They are as follows:

1. Static Channel Allocation
2. Dynamic Channel Allocation

• Static Channel Allocation: This is a traditional way of allocating a single channel among multiple users. In static allocation, the channel’s bandwidth is split into equal-sized portions among users, and each user gets a portion of the bandwidth.
• Dynamic Channel Allocation: In dynamic channel allocation, bandwidth is not allocated to the user permanently. The frequency is allocated to the devices when it is needed on the network. It uses little CPU power, which increases the optimal resource utilization of the network.

Static Channel Allocation

In static channel allocation, the network bandwidth is divided equally between devices and is permanent for devices.

• For example, there are 50 users on the network, and the network bandwidth is 100 Hz. The 100 Hz bandwidth would be divided into 50 equally sized parts that are 2 Hz. Each user will get a 2 Hz portion.
• Each device has a private frequency band, so there is no possibility of interference with other devices.
• A well-known example of a single channel allocation scheme is FM radio, in which different frequencies are assigned to different stations that are fixed.

The figure below shows the static channel allocation scheme.

• As shown in the figure, 1000 MHz bandwidth is present on a single channel. The 1000 MHz bandwidth is divided into four equally sized frequency bands of 250 MHz, permanent for the specific station.
• In addition, there is a gap between the frequency bands of the stations to avoid signal interference with other signals.

Disadvantages of Static Channel Allocation

As we saw in the above section that the bandwidth is divided into equal-sized portions, and a portion is assigned to each device. But it becomes a disadvantage on the network. Let us see the various disadvantages of the static channel allocation scheme.

• Let’s say the bandwidth is divided for 50 devices, and only 40 devices are active on the network, then the bulk of the valuable bandwidth will be wasted.
• If there are equal-sized portions of bandwidth for 50 devices and 60 devices want to communicate, 10 devices will be denied permission due to lack of bandwidth.
• Now, let’s say 50 equally-sized bandwidth portions are assigned to 50 devices. But the problem is that when some devices are idle, their bandwidth will be lost simply because they are not using it, and no one is allowed to use it.
• In a static allocation channel, most of the channels will be idle for most of the time.

Dynamic Channel Allocation

To overcome this problem, we can use dynamic channel allocation, in which the bandwidth is not allocated to the device permanently.

• In dynamic channel allocation, the bandwidth is not allocated to the device permanently, the frequency band is allocated to the device whenever required.
• When the device completes its communication on a channel, it is de-allocated, and the same channel can be assigned to another device.
• In dynamic channel allocation, a channel is dynamically allocated between devices.

The figure below shows five assumptions for dynamic channel allocation.

As shown in the above figure, Independent Traffic, Single Channel, Observable Collisions, Continuous or Slotted Time, and Carrier Sense or No Carrier Sense are the five assumptions of the Dynamic channel allocation.

Independent Traffic, Single Channel, and Observable Collisions

The first three assumptions for dynamic channel allocation are independent traffic, single-channel, and observable collisions. Let us understand all these one by one.

Independent Traffic: In this assumption, there are n independent devices controlled by the program or user that generate frames for transmission. The expected number of frames generated in the channel depends on the length of the interval and the rate of arrival of new frames.

• After creating a frame, the device is blocked and does nothing until the frame is successfully transmitted over a channel.

Single Channel: There is a single channel between devices for communication, in which all devices can send and receive from it.

• In single-channel assumption, protocols are used to prioritize frames from less important frames to more important frames.
• The channel transmits frames from one device to another according to the priority of the frame.

Observable Collisions: The collision occurs when two frames are traveling on a shared channel and overlap with each other. When a collision occurs between frames, all devices can detect that a collision occurred on a channel. The lost frame can be retransmitted after some time. No error is produced here compared to that generated by the collision.

Continuous/Slotted Time and Carrier Sense/No Carrier Sense

The last two assumptions for dynamic channel allocation are continuous or slotted time and carrier sense or no carrier sense. Both of them play major roles in the channel. Let us understand one by one.

Continuous or Slotted Time: We can consider the time constant as when frame transmission can be initiated at any instant. Time can be slotted or divided into discrete intervals known as slots. When the slot is allocated to the device or channel, the frame transmission must start at the beginning of the slot.

• For the idle slot, the slot has 0 frames for transmission. Similarly, successful transmission occurs if the slot contains one frame, and a channel collision will occur if the slot contains more frames.

Carrier Sense or No Carrier Sense: In the carrier sense, the device checks whether the channel is in use before using the channel. If the station finds the channel busy, it will not broadcast any frame on the channel. If there is no carrier sense, the station may not understand the channel before use, leading to frame collisions.

Concept of Dynamic Channel Assumptions

Now, we have understood the five assumptions of the Dynamic Channel Allocation Scheme. Some assumptions can either be good or bad, and they also affect the performance of the network. Let’s discuss the assumptions.

• We have seen that in independent traffic assumption, frames are generated according to the length of the interval and the arrival rate. In independent traffic, frame arrivals occur independently and are generated unexpectedly, which is not good for network traffic because the packets sent by the network layer are massive.
• The single-channel assumption is the heart of the dynamic channel model as no external way to communicate exists. Since it uses protocols to prioritize the packets, performance of the network increases.
• The collision assumption provides a way to detect collisions if frames collide and frames need to be retransmitted.
• Slotted time can be used for better performance. Splitting the time into discrete intervals requires the devices to be synchronized with each other.
• A network may or may not have carrier sensing. In wireless networks, not every device will be within radio range of another device, they may not use carrier sense effectively.
• Even if there is no network conflict after sensing the channel, the receiver may receive some frames incorrectly for different reasons. The data link layer protocol and other higher layers provide reliability.

The diagram below explains the concept of dynamic channel allocation assumptions.

• As shown in the figure, all the stations are connected on a single channel and sense the channel continuously.
• Stations sense the channel to see if the channel is busy or idle. If a station finds the channel idle, it sends data to the channel. If a collision has occurred with data from another station, the collision detection method will notify the station that the collision occurred on the channel.
• When a station sends data to another station, a clock is synchronized, which helps the sender and receiver identify the data.