Static Channel Allocation in LANs and MANs :
The most typical way of allocating a single channel among multiple competing users is Frequency Division Multiplexing (FDM). If the number of users are N, the bandwidth is divided into N equal-sized portions. Each user is assigned one portion. If the number of users are small and constant, FDM is a simple and efficient allocation mechanism.
A telephone trunk can be a simple example of this type. However, when the number of senders is not small and constant or the traffic is heavy, FDM presents some problems. If the spectrum is divided into N regions the number of users currently interested in communicating is less than N, a large piece of valuable spectrum will be wasted.
And due to this problem If more than N users want to communicate, some of them will be denied permission for lack of bandwidth.The main reason for lack of bandwidth is that some of the users who have been assigned a frequency band hardly ever transmit or receive anything. So, dividing a single channel into static sub channels is quiet inefficient.The poor performance of static FDM can easily be seen from a simple queuing theory calculation. If the mean time delay is T,for a channel of capacity C bps, with an arrival rate of lambda frames/sec, each frame having a length drawn from an exponential probability density function with mean 1/1/µ bits/frame. With these parameters the arrival rate is lambda frames/sec and the service rate is µC frames/sec.
From queuing theory , T = 1/(µC-lambda) For example, if C is 100 Mbps, the mean frame length, 1/µ, is 10,000 bits, and the frame arrival rate, lambda, is 5000 frames/sec, then T = 200 µsec.
Dynamic Channel Allocation :
Unlike the static channel,dynamic channel allocation is efficient and is used in areas where the traffic is nonuniform and heavy. Five main assumptions must be considered in dynamic channel allocation.
1. Station Model:
Stations are also called terminals. The number of independent stations are N, with independent constant arrival rates lambda, and probability of a frame being generated in a time interval of (delta t) is (delta t x lambda). Once a frame has been generated the station does nothing until the frame has successfully been transmitted.
2. Single Channel Assumption:
From hardware point of view, all stations are equal. A single channel is available for communication on which all stations can transmit on it and all can receive from it.
3. Collision Assumption:
If two frames are transmitted simultaneously, they will collide resulting in a false signal. Each station has the ability to detect collision and it must be kept in mind that collided frame must be retransmitted later.
4. Time Management:
Frame transmission can begin at any instant as there is no master clock diving time into discrete intervals.
Frame transmission start at the beginning of the time slots. A slot may contain 0,1, or more frame corresponding to an idle, successful or collision transmission respectively.
5. Sensing of Channel:
A channel can be sensed by station before trying to use it. If a station senses the channel as busy, no station will attempt to use it, until it goes idle.
No Carrier Sense:
Stations cannot sense the channel before trying to use it. First they transmit and then they came to know where the channel is busy or idle.
|← Data Link Layer - Sublayers, Services & Framing||Medium Access Control Sublayer MAC →|