lec9 - ECE 333: Introduction to Communication Networks Fall...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
1 ECE 333: Introduction to Communication Networks Fall 2002 Lecture 9: Data Link Layer V Sliding window ARQ protocols § Go-back-N § Selective repeat 2 Previously we presented a basic ARQ protocol, the stop-and-wait protocol. Recall, in a stop-and-wait protocol, the sender waits until a packet is correctly acknowledged before sending the next packet. When the round trip time is large relative to the packet transmission time, this protocol has low efficiency. In this lecture, we begin discussing more efficient approaches called sliding window ARQ protocols . In these protocols the sender is allowed to transmit more that one packet before receiving an acknowledgment. The total number of unacknowledged packets that may be sent is referred to as the sender's maximum window size . For example, suppose that the sender has a window of size W . In this case, the transmitter can send packets numbered 1,2,…, W , before receiving an acknowledgment for packet 1. Before discussing specific sliding window protocols, we consider how large the maximum window size should be.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
3 Filling the Pipe To maximize efficiency, the window should be chosen big enough to minimize the time the transmitter is idle. Recall in Lecture 8, we considered an example of a satellite link with a RTT of 500 msec and a packet transmission time of 20 seconds. Thus in a RTT, the transmitter can send 25 additional packets. If no errors occur, the acknowledgment from the first frame should arrive immediately after the 26th frame is sent. This allows the transmitter to send frame 27. In this case a window size of 26 is sometimes said to fill the pipe . 26 Ack 1 25 Ack 2 Ack 13 14 With no errors, this gives good efficiency, because it allows the sender to transmit continuously. What about when errors occur? 4 Error Recovery in sliding window protocols When a sliding window protocol is used, we still want to have a retransmission protocol that provides a reliable service, i.e. delivers packets in order, correctly, and only once. There are two classes of such protocols that are commonly used, the first is called go-back-n, and the second is called selective repeat . For both types of protocols, as with stop-and-wait, the sender numbers each packet, and the receiver returns numbered acknowledgments when a packet is correctly received. However, notice that with a sliding window protocol, as opposed to stop-and-wait, more than 1 bit will be needed for sequence numbers. Recall the sender's maximum window size is the maximum number of unacknowledged frames it is allowed to have sent at any time. We will also refer to the sender window as the set of frames that have been sent but are not yet acknowledged . Therefore, the number of frames in the sender window must be no greater than the maximum window size.
Background image of page 2
5 Go Back N In a Go Back N protocol, the transmitter has a maximum window size of N packets. The receiver behaves essentially the same as in stop-and-wait. That is it keeps a counter of the current frame it is waiting to accept, and will
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 10/12/2009 for the course ECE ECE 333 taught by Professor Randallberry during the Fall '02 term at North-West Uni..

Page1 / 9

lec9 - ECE 333: Introduction to Communication Networks Fall...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online