selective repeat || ACK falling outside of sender's window

Question

.With the SR protocol, it is possible for the sender to receive an ACK for a packet that falls outside of its current window.??

pleasee explain with example

Answer 1

obviously its possible...u sent a packet ,didnt recieved its ack...due to some delays say...then u retransmit it and everything goes on happily...but,
average lifetime of a packet is about 3mins...ur ack can arrive any where in between and its nt neccessary that u have to wait for its life time(genrally we wait)...if u get 3 duplicate acks then we can do fast retransmit also...

conclusion: its possible to recieve...but u'll nt accept that ofcourse because u must have retransmitted it already or have setup a new conenction!!

its like u are living in a house and a letter has to arrive for u...before it arrives,u changed ur location..then also letter will arrive though u dont accept it or not present..if letter has to arrive than it will unless its nt lost!!

Comments

I guess the answer holds from a TCP perspective. In a more general case, SR protocol slides it's window only when an ACK for a packet is received. Depending on the way duplicate packets are handled at the receiver side the answer can be YES.

Whereas in case of GBN, this definitely holds true due to the usage of cumulative ACKs.

Please refer: http://web.eecs.utk.edu/~qi/teaching/ece453f06/hw/hw7_sol.htm

---

@krish there is no proper reasoning given there why it cant happen
http://www.ics.uci.edu/~keldefra/teaching/spring2013/uci_cs132/problemsets/CS132_EECS148_ProblemSet3_Solution.pdf
refer to this also they have shown a case similar to how akriti has mentioned below..it'll help!!

---

Yeah. So the duplicate packets have to be acknowledged by the receiver or else the sender won't slide it's window. That's a good point to note. :)

---

Delayed acknowledgement possible in SR protocol ??

Answer 2

Acknowledgement from the receiver in $SR$ protocol is selective. It means the receiver acknowledges the successful arrival of packet by giving acknowledgement of successful arrived sequence. It means sender has already sent the packet. Therefore I think it isn't possible.

Comments

Can you post their reason or example?

---

here it is -

Suppose the sender has a window size of 3 and sends packets 1, 2, 3 at t0 . At t1 (t1 > t0) the receiver ACKS 1, 2, 3. At t2 (t2 > t1) the sender times out and resends 1, 2, 3. At t3 the receiver receives the duplicates and re-acknowledges 1, 2, 3. At t4 the sender receives the ACKs that the receiver sent at t1 and advances its window to 4, 5, 6. At t5 the sender receives the ACKs 1, 2, 3 the receiver sent at t2 . These ACKs are outside its window.

---

But still we can get. I thought only about the forward packets.