Reliable data transfer over a channel bit error rdt : 20, Computer Networking

Reliable  data  transfer over a channel bit error  rdt : 20

A more realist model  of the underlying  channel  is one  in which  bits in packet may be  computed. Such  bit errors typically  occurs  in the physical   components  of a network as a packet  is transmitted propagates or is buffered.

Before  developing a  protocols  for reliably  communication over  such a channel  receiver might  say ok after  each  sentence has bee beard understood an recorded. It the receiver hears a garbled sentence he asked to  repeat the garbled sentence, this message  dictation protocols uses both positive acknowledgment and negative acknowledgments  these control  messages allows  receiver correctly and what has  been received in error  and thus  requires repeating.

Figure  shows  the FSM representation of rdt 2.0 a data  transfer  protocols  employing  error   detection positive  acknowledgments  and negative  acknowledgments.

545_Reliable  data  transfer.PNG

                                                     Figure  rdt 2.0 FSM specification

The send  side of  rdt 2.0  has two states. In the left  state  the send  side  protocols  is waiting  for data  to be passed  down  form the  upper  layer  when  the rdt send event occurs the sender will create a packet containing  the data  to be sent along  with  a packet  checksum  and then  send  the packet  via  duct send operation. In the  right most  state the sender protocols is waiting  for an ACK or a NACK  packet  from the  receiver. If an ACK  packet  is received. The sender  knows that the most recently transmitted  packet  has been  received correctly  and thus  the protocols  returns to  the state  of waiting  for data  from the upper layer. If  a NACK is received the  protocols retransmits  that last  waits  for an ACK or NACK to be returned by the receiver in response to the  retransmitted data  packet. It is  important to note that when  the sender  is in the  wait for  ACK or NACK state it  cannot get  move data from the  upper layer that is  the rdt send event  cannot occur that will happen  only after the sender receives an  ACK  and leaves this state. Thus  the sender will  not send  a new  piece  of data until. It is sure  that the  receiver has correctly received the current  packet  because  of this  behaviour  protocols  such  as rdt 2.0  are known as stop  and  wait  protocols.

The receiver  side FSM for rdt 2.0  still  has a single  state on  packet  arrival the  receiver replies  with either an ACK  or a NACK. Depending  on whether  not  the  received  packet in corrupted. In  figure the notation rdt rev and corrupt  corresponds to the event in which  a packet  is received and is  found to be in error. Protocols rdt 2.0 many  look  as if  it works  but  unfortunately it has  a fatal flaw  or may be possible  that the  ACK or NACK  packet  could be corrupted.

Possibilities  for handling corrupted ACKs or NACKs

a.For  the first  possible it the sender didn't  understand  the ok or please repeat that reply  from the  receiver the sender  would  probably ask what did you say? The  sender  would  then repeat the reply. But  what if the  speaker what did you  say? Is corrupted  the receiver having  no idea whether difficult to understand the garbled sentence was part  of the dedication or a request to repeat he last reply, would  probably  then respond  with  did you  say?  And then  of course  that  response  might be garbled.

b.A second  alternative  is to  enough checksum bits to allow  the sender not only  to defect but also  to recover from bit errors. This solves  the immediate problem  for a  channel  that can  corrupt  packets  but not  lose them.

c.A third  approach  is for the  sender simply  resend the current  data packet when it  receives a garbled ACK or NACK  packet. This  approach  introduces duplicate packets  into  the sender to receiver channel. The fundamental problem with  duplicate packets is that the  receiver doesn't  know  whether the ACK or NACK  it last  sent was  received  correctly at eh sender. To  handling  the duplicate packets sender. Sender  adds sequence number  to each packet  retransmits  current  packet  if ACK / NACK  garbled  receiver discards  duplicate  packets.

Posted Date: 3/11/2013 7:12:29 AM | Location : United States







Related Discussions:- Reliable data transfer over a channel bit error rdt : 20, Assignment Help, Ask Question on Reliable data transfer over a channel bit error rdt : 20, Get Answer, Expert's Help, Reliable data transfer over a channel bit error rdt : 20 Discussions

Write discussion on Reliable data transfer over a channel bit error rdt : 20
Your posts are moderated
Related Questions
What is the Power Of Semantic Web Languages? Ans) The major power of Semantic Web languages is that anyone can create one, simply by publishing some RDF that defines a set of

Question 1: (a) In networking, there are different possible topologies. Explain the Linear bus, Ring and Star topology along with their advantages and disadvantages. (b) W

State the latest technology in the line of Intranet The latest technology in the line of Intranet tools has been the Web-based Distributed Authoring and Versioning (WebDAV), w

ROLE OF DESIGN IN ACHIEVING RELIABILITY One of the most important methods for achieving high reliability is through design.  Use reliability cannot be higher than the design re

BRIDGES:  A bridge is a hardware device also needed to connect two LAN code segments to extend a LAN. A bridge uses two NICs to connect two code segments. It listens to all tr

What everyday object is the same as The RAM function to the computer?

a)  What does the "CD" mean in the CSMA/CD? How is it implemented in the Ethernet?   b)  Consider an Ethernet hub with 8 ports and an Ethernet switch with 8 ports. In both cases

What is a Management Information Base (MIB)? A Management Information Base is part of each SNMP-managed device. Every SNMP agent has the MIB database that having information ab

What is a TP Monitor? There is no commonly accepted explanation for a TP monitor. According to Jeri Edwards' a TP Monitor is "an OS for transaction processing".

Packing and Unpacking Data pvm_packs - Pack the active message buffer with arrays of prescribed data type: int info = pvm_packf( const char *fmt, ... ) int info = pv