create new tag
, view all tags
  1. A Request to Send (RTS) message is sent by initial sender.
  2. Any nodes that provide progress towards the destination begin a countdown timer.
  3. The timer is calculated using the same equation described previously.
  4. When the countdown is complete, a Clear to Send (CTS) message is broadcast.
  5. Any potential hops that overhear this CTS will cancel their own timers.
  6. The node that provides the most progress should respond with a CTS frame first, when this is received by the initial sender the data packet is sent to that particular node.
  7. After the full data packet is received and acknowledgement is sent back to the initial sender.
  8. The selected node then begins the process again.

If the initial sender does not receive a CTS in response to its first RTS message by time Tmax, it will broadcast another 5 RTS frames before discarding the data packet.

After a next hop is selected via the RTS to CTS handshake and the full data packet is sent, if an acknowledgement isnít received by time , the node will attempt to send the data packet up to a maximum of 3 times until it will retransmit a RTS frame, ignoring any CTS messages sent by the initial next hop selected.

The use of RTS and CTS control messages means that the initial sender will unicast the data packet to the first hop. As the next hop is decided by the sender there is virtually no chance of multiple forwarders being selected.

Using control messages in the algorithm means that the number of messages sent per transfer is increased. This is undesirable as the purpose of beacon less routing is to minimise the total number of messages being sent to provide a more efficient communications protocol.

Generally the RTS and CTS packets being sent are relatively small. This means that although control messages can be exchanged between the initial forwarder and a potential next hop, when the full data packet is transmitted thereís a chance that it may not be able to traverse the link. This will result in redundant messages being sent as well as a delay in the transfer as the initial sender will have to begin the process all over again, selecting a different node as the first hop.

-- XiaohongWu - 2012-01-31

Topic revision: r1 - 2012-01-31 - XiaohongWu
This site is powered by the TWiki collaboration platformCopyright © 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback