create new tag
, view all tags
  1. An initial packet is sent. This contains the full data message, the position of the initial sender and the position of the destination.
  2. Any node that provides progress towards the destination stores the message.
  3. A timer is then started, with the time being calculated as with the previous protocols.
  4. A response message containing the nodes position is broadcasted.
  5. Any neighbours that overhear the broadcast cancel their own timer and delete the stored packet.
  6. When the initial sender receives the response, it broadcasts a selection frame notifying neighbouring nodes of which will be selected to forward the message.
  7. Any nodes that overhear the selection message and haven’t already cancelled their timers will do so.
  8. The selected node then forwards an initial packet as mentioned in step 1, beginning the process again.
During the forwarding process, when the selected node forwards an initial packet the initial sender will over hear this and interpret it as an acknowledgement. As with BLR, this reduces the number of control messages required. The only drawback is when the packet reaches the destination. As the destination doesn’t forward the message, the node that passed it on will assume the packet wasn’t received. Therefore an acknowledgement needs to be sent when the message arrives at the destination.


This protocol aims to combine the advantages of the previously mentioned protocols. By including the full data message in the first packet being sent, this will prevent the initial sender from choosing a next hop that may have an unstable connection. This results in less retransmissions and a lower end to end delay.

As with the IGF protocol, this protocol relies on control messages sent between nodes. This ensures that the forwarding of the packet is done by a single node and prevents multiple packets arriving at the destination.


In theory this protocol should outperform both of the previously mentioned protocols; though by combining their strengths it will also inherit minor comparative weaknesses.

Although it will produce fewer duplicates than the BLR protocol, it requires more control messages to do so. And although it will utilise more stable links than the IGF protocol, it must also send larger messages when transferring data from source to destination.

Algorithm Problem IGF BLR BOSS
Unreachability High Low Low
Unreliability Med Med Low
Duplicates Low High Low
Contention Med Med Low

-- XiaohongWu - 2012-01-31

Topic revision: r1 - 2012-02-01 - 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