Abstract:
Opportunistic network is network of wirelessly connected nodes in specific area and the
topology of the network is changed due to the mobility of the nodes. Opportunistic networks use
store-and-carry forward mechanism to deliver message from source to destination. This network
is extension of wireless multi-hop mobile ad-hoc networks and a part of delay tolerant networks
(DTN). Routing in opportunistic network usually utilizes multiple message copies in order to
guarantee the data delivery. But, data delivery is affected by limited storage in the network due
to storage congestion problem. To overcome this problem proper buffer management should be
used on the selected routing protocol.
Authors in this area proposed different message dropping policies for buffer managment to drop
the right message from the buffer when buffer is full in order to accept the incoming message.
But the problem is which message should be dropped frist without affecting network
performance is an important task. This research work proposed social based message dropping
policy for SimBET routing protocol by combining message remaining time to live with node
SimBET utility. Therefore Social based dropping policy drops a message having least Remaining
time to live and node SimBET utility because they have lowest delivery probability.
The study uses the ONE network simulator tool to conduct experiment in order to evaluate the
performance of proposed dropping policy with existing dropping policy. the simulation result
shows the proposed SMDP dropping policy performs better interms of message delivery in the
network by varying message time to live and buffer size of the nodes then the result report shows
the Average delivery of message by varying TTL is 73.93% in number 5401 messages are
delivered for SMDP and 73.79% or 5391 messages are delivered for TTL based dropping policy
and by varying buffer size 75.85% in number 5541 messages are delivered for Social based
Message Dropping Policy and 75.52% in number 5517 messages are delivered for TTL based
dropping policy for simulation time 43200s or 12hr. Thus the result shows the proposed SMDP
performs better than the existing dropping policy interms of Delivery ratio, Overhead ratio and
Average hop Count using two scenarios TTL varying and Buffer size Varying
