To the best of our knowledge, SEM is the most recent and efficient explicit multicast
protocol proposed. It adopted the idea of keeping the state information only in the branching
node routers instead of keeping this information in all group members as in traditional
multicast protocols. However, it still has a number of limitations that discussed before.
To overcome these limitations found in SEM, this thesis proposes new protocol called
SReM, which has the same objectives as SEM with major enhancements. The basic idea
behind SReM is to forward data between dynamically selected nodes called Branching Node
Routers (BNRs). Its goal is to reduce the state information kept for the multicast group
members and to reduce the routing overhead by providing a local join/leave and tree
maintenance procedures using fixed size messages. Hence, it is achieving higher degree of
scalability. The main difference between SReM and SEM is that in SReM there is no
reconstruction of tree when group member leaves as in SEM.
The second part of the discussion is concerned with the explicit multicast protocols for
MANETs. DDM, E2M and EM2NET are the most recent and efficient protocols proposed in
this area of research. DDM is a source based explicit multicast protocol, which keeps the
receivers addresses as list in the source and the list is kept up to date by periodical Join
messages. These messages will increment or decrement the contents of the multicast list in
the source node. In this protocol if the mobility is high then the changes in the list will be too
much which will cause a huge overhead. Other problem may be faced by this protocol is the
huge overhead in the source because the source is fully responsible for joining and leaving of
members and forwarding data packets.
E2M employs Xcast forwarders (XFs) to take part of the job taken in the previous protocol
by the source. The Xcast forwarder(s) keep information about a set of receivers and at the
same time this XF is known to the source. This idea will reduce the number of control
packets arrived to the source. This protocol did not give an efficient solution for mobility
issue especially when the XF moves. The way proposed to solve the movement of XF is to
give the source the responsibility to take XF job by again encoding the receivers that was
53