UNIT II

MAC AND ROUTING PROTOCOLS

MAC protocols – fundamentals, low duty cycle protocols and wakeup
concepts, contention and Schedule-based protocols - SMAC,
BMAC,TRAMA, Routing protocols – Requirements, Classification -SPIN,
Directed Diffusion, COUGAR, ACQUIRE, LEACH, PEGASIS.

Classification of Routing Protocols:

The following routing methods are discussed in this chapter.

i. Energy Efficient Unicast Routing

ii. Geographic Routing

Energy Efficient Unicast Routing

Based on the use of routing tables, energy efficient unicast routing is of two types.

i. Energy Efficient Unicast Routing without Routing Table

ii. Energy Efficient Unicast Routing with Routing Table

Energy Efficient Unicast Routing without Routing Table

This uses forwarding schemes to send the packet. Flooding is mainly used which forwards
the incoming packets to its neighbor. Topology-control forwarding scheme is also used and
this have improved efficiency than flooding. Another scheme is rumor mongering, in which
an update is periodically and randomly updated to another node until sufficient number of
nodes have updated the same information. In wireless applications a single transmission can
be received by all the neighbor nodes in the radio range. This property is called wireless
multicasting. This option is dependent on the deployed MAC protocol and the relative cost
for sending and receiving the packets.

3.6.1.1.1 Randomized Forwarding

This uses gossiping mechanism and with certain probability a node retransmits the incoming
packets. Normally the probability is constant. There is a critical threshold or critical
probabiliny pake beat the probability is donand the incoming packets reaches only few
numbers of nodes. If a node uses probability higher than critical probability value, then
gossiping reaches almost all the nodes in the network. The value of critical probability is
about 65 to 75 percem. The nodes present near the boundary region are critical because
these nodes have less number of neighbors than the nodes in the center region. The ratio of
nodes that receive a gossip is increased by the following ways.

i. Neighbor of a node near boundary region should have few neighbors with high
retransmitting probability.

ii. Prevent fast gossiping by retransmitting packets with probability one over first few hops, iii.
Retransmit a packet if the node does not overhear the packet from atleast one neighbor.

In broadcasting, a node decides when to retransmit a received or overheard packet based

on looking at the following.
i. Counters - No retransmission when packet is overheard few numbers of times.

ii. Distance No retransmission if the distance from sender is small.

iii. Location Additional coverage is obtained by retransmission.

Random Walks

This method considers data packet as an agent that wanders throughout the network to
reach its destination. The packet from a source is forwarded to its neighbor or next hop
randomly. The neighbor node will again forward to its neighbor and so on. Random walk
method is inefficient for WSNs. The following are examples of extensions to random walks.

i. Rumor Routing

ii. Random Walks with known Destination

Rumor Routing This method is based on event notification. An event may occur when the
temperature exceeds a predefined value. A sensor node is capable of observing the
occurrence of an event. Previous methods flood either a query for the event or notify the
occurrence of an event to the entire network. Rumor routing will not flood the occurrence of
event information to the entire network, but installs the information about the occurrence of
an event to few paths by using agents The agents propagate from node to node along few
paths and install the routing information to the visited node about the event. Figure 3.10.

Figure 3.10

Setting Two Event Paths from an Event Source (Middle Node)

If a node wants to get information about the event, it sends a query in the form of one or
more agents. The agents are forwarded until it intersects a node in the preinstalled event
path as shown in figure 3.11. The principle behind this is that two random lines in a square
intersect each other with a probability of about 69 percent. But in reality the search path or
the event path may not be ir. straight line. So by using five event paths the probability
can be increased

SPIN:
One of the first data-centric dissemination protocols for wireless networks is Sensor Protocol
for Information via Negotiation (SPIN) (Figure 1). The target scenario is a network where
one, several, or possibly all nodes have data that should be disseminated to the entire
network. Moreover, the data per node is relatively large such that a unique name for each
piece of data that a node holds can be easily created and is of small size relative to the data
itself. When applying, for example, simple flooding to such a scenario, the network will suffer
from implosion and from overlap – the same area is observed by different nodes, each
independently and needlessly reporting that data. Moreover, simple flooding is unaware of
resource limitations in different nodes. To overcome these problems, the names of the data
to negotiate which nodes should forward which data is used.
This negotiation replaces the simple sending of data in a flooding protocol by a three-step
process. First, a node that has obtained new data – either by local measurements or from
some other node – advertises the name of this data to its neighbors. The receiver of an
advertisement can compares it with its local knowledge and, if the advertised data is as yet
unknown, the receiver can request the actual data. If the advertisements describe already
known data (for example, because it has been received via another path or another node
has already reported data about the same area), the advertisement is simply ignored. Only
once a request for data is received, the actual data is transmitted.
The savings in this approach rest on the small size of the data description compared to the
data itself. Once description of the data becomes comparable to data itself, it is not useful to
first announce the data instead of simply sending it. The advantage is a relatively simple rule
how to constrain, on the basis of the actual data, the flooding of data within the network. It is
actually more powerful than the broadcasting constructions since it can take into account
which data is actually missing and not only which node has not yet reported (possibly
redundant) data. This basic protocol idea is made more concrete by developing protocols for
point-to-point networks and networks that enjoy a wireless multicast advantage (depending
on the MAC, a WSN can be regarded as either of these two cases). Moreover, variants of
the protocols that adapt node behavior to the remaining battery capacity are presented;
nodes with low reserves reduce their participation in the protocol. While details differ, these
protocols are claimed to be able to transmit 60 and 80% more data for a given amount of
energy than conventional protocols.