Modeling for Geo-Routing in Multi-Hop Wireless Networks - Application and Multidisciplinary Aspects of Wireless Sensor Networks - page 148

on the basis of incoming signal strengths. Relative coordinates of neighboring
nodes can be obtained by exchanging such information between neighbors [ 13 ].
The other assumption, which is made for this project, is that the source node is
aware of geographic position of destination. Such algorithms, where the source
node is aware of the geographical position of its destination are known as “ Position
or Destination ” based routing protocols.
In comparing the two models, the UDG model and the correlated shadow fading
model, “ Localized Destination Based” algorithms have been considered. The reason
for comparing the UDG model with the proposed model is that the designers of
network layer protocols for ad hoc and sensor networks assume the Unit Disk
Graph (UDG) communication model while designing the protocols.

7.6.1 Adopted Unit Disk Graph (UDG) Model

In the UDG model of § 7.1, we choose all nodes to transmit with equal transmission
power. Thus all nodes are taken to have a fixed and equal transmission radius R .
Furthermore, for comparison purposes with the remaining models, we may define an
equivalent UDG transmission range, R , by either of the two methods given below:

• For adopted propagation path-loss model, we set =σ to recover the fully

deterministic, circularly symmetric model for the probability of packet reception
shown in Fig. 7.1 . We then define the equivalent UDG radius to be the radial
distance, R , from the transmitter where ppr = 0.5.
The transmission radius “

0dB

dB

• R ” can alternatively be chosen on the basis of the node

receiver sensitivity, which in the case of IRIS mote is (according to its data sheet
[ 16 ]) ( 131dBW ). This transmission radius is calculated only on the - 101 dBm -

basis of mean path loss (i.e., =σ ) and the received signal strength, without

resorting to computation of the probability of packet reception.

0 dB

dB

The latter method was used in the following simulations, but it is worth pointing
out that the difference between the value of the equivalent UDG transmission
radius, R , given by the above two methods was very small, 63 and 61 m, respec-
tively. Since all nodes are assumed to have the same transmit power and antenna
gain and height, R is equal for all nodes.

7.6.2 Localized Greedy Algorithm

The localized greedy position-based algorithm as proposed by Finn [ 14 ] suggests
that the node that is currently holding the packet will forward it to the neighbor that
is closest to the destination. This algorithm makes unambiguous and explicit use of
the notion of a transmission radius as defined in the idealized UDG model; however,
routing fails if no neighbor is closer to the destination than the current node.