Energy and Routing Optimization in Wireless Sensor Networks

swarmtellingΚινητά – Ασύρματες Τεχνολογίες

21 Νοε 2013 (πριν από 3 χρόνια και 4 μήνες)

70 εμφανίσεις

LABORATOIRE DE RECHERCHE EN
I NFORMAT I QUE
Energy and Routing Optimization in Wireless Sensor Networks
Joseph Rahmé Advisors:Aline C.Viana and Khaldoun Al Agha
hipercom.lri.fr
Introduction
Energy is a scarce resource in sensor networks.It is thus important to opti-
mize energy consumption in order to extend the life of the sensor network.
We introduce a model that quantifies per-state energy consumption (Recep-
tion,Sleep,Transmission and Overhearing),and approximates the remaining
energy of a node.
Our cost functions use the information about the remaining energy to calcu-
late energy-efficient routes.
The Energy Model
We use our model,represented by the following formulas,to quantify the
energy consumed for each node state:
(k) = I
k
[
k
+
2

2
(

2
6
ln(
e

2

k
e

2

k
1
) +
1
2e
2
2

k
+
1
3e
3
2

k
)]

r
(k) = I
k
[
k
+
2

2
(ln(
e

2
(L
k
)
e

2

k
e

2
(L
k
)
1
)
1
2
(
1
e
2
2
L

1
e

2
(L
k
)
)
1
3
(
1
e
3
2
L

1
e
3
2
(L
k
)
)]
Where
(k);
r
(k) are the energy consumed per state with and without recovery,re-
spectively.I
k
represents the current value during state duration 
k
L is the total duration of a node state including idle time, is the rate at
which the active charges are replenished at the battery’s electrode surface
Our Cost Functions
Our approach consists of four cost functions,and takes the following criteria
into account:
First Cost Function
Goal:minimize the end-to-end energy needed to route a packet between a
source and a destination
E

1
(k;i) = E
TX
+
X
n
1
2N
1
(k)
E
RX
+
X
n
2
2N
2
(k)
E
I
(1)
,where
• N
1
(k) is the set of 1-hop neighbors of node k;
• N
2
(k) is the set of 2-hop neighbors of node k;
• E
TX
,E
RX
and E
I
are the energy consumed for a transmission,recep-
tion and overhearing respectively.
Second Cost Function
Goal:Find a route in which intermediate nodes,between a source and a
destination,possess high remaining energy:
E

2
(k;i) = minf(E
r
(k) E
TX
);
min
n
1
2N
1
(k)
(E
r
(n
1
) E
RX
)
min
n
2
2N
2
(k)
(E
r
(n
2
) E
I
)g
(2)
,where
• E
r
(k) is the remaining energy of node k;
• E
r
(n
1
),E
r
(n
2
) are the remaining energy of the 1-hop and 2-hop neigh-
bors affected by the transmission of node k
Third Cost Function
Goal:Find a path with intermediate nodes that have high remaining energy
E

3
(k;i) = min f
E
r
(k)
E
TX
;min
n
1
2N
1
(k)
E
r
(n
1
)
E
RX
;min
n
2
2N
2
(k)
E
r
(n
2
)
E
I
g
(3)
Fourth Cost Function
Goal:Optimize the first cost function by considering the remaining energy of
the neighbors
E

4
(k;i) = c
f

E

1
(k;i)
E
r
(i)
;8i 2 N
1
(k) (4)
Results
Illustration of the First and Fourth Cost Functions
Conclusion and Future Work
Simulation results demonstrate that the fourth cost function signifi-
cantly increases the network’s lifetime.This is expected since it con-
siders both the remaining energy and the end-to-end energy consumed.
We next plan to implement the fourth cost functions as a routing metric,
using the OLSR routing protocol,in order to make the routing protocol
energy efficient.
1