Mobility Aspects in WSN - Application and Multidisciplinary Aspects of Wireless Sensor Networks - page 128

Table 6.1 (conti nued)

Goals and
performances Algorithm Advantages Disadvantages

ZONER Coverage improvement

with energy efficient
approach for sensor
relocation

Guaranties replacement
discovery and closest
node replacing, void
area capability

Local protocol, sometimes
increased message
overhead

MSRP Coverage improvement

with improved
mitigation path
discovery

Low message complexity
and constant storage
load

Local protocol, sometimes
can cause increased
message overhead

SMART Even sensor deployment Distributed protocol,

improved
convergence rate

Works for dense networks,
message overheads

SNAP &
SPREAD

Uniform and regular
sensor deployment,
coverage uniformity

Local knowledge, avoids
useless movements,
energy effective

Does not guarantee hole
elimination, possible
network partitioning

MaxNetLife Increasing cumulative

connected coverage

Preemptive relocation of
sensor nodes, scalable
and applicable

Localized protocol, long
termination time

GA & GRG Connectivity and

coverage

Finite execution time Obstacles are not taken into

consideration

Distributed, minimal
communication and
computation

Requires knowledge of
nodes positions

EBD Coverage maintenance,

convergence towards
event distribution

SEC Bounding event loss

probability

Good analytical
algorithm design

Only analytical analyzes,
PoI are known a priori

EDSSA Jointly achieves

network
deployment and
relocation

No centralized
control, no
communication
between distant
nodes

Similar constraints as virtual
force based algorithms

disadvantages of the approaches, which were described in the algorithms' defini-
tions. The table gives a list of several algorithms, with their major characteristics
and specifics.

6.3.2 Sensing Enhancement

Mobile sensors have the advantage to move to locations that meet sensing coverage
requirements. Enhancing sensing in WSN can be achieved with repositioning sen-
sors to mitigate sensor failure and moving toward events that occur in the network
and that need to be captured in real time. Strategies about the specific problems are
further discussed, while their comparisons are added on the previous Table 6.1 .
An event-based deployment (EBD) framework is introduced in [ 31 ], where
sensors move to locations where events are detected, while maintaining a desirable
coverage for the entire field. The presented approach consists of two distributed