MACHINE LEARNING TECHNIQUES USED IN WIRELESS SENSOR NETWORKS

Abstract:
Wireless sensor network (WSN)
refers to a group of spatially dispersed
and dedicated sensors for monitoring
and recording the physical conditions
of the environment and organizing the
collected data at a central location.
Clustering is applied to divide the
network into clusters and each cluster
is having a cluster head in which we
perform data collection. In many
cases, while we are collecting
different types of data there is chance
of redundancy in the data while
forming a cluster and thus the size of
the data packets are increased. There
is a need to aggregate the data at the
cluster head such that data can be
compressed into less number of
packets and redundancy is also

decreased. The Clustering Techniques
used are “Affinity Propagation” and
“Spectral Clustering” and finally a
comparision under various
performance metrics like delay , CPU
usage, execution time are done
between the algorithms under the data
aggregation protocol is done and the
efficient algorithm of the two is
concluded. The most efficient
algorithm of the two is considered as
the one used to reduce redundancy in
Wireless Sensor Networks(WSN)