Abstract:
Wireless sensor network (WSN) technology is one of the popular applied tech-
nologies due to its various applications. Data collection or data gathering is a
vital part of WSN. To make the network workable and to maintain the quality
of services, energy management is also essential. To collect enough data, it is
desirable to keep as many as sensors active which causes network energy failure.
Trade-o between data collection and energy saving has made data collection
scheme in WSN enough challenging. This research addresses the problem of en-
ergy limitations in data gathering. To handle the challenges of data gathering,
we present an energy e cient data gathering approach ZDG which increases the
network lifetime and packet delivery by partitioning the network into some zones.
The tasks of data collection and aggregation are distributed among the nodes of
the zones which resist the nodes from going to the bottleneck state by minimizing
the data aggregation latency. Since the bu er capacity of a sensor node is limited,
it is necessary to reduce data aggregation overhead also. To reduce the overload,
we design a load balanced data collection scheme by dividing zone into clusters.
We select some cluster heads in the zone to handle e ciently the bu er limitation.
The clusters are on-demand clusters which helps to reduce energy consumption.
In this thesis, we provide the theoretical analysis of energy consumption by our
method. We simulate our technique with a popular sensor network simulator,
OMNeT++. We compare our results with some popular zone-based data collec-
tion schemes for WSN. Simulation results show that our method improves the
network lifetime and our method provides the higher throughput or packet de-
livery compared to those methods. We nd that about thirty percent energy
consumption is less in our method compared to the other popular methods. Our
method has better packet delivery compared to those methods also.
