print page

Diploma Thesis

Energy-Efficient TDMA Schedules for Data-Gathering in Wireless Sensor Networks

In recent years, wireless sensor networks have been frequently adopted for data-gathering. The inherent request for prolonging network lifetime gives rise to the demand for energy-efficiency. In multi-hop networks, this end is met by applying dedicated data-collection phases in combination with Time-Division-Multiple Access (TDMA) protocols for scheduled transmission. Despite the existence of a variety of different scheduling schemes, a detailed comparison has not been carried out. In this thesis, an analytical investigation is therefore provided to reveal the strengths and weaknesses of existing schemes. As the latter exhibit severe disadvantages, a new scheduling scheme, named Spatial Path-Based Reuse (SPR), is devised. Its distributed implementation is frugal and highly efficient, as it combines a small memory footprint with low communication overhead. To permit an in-depth comparison under realistic conditions, a simulation framework for the ns-2 simulator is also developed. The results obtained from extensive simulation with more than 400,000 individual runs substantiate the advantages of SPR, particularly in large networks. Moreover, they reinforce the strengths and shortcomings of the different schemes. The thesis concludes with recommendations for the most suitable scheme for a given data-gathering scenario.

Start date 15. December 2007
End date 15. June 2008
Documents Thesis | Presentation
Projects SomSeD | SensorNet
Supervisor Christoph Weyer