Comparison of Cell-Based and Topology Control-Based Energy Conservation in Wireless Sensor Networks

In this paper, we compare the effectiveness of two popular energyconservation strategies, namely topology control and cooperativecell-based approaches, to extend the lifetime of ad hoc networks. To this end, we define a realistic, though necessarily idealized in some respects, and unified framework of investigation. Using this framework, we perform a number of simulation experiments taking into account different dimensions of the analysis. In particular, we investigate realistic traffic models and energy parameters that fit the potential application settings of general ad hoc networks and sensor networks. We also consider alternative definitions of network lifetime and, in case of general ad hoc networks, stationarity and mobility. Despitethe growing number of papers on energy conservation, this is the firstattempt at a comprehensive understanding of the conditions under whichone approach outperforms the other. Indeed, our study reveals a number of properties of the techniques investigated, some of which are not at all obvious. We have found that cell-based cooperative approaches, which completely power down network interfaces of certain nodes for extended time periods, produce significantly longer network lifetimes when node density is very high and in the presence of mobility. While it is not surprising that cell-based approaches do not extend lifetimeunder low density conditions, we find that, for certain parameter choices, topology control techniques {\em can} significantly increase network lifetime under those conditions and, in fact, they substantially outperform cooperative approaches in this respect.