A common assumption in sensor networks is that the sensors are located according to a uniform random distribution. In this paper we show that uniform random points on the two dimensional unit square are almost a “grid”. In particular, for a synchronous geographic sensor network we show how to emulate any grid protocol on random sensor networks, with high probability. This suggests the following framework. In order to solve a problem on a random sensor network we solve the same problem on the grid. Then we use our emulation to make the obtained solution suitable for random sensor network. We analyze the cost of the emulation in terms of consumed energy and time. Finally we provide three examples that illustrate our method.
Managing Random Sensor Networks by means of Grid Emulation
NAVARRA, Alfredo
2006
Abstract
A common assumption in sensor networks is that the sensors are located according to a uniform random distribution. In this paper we show that uniform random points on the two dimensional unit square are almost a “grid”. In particular, for a synchronous geographic sensor network we show how to emulate any grid protocol on random sensor networks, with high probability. This suggests the following framework. In order to solve a problem on a random sensor network we solve the same problem on the grid. Then we use our emulation to make the obtained solution suitable for random sensor network. We analyze the cost of the emulation in terms of consumed energy and time. Finally we provide three examples that illustrate our method.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
