Efficient multi-hop broadcasting in dense nanonetworks

Broadcasting is a widely used dissemination technique in wireless multi-hop networks. The simplest broadcasting technique, the pure flooding, is quite inefficient in terms of number of packets generated. Whereas numerous ways to optimize it in classical networks have been proposed, we found no proposition appropriate for very dense networks (nodes having thousands of neighbors or more) with resource-constrained nodes. Typically, nodes have to prevent the broadcast storm problem using only a partial or even no information about their neighborhood due to memory scarcity. At the same time, they have to avoid the die out problem commonly found in broadcasting techniques. In this article we propose an efficient broadcasting scheme for dense nanonetworks. It combines an estimator of the number of neighbors, a backoff window and a counter of packets. We explain why this method is efficient in dense networks. Based on simulation results, we show that on a random dense network it reduces by 71% the number of packets exchanged in the network compared to an optimized method found in the literature, while at the same time avoiding the die out problem.