查看文章

In this paper, we describe a scalable and distributed framework for minimizing congestion and assuring reliable data transmissions in event based networks. Event based networks are a particular category of sensor networks on which reports are produced only upon the observation of a specific event. This event should satisfy a pre-specified condition. Whenever this condition is satisfied, a sudden traffic increase occurs which may lead the network into congestion. This is particularly undesirable because the data generated during this situation are of great importance, often critical, to the applications. We propose a novel algorithm which is able to control a congestion situation and which is efficient enough to safely transmit almost all the data, generated by the sensors due to an event, back to the sinks. The algorithm does that without throttling the source nodes' data rate. Throttling the data rate could prove fatal for critical networks, due to the fact that each data packet provides the network with updated information concerning the monitored event.