Robust Probabilistic Information Dissemination in Energy Harvesting Wireless Sensor Networks

Modern network environments, like energy harvesting wireless sensor networks in which network lifetime can be prolonged due to ambient energy collection, necessitate the revisit of classical networking problems like information dissemination. However, as observed in this paper, flooding-based information dissemination mechanisms suffer certain limitations due to the idiosyncrasies of nodes' operational states in energy harvesting network environments. Certain observations motivate the introduction of Robust Probabilistic Flooding, i.e., a robust version of Probabilistic Flooding, capable of dealing with non-operating nodes due to exhausted batteries that later resume their operation due to successful ambient energy collection. A Markov chain model is also introduced to capture the qualitative aspects of such environments in which nodes may or may not operate. This Markov chain is simplified in the sequel, based on certain observations and assumptions presented here, and subsequently used for evaluating the proposed Robust Probabilistic Flooding through simulations. In particular, simulation results demonstrate the inefficiency of Probabilistic Flooding to achieve full coverage in energy harvesting environments. On the other hand, it is shown that Robust Probabilistic Flooding is capable of fully covering the network on the expense of increased termination time. Furthermore, no extra overhead is introduced with respect to the number of messages, thus avoiding extra transmissions and therefore, no additional energy is consumed.