Self-Organization of Wireless Sensor Networks for Wildfire Detection Using Shapley Values

Wildfires have devastating impacts on the environment, economy and public health. Early detection can help mitigate damage but detection can be challenging in remote and inaccessible areas. Wireless sensor networks provide an effective solution. These networks consist of inexpensive nodes that monitor the environment and transmit the sensed data over wireless links to a base state. Sensor nodes are usually randomly distributed via air and must self-organize to create the network. This paper proposes a new self-organizing method that forms a hierarchical sensor network suitable for wildfire detection. Clusters are formed without a predefined structure or the need for localization. Shapley values are used to select cluster heads as new clusters are dynamically created during self-organization. Routing paths for aggregated data are naturally defined as clusters are created. Simulation results are presented to show the effectiveness of this approach.