Adaptive random access for mobile sensors based field reconstruction

We consider an adaptive scheme to determine the access probability for random access when the number of sensors is varying in a wireless sensor network (WSN) to reconstruct a target field (i.e., a 2-dimensional process). In particular, we consider mobile sensor units (MSUs) that are sensor units on cars or unmanned aerial vehicles (UAVs), The access probability is adaptively decided within a session that consists of multiple time slots by estimating the number of MSUs. An access point (AP) is to estimate the number of the MSUs that transmit preambles and fed back this information so that the MSUs can determine their access probability to maximize the throughput. Simulation results confirm that the proposed adaptive scheme can provide a near optimal throughput although the initial estimate of the total number of MSUs is overestimated.