Efficient Data Aggregation Method Based on Function Approximation and Characterization in UWSNs

In order to solve the problem of large energy consumption and extended time of data aggregation in Under-water Wireless Sensor Networks(UWSNs) target detection and positioning, we propose an efficient data aggregation method for underwater acoustic sensor network based on function approximation and characterization. Firstly, the spatial function model is established by using the spatial variation characteristics of feature-level data in UWSNs to characterize the sensor nodes in the network. Secondly, by making full use of the wireless broadcasting characteristics of underwater acoustic signals, a sequential optimal subset selection method and an optimal local error criterion are proposed, so as to realize the optimal distributed approximation of the spatial function with the least sensor feature-level data. In addition, three schemes are proposed: distributed threshold separation, probabilistic competition mechanism of node self-selection and dynamic backoff timer mechanism based on MAC layer, so as to realize the distributed fast approximation of spatial functions. Finally, the simulation results prove the excellent performance of the method, which can break through the bottleneck of energy consumption and delay of data aggregation in underwater acoustic sensor network, greatly extend the network lifetime, and reduce the aggregation delay.