摘要

This paper studies an unmanned aerial vehicles (UAV) based data aggregation of wireless sensor networks (WSNs). Data aggregation approaches based on Compressive Sensing (CS) enjoy the benefits of having their traffic volume reduced. However, existing CS-based data aggregation schemes have drawbacks, such as errors in data reconstruction process and the large overheads in regenerating and broadcasting feasible measurement matrices when the network is scaled up. To this end, we propose a topology-aware data aggregation (TADA) protocol, which preserves the advantages of CS-based schemes while alleviating the aforementioned issues. Extensive performance comparisons demonstrate that the data reconstruction error rate of TADA is much lower than other CS-based schemes under test. More important, for most WSN applications, since the underlying physical layer prefers frames with small payloads, the proposed TADA utilizes less transmissions from sensors to UAV in each round of data aggregation, and thus can yields a lower energy consumption.