An Aerial-Computing-Assisted Architecture for Large-Scale Sensor Networks

Wireless sensor networks have been widely adopted in various areas. However, large-scale sensor networks face four critical problems. Information transmissions through the network would cause the energy hole problem; deploying sensors in a distributed manner might cause the isolated problem; simultaneous transmissions lead to the communication collision problem; and the lack of intelligent computing incurs the limited computation problem. Due to drones' high mobility and flexibility, drone-based aerial computing emerges as a new paradigm to address these problems. In this article, we propose an aerial-computing-assisted architecture for large-scale sensor networks; the core layer contains the following modules: sensor localization, trajectory planning of drones, communication control between drones and sensors, and enabling intelligent computation on drones. The architecture significantly advances large-scale sensor networks with higher efficiency, smaller delay, larger lifetime, and more powerful computing capability.