Energy-Efficient BLE Device Discovery for Internet of Things

As a short-range and peer-to-peer wireless communication technology, Bluetooth low energy (BLE) receives lots of attentions recently. By enabling discover devices in the proximity, BLE could support various mobile Internet of Things (IoT) applications. Obviously, device discovery plays an essential role on the trade-off between the number of devices discovered and the energy consumption for a mobile IoT device with a constrained resource. This paper subsequently proposes a smart device discovery mechanism, named sDiscovery, to adaptively adjust scan window and scan interval according to the variant of the environment. In particular, we record the identities of devices discovered during each scan window and find out the redundant ones. Intuitively, a larger number of redundant devices encountered implies a more stable environment. In this case, the number of unknown devices encountered becomes smaller and the scan window can be shortened as well as the scan interval can be enlarged to preserve energy. The simulation results show that sDiscovery outperforms the existing solution and conventional BLE mechanism from the perspective of power saving. Moreover, we verify the correctness of simulation model through experiments in TI CC2540 development board, where a prototype of the protocol is implemented.