Multichannel Neighbor Discovery in Bluetooth Low Energy Networks: Modeling and Performance Analysis

Zhong Shen,Qinghai Yang,Hai Jiang
DOI: https://doi.org/10.1109/tmc.2021.3113349
IF: 6.075
2021-01-01
IEEE Transactions on Mobile Computing
Abstract:Bluetooth Low Energy (BLE) has become one of the enabling wireless technologies to facilitate the Internet of Things. Neighbor discovery is critical in BLE communications. BLE uses multiple (three) channels in neighbor discovery. It is challenging to achieve low-latency and low-energy-consumption BLE neighbor discovery due to the lack of analytical models for multichannel neighbor discovery. In this paper, we study BLE multichannel neighbor discovery for two advertising modes specified by BLE: periodic deterministic advertising (PDA) and pseudo-random delay advertising (RDA). We build two generic models, BLE 3-Circle model and BLE 1-Circle model, for characterizing BLE multichannel neighbor discovery. For PDA mode, we present a necessary and sufficient condition for BLE multichannel neighbor discovery, and provide a guideline for parameter setting. With the guideline, we derive the expected discovery latency in closed form, and demonstrate that the expected discovery latency is very close to a theoretical lower bound. For RDA mode, we build an analytical model based on Markov chain to accurately compute the expected discovery latency. Simulation and experimental results show accuracy of our analytical works. Interestingly, our parameter setting guideline works well for both PDA and RDA modes.
What problem does this paper attempt to address?