BCS: Bidirectional Cross-Interface Scheduling for Sustainable End-to-End Data Delivery in IoT

Abstract

In this era of Internet of Things (IoT), a variety of battery-powered devices communicate with each other or some remote gateways via wireless multihop communications. Although Wi-Fi has been shown to be of great benefit to a wide range of IoT applications due to its high data rate, high reliability, and native IP compatibility, it is inherently energy-hungry and potentially suffers from a dilemma: more (less) frequent use of Wi-Fi results in shorter (longer) delay but higher (lower) energy consumption. This dilemma becomes the major obstacle in the use of Wi-Fi in long-term but delay-bounded IoT applications. To address this, we propose a bidirectional cross-interface scheduling (BCS) scheme, which leverages the widely used low-power ZigBee to collaborate with the high data rate and reliable Wi-Fi for sustainable and reliable end-to-end (E2E) data delivery in IoT. The objective is to maximize network lifetime while satisfying certain required E2E delay bound. Extensive simulations and testbed experiments have been conducted. The results show that the network lifetime of BCS is 33.4% and 346.9% longer than those of a state-of-the-art scheme and the IEEE 802.11's standard power-saving scheme with the default settings, respectively.