LUCID: Receiver-aware Model-based Data Communication for Low-power Wireless Networks

Abstract: In the last decade, the advancement of the Internet of Things (IoT) has caused unlicensed radio spectrum, especially the 2.4 GHz ISM band, to be immensely crowded with smart wireless devices that are used in a wide range of application domains. Due to their diversity in radio resource use and channel access techniques, when collocated, these wireless devices create interference with each other, known as Cross-Technology Interference (CTI), which can lead to increased packet losses and energy consumption. CTI is a significant problem for low-power wireless networks, such as IEEE 802.15.4, as it decreases the overall dependability of the wireless network. To improve the performance of low-power wireless networks under CTI conditions, we propose a data-driven proactive receiver-aware MAC protocol, LUCID, based on interference estimation and white space prediction. We leverage statistical analysis of real-world traces from two indoor environments characterised by varying channel conditions to develop CTI prediction methods. The CTI models that generate accurate predictions of interference behaviour are an intrinsic part of our solution. LUCID is thoroughly evaluated in realistic simulations and we show that depending on the application data rate and the network size, our solution achieves higher dependability, 1.2% increase in packet delivery ratio and 0.02% decrease in duty-cycle under bursty indoor interference than state of the art alternative methods.