Research
Dependable Wireless Power Transfer for Batteryless IoT
Wireless power transfer (WPT) is a promising technology for powering batteryless and energy-autonomous IoT devices. However, reliable energy delivery remains challenging due to dynamic wireless channels, asynchronous power transmissions, and the nonlinear behavior of energy-harvesting circuits. Counterintuitively, adding more power sources may not always improve energy availability and can even cause energy black holes, significantly degrading system performance. To address these challenges, we investigate dependable WPT, with a focus on understanding and mitigating energy unreliability through analytical modeling, system design, and experimental evaluation, ultimately enabling predictable and reliable energy delivery for batteryless IoT systems. The related papers have been published in IEEE WCM 2022 [PDF], IEEE WCM 2024 [PDF], IEEE WCM 2025 [PDF], IEEE TMC 2025 [PDF], IEEE TII 2026 [PDF], and IEEE JSAC 2026 [PDF].
ChirpBox: An Infrastructure-Less LoRa Testbed
A key obstacle hindering the development of large-scale LoRa testbeds outdoors is the lack of a backbone infrastructure allowing them to communicate with the nodes and supply them with power easily. As a result, many LoRa installations are deployed indoors or only support a handful of outdoor devices, which does not allow proper testing. To bridge this gap, we built ChirpBox, an infrastructure-less LoRa testbed. ChirpBox is open source, and the tutorial website can be found at [Link]. The related papers have been published in EWSN 2021 [PDF], SenSys-DATA 2021 [PDF], IPSN 2022 [PDF], and IEEE TCCN 2025 [PDF].
Dynamic Mapping of Environmental Noise with IoT
Environmental noise poses a growing threat to public health and sustainable urban development. However, conventional noise-mapping approaches largely rely on costly manual measurements or computational models that provide coarse-grained and often static estimates. To address this gap, we investigate DAMPEN, a versatile wireless acoustic sensor network for real-time, fine-grained, and large-scale environmental noise mapping. DAMPEN integrates acoustic sensor networks, mobile crowdsensing, edge-cloud computing, and AI-assisted analytics to support dynamic visualization, noise-source recognition, personal exposure assessment, and data-driven noise mitigation. Our research further explores dependable multimedia transmission, energy-efficient and event-triggered sensing, and defenses against inaudible sound attacks. The long-term goal is to build a low-cost, scalable, easy-to-maintain, sustainable, dependable, and trustworthy sensing infrastructure for smart cities and industrial environments. Related papers have been published in IEEE Network 2020 [Paper] and IEEE Industrial Electronics Magazine 2021 [Paper].
Dependable Wireless Network Protocols for IoT
Wireless networking in the Internet of Things is challenging because a massive number of devices in a relatively small region need to be interconnected. Particularly, the CSMA/CA operation is not a viable solution since a dense network leads to high channel contention. Moreover, external radio interference can undermine network dependability. Thus, we proposed and implemented protocols to address these challenges. This research work was supported by NSFC. The related papers have been published in PPNA 2018 [PDF], INFOCOM 2020 [PDF], and ICNP 2020 [PDF].
On-Demand Railway Bridge Structural Health Monitoring
Over 73,000 railway bridges are older than 110 years in Europe. The IoT is promising for railway bridge health monitoring. However, existing energy-efficient approaches, such as duty cycling and energy harvesting, face challenges in this application due to the unpredictability of train passages and insufficient ambient energy around bridges. We proposed EcoVibe, the first railway bridge monitoring IoT system that provides on-demand sensing with near-zero idle energy dissipation. This research work was supported by VINNOVA. The related papers have been published in IEEE ComMag 2016 [PDF] and IEEE IoTJ 2019 [PDF].
Harvest Energy from Water: A Green Water Quality Sensing System
Water quality data is crucial and valuable, but its acquisition is not always trivial. A promising solution is to distribute a wireless sensor network in water to measure and collect the data. However, a drawback exists in that the system's batteries must be replaced or recharged after being exhausted. To mitigate this issue, we designed a self-sustained and on-demand water quality sensing system powered by renewable bioenergy from microbial fuel cells. This research work was supported by the Exploratory Advanced Research Program of FHWA. The related papers have been published in IEEE ComMag 2016 [PDF] and ACM TECS 2017 [PDF].