CAREER: LoRa Enabled Space-air-ground Integrated Networks for Next-Generation Agricultural IoT

Project Overview

The Internet of Things (IoT) is poised to play a critical role in the future of precision agriculture, aiming to boost crop yields while reducing management costs. In particular, agricultural IoT bridges the gap between traditional weather-based decision-making and modern data-driven intelligence by enabling low-cost, low-power data collection in natural, often harsh, environments. Long Range (LoRa) technology—a promising wide-area IoT solution—has the potential to drive this transformation. However, deploying LoRa for next-generation agricultural IoT poses significant challenges, including heterogeneous deployment environments, limited power availability, and the lack of network backhaul in rural areas. This project seeks to make foundational contributions toward the development of space-air-ground integrated agricultural IoT systems in the emerging era of 6G. As part of this CAREER project, I will explore this pioneering domain by designing reliable, sustainable, and ubiquitous agricultural IoT systems. In collaboration with a partner, I plan to deploy the proposed system across several large-scale farms in Michigan to collect rich sensory data, enabling cross-disciplinary research, education initiatives, and industry engagement. Additionally, the project includes the development of comprehensive educational components and active participation in outreach efforts to broadly impact graduate and undergraduate students, underrepresented groups, and K–12 teachers and students.

Intellectual Merits

The limitations of existing approaches inspire me to fundamentally rethink the role of LoRa in agricultural IoT by pursuing a novel design paradigm that enriches the LoRa network architecture with three key components: reliable cross-soil communication, energy-efficient drone-based energy excitation, and scalable low Earth orbit (LEO) satellite-based opportunistic backhaul. In this CAREER proposal, I aim to advance LoRa technology by introducing a new network framework tailored to the unique demands of agriculture—enabling cross-soil, ultra-low energy, and large-scale IoT deployments. The ultimate goal is to build a reliable, sustainable, and ubiquitous IoT infrastructure for precision agriculture. The intellectual merits of this project are threefold:

  • Beyond surface-level deployments – I propose to expand the design space of LoRa configurations to support reliable cross-soil communication for underground sensors, addressing a critical gap in current systems.
  • Beyond conventional gateways – Rather than merely using drones and satellites as mobile LoRa gateways, I will assign them new roles: drones as sources of targeted energy excitation and satellites as opportunistic mobile backhaul. This requires the development of novel communication protocols to significantly enhance network energy efficiency and throughput.
  • System-wide optimization – I will co-design hardware and communication protocols across nodes, gateways, drones, and satellites to reduce computational complexity and energy consumption, thereby enabling the practical deployment of large-scale agricultural IoT systems.

Broader Impact

  • Multidisciplinary Impacts: The successful completion of this project will not only advance wireless IoT communication technologies in agriculture but also lead to the development of novel networking infrastructure and hardware that significantly improve power efficiency and system scalability in complex deployment environments. The proposed research is expected to have broad impact beyond agriculture, benefiting a wide range of IoT and cyber-physical system (CPS) applications such as smart cities and Industry 4.0. By introducing new classes of low-power wireless systems, this work will reduce the power consumption of existing IoT sensors, minimize maintenance requirements, and enhance network throughput and scalability.
  • Impact on Undergraduate and Graduate Research: Undergraduate and graduate research is a very important component of the PI’s career as educators and researchers. For the proposed project, the PIs will actively recruit undergraduate and graduate students and work closely with them as collaborators and mentors, and will co-author with them on potential publications.
  • Dissemination: The research outcomes of this project will be disseminated through multiple channels to ensure broad visibility and impact. These include: (1) publication in top-tier conferences and journals, (2) release of open-source hardware and software developed during the project, and (3) distribution of anonymized and ethically approved datasets. Additionally, deployment experiences and findings from this agricultural IoT work will be shared with the broader community through presentations at key regional outreach events, including the Great Lakes Fruit, Vegetable, and Farm Market Expo, the Michiana Irrigation Winter Workshop, the Southwest On-Farm Demonstration Field Day, the Michigan Soybean and Potato Field Days, and the Potato Winter Conference.

Reseaerch Activities

  • Year 1:
    • For the research of LEO IoT satellite (Thrust 3), we first introduce a bursty link model that predicts the number of transmittable packets within a transmission window, reducing energy waste from failed uplink transmissions. Moreover, we enhance the model by selecting informative features and optimizing the window length. Additionally, we develop a multi-hop flooding protocol that enables gateways to buffer and share data packets across the network while incorporating a priority data queue to avoid duplicate transmissions. We implement the designed protocols with commercial-off-the-shelf (COTS) IoT satellites and LoRa radios, then evaluate its performance based on real deployment and real-world collected traces.
    • For the research on leaf wetness sensing (Evaluation Application), we integrate millimeter-wave (mmWave) radar with camera technology to detect leaf wetness by determining if there is water on the leaf. Firstly, we design a Convolutional Neural Network (CNN) to selectively fuse multiple mmWave depth images with an RGB image to generate multiple feature images. Then, we develop a transformer-based encoder to capture the inherent connection among the multiple feature images to generate a feature map, which is further fed to a classifier for detection. We also implemented the sensing system with COTS mmWave radar and RGB cameras and evaluated its performance in controlled environments and real fields at Michigan State University.
  • Year 2:
    • For the research of drone-based LoRa backscatter networking system (Thrust 2), we integrate aerial excitation sources and backscatter tags to enable efficient agricultural IoT. Firstly, we co-design the excitation source and tag with a customized packet format to enable decoding for multiple tags. Secondly, we propose excitation cells to achieve optimal throughput and symbol error rate. Finally, we devise two aerial routing strategies to optimize system energy efficiency and coverage reliability for arbitrary agricultural sensor deployments. We implement the designed system with customized lowcost hardware, signal processing via software-defined radio on TV white space spectrum, and evaluated in real-world scenarios
    • For the research on leaf wetness sensing (Evaluation Application), beyond the multi-modality design, we focus on signle modality - mmWave only to improve system cost and efficiency. First, we utlize the temporal-correlation in the continuously changing process of leaf wetness to develop a sensing system that measures the fine-grained levels of leaf wetness. Moreover, we use the idea of knowledge transfer to teach mmWave-based model with the RGB-image based model, enhancing the detection accuracy with single mm-Wave modality.

Education Activities

  • Year 1:
    • Jadon Jones, recruited from Pathway to Research (PTR) program at MSU, joins the leaf-wetness project to learn data analysis as an undergraduate researcher.
  • Year 2:
    • A benchmark of leaf wetness detection, Hydra-Bench, has been developed as a course project for CSE 891.
    • Soham Bamane and Jacob Patton from the CSE department at MSU, joins our team to develop the end-to-end agricultural IoT system as undergraudate researchers.

Outputs

  • Year 1:
    • Yidong Ren, Amalinda Gamage, Li Liu, Mo Li, Shigang Chen, Younsuk Dong, Zhichao Cao, "SateRIoT: High-performance Ground-Space Networking for Rural IoT", accepted by Proceedings of ACM the 24th Annual International Conference on Mobile Computing and Networking (MobiCom 2024)
    • Yimeng Liu, Maolin Gan, Huaili Zeng, Li Liu, Younsuk Dong, Zhichao Cao, "Hydra: Accurate Multi-Modal Leaf Wetness Sensing with mm-Wave and Camera Fusion", accepted by Proceedings of ACM the 24th Annual International Conference on Mobile Computing and Networking (MobiCom 2024)
  • Year 2:
    • Yidong Ren, Gen Li, Yimeng Liu, Younsuk Dong, Zhichao Cao, "AeroEcho: Towards Agricultural Low-power Wide-area Backscatter with Aerial Excitation Source", accepted by Proceedings of IEEE International Conference on Computer Communications (INFOCOM 2025)
    • Yimeng Liu, Maolin Gan, Gen Li, Younsuk Dong, Zhichao Cao, "Adonis: Neural-enhanced Fine-grained Leaf Wetness Sensing with Efficient mmWave Imaging", accepted by Proceedings of IEEE International Conference on Computer Communications (INFOCOM 2025)
    • Yimeng Liu, Maolin Gan, Huaili Zeng, Yidong Ren, Gen Li, Jingkai Lin, Younsuk Dong, Xiaobo Tan, Zhichao Cao, "Proteus: Enhanced mmWave Leaf Wetness Detection with Cross-Modality Knowledge Transfer", accepted by Proceedings of ACM the 23rd Conference on Embedded Networked Sensor Systems (SenSys 2025)
    • Yimeng Liu, Maolin Gan, Yidong Ren, Gen Li, Jingkai Lin, Younsuk Dong, Zhichao Cao, "Hydra-Bench: A Benchmark for Multi-Modal Leaf Wetness Sensing", arXiv preprint arXiv:2507.22685

Personnel

Collaborator

  • Dr. Younsuk Dong (The Department of Biosystems and Agricultural Engineering at MSU)
  • Dr. Xiaobo Tan (The Department of Electrical and Computer Engineering at MSU)