Portrait of Jialuo Du

Jialuo Du

Research Associate · Department of Computer Science and Engineering · Michigan State University

dujluo@gmail.com Google Scholar GitHub

I am a Research Associate in the Department of Computer Science and Engineering at Michigan State University, advised by Dr. Zhichao Cao.

I earned my Ph.D. in Software Engineering from Tsinghua University in 2025, where I was advised by Professor Yunhao Liu. I also received my Bachelor of Engineering in Computer Science and Technology from Tsinghua University in 2020.

My research focuses on the intersection of the Internet of Things and Artificial Intelligence.

Publications

Highlights

DeRa: Deep-Range Low-Power Wide Area IoT via Ultra-sensitive Signal Detection

Jialuo Du, Jingkai Lin, Zhichao Cao

MobiCom 2026To appearACM International Conference on Mobile Computing and Networking

Weak LoRa packets are lost at detection, before decoding even starts. Carrier frequency offset rotates the phase of every preamble chirp, so existing detectors add only magnitudes and waste signal power. DeRa models that phase evolution, compensates the CFO-induced timing, frequency and leakage distortions (up to 40 kHz CFO), and accumulates the preamble coherently. On COTS devices across SF 7–12 it lowers the detection threshold by 4.3–4.8 dB and the decoding threshold by 3.3–3.9 dB.

▶ Video (1:30)Paper and code coming soon

LoRaTrimmer: Optimal Energy Condensation with Chirp Trimming for LoRa Weak Signal Decoding

Jialuo Du, Yunhao Liu, Yidong Ren, Li Liu, Zhichao Cao

MobiCom 2024ACM International Conference on Mobile Computing and Networking

Real LoRa symbols contain a frequency jump that standard dechirp-and-FFT demodulation ignores, adding noise and a random phase jump. LoRaTrimmer trims the FFT window at the jump and handles the phase jump probabilistically, gaining 1.70–2.49 dB SNR over the state of the art (up to 1.67× coverage area). Splitting the DFT matrix into two anti-diagonal triangles computes every bin's trimmed FFT at once, in O(n3/2) time.

Interactive demo Paper Code Kaggle Slides

SRLoRa: Neural-enhanced LoRa Weak Signal Decoding with Multi-gateway Super Resolution

Jialuo Du, Yidong Ren, Zhui Zhu, Chenning Li, Zhichao Cao, Qiang Ma, Yunhao Liu

MobiHoc 2023ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing

A neural LoRa decoder that fuses the same weak packet as heard by multiple gateways. Interleaved denoising and merging layers exploit spatial diversity to decode signals far below the noise floor.

Paper

2026

Kairos system overview: gateway decides when to calibrate, node decides how

Kairos: Soil-Aware Transmission Control For Long-Term Agricultural Cross-Soil LoRa

Jingkai Lin, Yidong Ren, Jialuo Du, Yimeng Liu, Yawen Wang, Younsuk Dong, Tianxing Li, Zhichao Cao

ICNP 2026To appearIEEE International Conference on Network Protocols

LoRa antennas buried in farmland drift out of polarization alignment as soil moisture changes. Kairos predicts when recalibrating will actually save energy and then finds a good polarization state with few probes, cutting energy use by up to 3.71× in a real agricultural field.

Paper coming soon
Overview of AURA: driving telemetry, multi-scale learning, context fusion and actionable intelligence

AIoT-based Continuous, Contextualized, and Explainable Driving Assessment for Older Adults

Yimeng Liu, Fangwei Zhang, Maolin Gan, Jialuo Du, Jingkai Lin, Yawen Wang, Fei Sun, Honglei Chen, Linda Hill, Ruofeng Liu, Tianxing Li, Zhichao Cao

arXiv 2026arXiv:2603.00691

AURA is a vision for assessing older adults' driving safety continuously from everyday trips rather than occasional clinic visits. It combines in-vehicle sensing, multi-scale behavior modeling and context-aware analysis on a privacy-preserving edge architecture.

arXivPDF

2025

LoRaSeek denoising stages: noisy input spectrogram, global features, decoder output, denoised chirp

LoRaSeek: Boosting Denoising Ability in Neural-enhanced LoRa Decoder via Hierarchical Feature Extraction

Khang Nguyen, Yidong Ren, Jialuo Du, Jingkai Lin, Maolin Gan, Shigang Chen, Mi Zhang, Chunyi Peng, Zhichao Cao

MobiCom 2025ACM International Conference on Mobile Computing and Networking

A lightweight denoiser for neural LoRa decoding that combines CNNs, a compact Transformer and a hierarchical U-Net. It gains up to 3.03 dB over prior neural decoders while cutting model storage by up to 7.4×.

Paper

2024

Schematic: a buried node with a fixed antenna loses polarization alignment with the gateway; Demeter's programmable antenna re-aligns it

Demeter: Reliable Cross-soil LPWAN with Low-cost Signal Polarization Alignment

Yidong Ren, Wei Sun, Jialuo Du, Huaili Zeng, Younsuk Dong, Mi Zhang, Shigang Chen, Yunhao Liu, Tianxing Li, Zhichao Cao

MobiCom 2024ACM International Conference on Mobile Computing and Networking

A low-cost, low-power programmable antenna that realigns signal polarization on single-RF-chain COTS LoRa radios, keeping links from sensors buried in soil reliable as soil conditions change.

Paper

2023

Schematic: backscatter tags convert a COTS linear chirp into different non-linear chirps that share one channel

Prism: High-throughput LoRa Backscatter with Non-linear Chirps

Yidong Ren, Puyu Cai, Jinyan Jiang, Jialuo Du, Zhichao Cao

INFOCOM 2023IEEE International Conference on Computer Communications

Backscatter tags convert the linear chirps of ordinary COTS LoRa nodes into different non-linear chirps with a lightweight, low-power frequency-shift design, so many tags can transmit concurrently in the same channel.

Paper