Embedded Files
1. Bayesian inference and receiver algorithms for massive MIMO
We derived variational Bayesian receivers that estimate channels and detect data jointly while accounting for imperfect hardware: spatial sigma-delta and few-bit ADCs, quantized fronthaul links, and time-varying or near-field channels. The algorithms stay low-complexity, and I have extended them into deep-unfolding networks.
Refs:
T.-V. Nguyen, S. Nassirpour, I. Atzeni, A. Tolli, A. L. Swindlehurst, and Duy H. N. Nguyen, "MIMO Detection with Spatial Sigma-Delta ADCs: A Variational Bayesian Approach," IEEE Transactions on Signal Processing, vol. 74, Aug. 2025.
S. Nassirpour, T.-V. Nguyen, H. Q. Ngo, L.-N. Tran, T. Ratnarajah, and Duy H. N. Nguyen, "Variational Bayesian Channel Estimation and Data Detection for Cell-Free Massive MIMO with Low-Resolution Quantized Fronthaul Links," IEEE Journal of Selected Topics in Signal Processing, Jun., 2025.
T.-V. Nguyen, S. Nassirpour, Rose Q. Hu, and Duy H. N. Nguyen, "Joint Gridless mmWave Channel Estimation and Data Detection in Massive MIMO Networks," IEEE Transactions on Vehicular Technology, under revision, 2026.
2. Intelligent surfaces and advanced MIMO architectures
We optimized beamforming, user pairing, and power allocation for next-generation array architectures: RIS and STAR-RIS, stacked intelligent metasurfaces, fluid antennas, cell-free massive MIMO, and rate-splitting multiple access. Where the problem is too large for exact optimization, I use deep learning to approximate the solution.
Refs:
R. H. Y. Perdana, T.-V. Nguyen, Y. Pramitarini, Duy H. N. Nguyen, and B. An, "Enhancing Spectral Efficiency of Short-Packet Communications in STAR-RIS-Assisted SWIPT MIMO-NOMA Systems with Deep Learning," IEEE Transactions on Wireless Communications, vol. 24, no. 1, Jan. 2025.
3. Deep learning for wireless signal intelligence
We designed convolutional and attention-based networks that classify modulation in impaired MIMO-OFDM links, recognize radar waveforms, detect drones from RF signals, and sense the spectrum for 5G and LTE. The architectures aim for high accuracy at low cost, and I have also written a survey of deep architectures for modulation classification.
4. Short-packet, NOMA and energy-harvesting IoT networks
We analyzed and designed cooperative IoT networks that use non-orthogonal multiple access, energy harvesting, and short-packet transmission for low-latency links. My papers combine exact performance analysis (outage and fading models, imperfect CSI and SIC) with deep-learning evaluation, and a Korean patent covers hybrid user pairing.
Google Sites
Report abuse