OpenDPDv2: A Unified Learning and Optimization Framework for Neural Network Digital Predistortion
Preprint, accepted to the 2026 IEEE Global Communications Conference (GLOBECOM), 2026
Learning better signals.
AI for RF & mixed-signal systems
We bring neural networks into the signal chain: modeling and linearizing power amplifiers, identifying load conditions, and exploring learning-based calibration for data converters. Algorithm–hardware co-design connects model efficiency to implementation on FPGAs and silicon.
Selected work.
All publicationsDeep-Learning Based Hardware-Efficient Load Identification for Power Amplifiers
in IEEE Journal of Selected Topics in Electromagnetics, Antennas and Propagation (JSTEAP), 2026.
DPD-NeuralEngine: A 22-nm 6.6-TOPS/W/mm2 Recurrent Neural Network Accelerator for Wideband Power Amplifier Digital Pre-Distortion
accepted to 2025 IEEE International Symposium on Circuits and Systems (ISCAS).
SparseDPD: A Sparse Neural Network-based Digital Predistortion FPGA Accelerator for RF Power Amplifier Linearization
accepted to 35th International Conference on Field-Programmable Logic and Applications (FPL), Leiden, The Netherlands, 2025.
DeltaDPD: Exploiting Dynamic Temporal Sparsity in Recurrent Neural Networks for Energy-Efficient Wideband Digital Predistortion
in 2025 IEEE MTT-S International Microwave Symposium (IMS). (IMS 2025 Top 50 Paper, invited to IEEE MWTL)
TCN-DPD: Parameter-Efficient Temporal Convolutional Networks for Wideband Digital Predistortion
accepted to 2025 IEEE MTT-S International Microwave Symposium (IMS), 2025.