TCN-DPD: Parameter-Efficient Temporal Convolutional Networks for Wideband Digital Predistortion

Digital predistortion (DPD) is essential for mitigating nonlinearity in radio frequency (RF) power amplifiers, particularly for wideband applications. This paper presents TCN-DPD, a novel parameter-efficient architecture based on temporal convolutional networks. By integrating noncausal dilated convolutions with optimized activation functions, our approach achieves superior linearization performance while using significantly fewer parameters than existing deep neural network solutions. Evaluated on the \texttt{OpenDPD} framework with the DPA_200MHz dataset, TCN-DPD demonstrates superior linearization performance with only 500 real-valued parameters, achieving simulated ACPRs of -51.58/-49.26 dBc (L/R), EVM of -47.52 dB, and NMSE of -44.61 dB. These results establish TCN-DPD as a promising solution for efficient wideband PA linearization.