Scalable Double Crosstalk Canceling Digital Predistortion for MIMO Transmitters

Nonlinear distortion becomes more severe due to strong crosstalk between different transmitter branches in multiple-input-multiple-output (MIMO) arrays. A double crosstalk canceling (DCC) digital predistortion (DPD) algorithm is proposed in this paper to solve the power amplifier (PA) nonlinearity, nonlinear and linear crosstalk problems in MIMO transmitters. The algorithm includes two canceling modules to estimate and cancel the nonlinear and linear crosstalk respectively. We propose two different methods named iterative training (IT) and reference signal training (RST), to estimate the crosstalk coupling coefficients. Furthermore, we propose a scalable DCC single-input-single-output (SDCC-SISO) structure. The structure can easily generalize the proposed DCC DPD algorithm from <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2\times 2$ </tex-math></inline-formula> to larger scale MIMO. Simulated and measured results show that the DCC DPD algorithm outperforms other DPD algorithms at strong nonlinear and linear crosstalk levels for <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2\times 2$ </tex-math></inline-formula> MIMO case and its coefficients are reduced by nearly half or more compared with other algorithms. Furthermore, we conduct simulation experiments to test the performance of the SDCC-SISO DPD algorithm for the 3, 4, 5 and 10-branch MIMO cases. SDCC-SISO DPD always maintains good linearization performance and low complexity. Simulation results prove the excellent scalability of SDCC-SISO structure, which is difficult to be achieved by other DPD methods.