Channel Capacity Analysis with Nonlinear Effects of RF Power Amplifiers

The nonlinear effects from the RF circuits impact the channel capacity in wireless communication systems, particularly in scenarios where the power amplifiers introduce significant distortions in the output signals. In this work, an analysis of the channel capacity is presented by considering a complex-valued scalar nonlinear AWGN model with both peakamplitude and average-power constraints. Unlike prior works that focus on idealized linear models, realistic power amplifier nonlinearities are incorporated in this analysis. By following Smith's approach, the structure of the optimal input distribution is found to have discrete amplitude and independent uniform phase. While the presented analysis is general without relying on specific nonlinear models, we demonstrate the validity of the analysis by evaluating two commonly used power amplifier models - the Saleh model and the third-order polynomial model. We observe that the nonlinearity effectively limits the input peak-amplitude and thus the capacity is bounded at high SNR. This work brings practical constraints into fundamental communication analysis and offers insights into system performance under hardware nonlinearities.