Post-compensation of RF non-linearity in mobile OFDM systems by estimation of memory-less polynomial

The high peak-to-average-power-ratio (PAPR) of OFDM systems introduces inevitable nonlinear distortion in the transmitter and causes both in-band distortion and out-of-band spectrum re-growth. In this paper we propose a novel scheme to compensate for the nonlinearity of RF transmitters with solid-state-power-amplifiers (SSPA) in OFDM by least square (LS) estimation of the nonlinear parameters modeled by memoryless polynomials. The compensation is easily achieved by subtracting the higher-order polynomials of the input from the RF output. Simulation results for spectrum masking as well as the BER curve in 16-QAM OFDM show very stable and promising performance as well as simple implementation.