Adaptive digital predistorter for power amplifiers with real time modeling of memoryless complex gains

When the signals resulting from linear modulation methods, like M-ary QAM, are passed through a nonlinear power amplifier, their fluctuating envelope causes distortion and spectral spreading. In order to avoid these effects, while maintaining both power and spectral efficiency, the use of linearization techniques is necessary. This paper presents a digital predistorter with real time modeling of AM-AM and AM-PM characteristics of a power amplifier (PA). The input and output lowpass equivalent complex envelopes of the amplifier are sampled, scaled and updated into a lookup table to provide the predistorted signal. An improvement of 45 dB of out-of-band power is obtained when simulating with Signal Processing WorkSystem (SPW). The proposed technique is robust and efficient since no iterative procedure is needed, hence the convergence time is eliminated.