Nonlinear Compensation for High Power Amplifiers using Genetic Programming

Nonlinearities are inherent in a high power amplifier (HPA) resulting in undesirable distortion in communication systems. A predistortion module is usually cascaded in front of the HPA to compensate these nonlinearities. The structure of the predistorter is typically unknown and must be identified to attain a linear amplification. In order to achieve an autonomous system identification of the unknown predistorter structure, this paper presents an evolutionary algorithm based on genetic programming (GP). GP is used to search for an optimal rational model structure and is combined with a least-squares parameter estimator to estimate the parameters of the evolved models. The effectiveness of the proposed identification scheme has been verified through experiments and comparison with conventional predistortion techniques.