Comparative Analysis of Behavioral Modeling for Wireless Radio Systems

Over the past two decades, various behavioral modeling techniques have been proposed to compensate for the non-idealities in the wireless systems, including quadrature errors, channel imbalance, interleaved spurs, harmonic distortion and intermodulation distortion, to improve radio performance and efficiency. This paper aims to provide an overview of prevalent digital compensator models considering memory effects and bandpass non-linearities as well as model complexity reduction algorithms, including orthogonal matching pursuit and least absolute shrinkage and selection operator, in compact linear regression modeling. More specifically, polynomials, orthogonal polynomials, basis spline functions and radial basis functions are cross examined in time and frequency. Key performance indicators, such as feature space, normalized mean square error, adjacent channel leakage ratio, maximal coefficient value, basis matrix condition, and cumulative energy of leading singular values are demonstrated on a radio frequency power amplifier.