Search

Abstract Piecewise linearization techniques require dividing the signal into multiple pieces each linearized individually. Machine learning (ML) is one of the useful tools to perform the automatic division of these pieces. Complexity reduction in the classification of piecewise digital predistortion is possible through carefully constructing features from both the signal statistics and the power amplifier (PA) characteristics. Our paper introduces two low-complex classical ML-based methods that facilitate the classification of baseband input data into distinct segments. These methods effectively linearize PA behavior by employing tailored Volterra models corresponding to each segment. Moreover, we perform an in-depth analysis of the proposed schemes to further optimize their classification and regression complexities. The two proposed low-complexity approaches are validated by laboratory experiments and show up to 4 dB error vector magnitude (EVM) improvement over the conventional approach for a class A PA at 28 GHz. Similarly, the EVM improvement is up to 2 dB over the vector-switched general memory polynomial scheme. With only one indirect learning architecture iteration, the two proposed schemes obey the 5G new radio standard up to 6.5 dB and 7 dB output backoff, respectively.

Piecewise linearization techniques require dividing the signal into multiple pieces each linearized individually. Machine learning (ML) is one of the useful tools to perform the automatic division of these pieces. Complexity reduction in the classification of piecewise digital predistortion is possible through carefully constructing features from both the signal statistics and the power amplifier (PA) characteristics. Our paper introduces two low-complex classical ML-based methods that facilitate the classification of baseband input data into distinct segments. These methods effectively linearize PA behavior by employing tailored Volterra models corresponding to each segment. Moreover, we perform an in-depth analysis of the proposed schemes to further optimize their classification and regression complexities. The two proposed low-complexity approaches are validated by laboratory experiments and show up to 4 dB error vector magnitude (EVM) improvement over the conventional approach for a class A PA at 28 GHz. Similarly, the EVM improvement is up to 2 dB over the vector-switched general memory polynomial scheme. With only one indirect learning architecture iteration, the two proposed schemes obey the 5G new radio standard up to 6.5 dB and 7 dB output backoff, respectively. [ABSTRACT FROM AUTHOR]

The Last decade saw significant growth in the digital terrestrial television (DTT) subscribers. The High peak-to-average power ratio (PAPR) of the transmitted signal is implied by the large fluctuations in its amplitude, which is a serious issue in the contemporary and next generation DTT broadcasting systems. These large fluctuations prevent feeding the high power amplifier (HPA) at an operating point near its non-linear (NL) saturation region, thereby lowering its power efficiency. For a better future, energy-efficient green communications is the need of hour. This paper focuses on digital predistortion (DPD), which is among the widely used techniques for HPA linearization in order to minimize the NL effects. DPD technique requires a feedback signal. This paper proposes a novel adding signal-based digital predistortion, where the output of the HPA is gradually linearized by adding a signal at the input in an iterative manner. By Taylor series approximation, the DPD problem can be posed as a convex one. We prove that a tighter upper bound exists for the convergence factor and instead of directly solving the convex problem, we pick a convergence factor value within its bounds that offers a good trade-off between convergence speed and residual error value. The simulation results show that the proposed technique can quickly linearize the HPA with good convergence. [ABSTRACT FROM AUTHOR]