Digitally Assisted Feedforward Compensation of Cartesian-Feedback Power-Amplifier Systems

A Polynomial Analog Baseband Predistorter for Compensation of Wireless Amplifier’s Distortion

“…In complex format. The output of PA is given as: (5) It is important to note that, the amplitude and phase distortions only depend on the amplitude and independent of the phase distortion [19]. Therefore, we can first compensate for the amplitude distortion followed by the phase correction.…”

“…In order to compensate the exhibited distortion, one common approach relies on pre-distorting the input signal by mimicking the inverse nonlinear behavior of PAs, called predistortion (PD). A number of different linearization methods including the feed-forward [3], [4], feedback [5], digital PD [6] and analog PD [7], [8] have been widely employed to improve the linear characteristic of the PA. As for implementation, the digital PD will no longer be the optimum candidates. This is because of their bandwidth limitations and the need for high power supply [1], [2].…”

Power amplifier linearisation using digital predistortion and multi‐port techniques

“…where f F and f G are phase shifts of the predistorter and the amplifier, respectively, and also depend on the input magnitude. In other words, (8) suggests that the predistorter applies phase rotations against those of the amplifier (AM/PM). Amplifier output can be expressed as…”

“…There are different methods for linearising HPA such as feedforward, feedback, linear amplification with non-linear components and predistortion, which can be used to improve the HPA linearity without sacrificing its efficiency [1,[7][8][9][10][11]. The literature search shows that from all the linearisation techniques that have been developed, the predistortion is the most commonly used in the new systems today [3,5,[12][13][14][15][16][17].…”

Power Amplifier Behavioral Modeling by Neural Networks and Their Implementation on FPGA

“…It achieves high linearity, but it greatly reduces the power added efficiency. To increase efficiency and minimize distortions, many linearization techniques have been proposed in literature such as feedback [3], feedforward [3], linear amplification with nonlinear components (LINC) [4], digital pre-distortion (DPD) that uses Volterra series [2], [5], memory polynomials [6], lookup table (LUT) [7] and neural networks models.…”