Digital feedforward compensation scheme for the non-linear power amplifier with memory

Steady-State Performance Analysis and Step-Size Selection for LMS-Adaptive Wideband Feedforward Power Amplifier Linearizer

“…For arbitrary and , the output of the linearizer can be rewritten as (19) where and . Then the IMD suppression at the linearizer output is IMD (20) Note that only the accuracy of determines the amount of IMD suppression which was the case for memoryless nonlinearity as well.…”

“…Gradient based learning methods implemented either digitally or in analog form have been proposed in [5], [7]- [9], [13], [17], and [19] to adapt SCL and ECL coefficients. In [5], [9], and [19] there are variations in the architecture of feedforward implementation.…”

“…In [5], [9], and [19] there are variations in the architecture of feedforward implementation. In this article, we will focus on coefficient learning and adaptation of the feedforward structure using gradient methods and discrete-time low frequency signals as illustrated in Fig.…”

“…One of the practical problems, however, is that the above feedforward method is actually fairly sensitive to the misadjustment of the underlying component parameters as shown, e.g., in [8], [24], and [26]. Therefore, adaptive feedforward techniques have been developed [5], [7]- [9], [19] with online parameter monitoring, to have better IMD suppression performance under implementation inaccuracies or varying operating conditions. This is the general setup also in this article.…”

Performance of feed‐forward linearizers of power amplifiers in OFDM systems under complex gain errors

“…In the first approach, the amplifier is operated at a point further in the linear region, or equivalently, a point with a large power input back-off (IBO). These complex gains are used to compensate for drifts in the main and error amplifier gains [23][24][25][26][27]. The second approach is to use a linearization technique such as predistortion, feed-forward (FF), feedback, and envelope elimination and restoration approaches [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20].…”