Digital predistortion linearization and crest factor reduction for wideband applications

Power amplifiers are essential components in wireless communication systems and are inherently nonlinear. This nonlinearity generates spectral regrowth beyond the signal bandwidth, which in turn interferes with adjacent channels. Wideband code division multiple access (WCDMA) and orthogonal frequency division multiplexing (OFDM) systems are particularly vulnerable to nonlinear distortions; this is due to their high peak-to-average power ratios (PAPRs), which require a stringent linearity. One way to achieve the required linearity is to back-off the input signal. However, in the case of high PAPR signals, the efficiency of the power amplifier will be very low. In this dissertation, we are concerned with achieving high linearity and high efficiency. We first propose a predistorter based on piecewise pre-equalizers, for use in multi-channel wideband applications. This predistortion linearizer consists of piecewise pre-equalizers, along with a lookup table (LUT) based digital predistorter; together they compensate for nonlinearities, as well as memory effects of power amplifiers. Taking advantage of the multiple finite impulse response (FIR) filters, the complexity is significantly reduced when compared to memory polynomial methods. Furthermore, experimental results obtained when two WCDMA carriers were applied verified that our proposed method provides improvements comparable to those seen using the memory polynomial approach. Secondly, a unique baseband derived radio frequency (RF) predistortion system is presented, which uses LUT coefficients extracted at baseband to directly RF envelope modulate a quadrature vector modulator. The primary advantage of this architecture is that it combines the narrowband benefit of envelope predistortion with the accuracy of baseband predistortion. Finally, a novel efficient crest factor reduction technique for wideband applications is described. The technique uses peak cancellation to reduce the PAPR of the input signal. Conventional iterative peak cancellation requires several iterations to converge to the targeted PAPR, since filtering causes peak re-growth. The proposed algorithm eliminates several iterations and subsequently saves hardware resources. A direct performance comparison between a digitally predistorted and a feed-forward linearized Doherty amplifier is provided, under various crest factor reduction levels.