Optimum spacing for a quadratically interpolated look-up table predistorter for cellular power amplifiers

One of the major barriers in modern wireless communication system is the trade-off between the linearity and power efficiency of cellular power amplifier (PA). On the other hand, the non-constant envelope of spectrally competent modulated signals such as wideband code division multiple access (WCDMA) and long term evolution (LTE) interrelates with PA’s non-linearity, causing in-band and out-band distortions. These distortions violate the out-of band emanation requirements which consequents reduce the power efficiency. Digital predistortion (DPD) is one of the most promising techniques to linearize a non-linear power amplifier (PA) in order to reduce the intermodulation distortion (IMD) and enhance the power efficiency. By far, lookup table (LUT) predistorters are the most commonly used approach to alleviate the effects of non-linear PA. This dissertation provides a theoretical analysis of an optimum spacing of a quadratically interpolated LUT predistorter which reduces the quantization error introduced by the LUT approximation supported by extensive simulation experiments. This technique provides better rejection of intermodulation distortion compared to conventional quadratic interpolated LUT, linear interpolated LUT and non-linear interpolated LUT predistorters respectively. Simulation results show that the proposed technique provides an improvement of 20% to 30% of error vector magnitude (EVM) and an improvement of 3 to 4 dBc of adjacent channel leakage ratio (ACLR) at a minimal memory usage for the wideband code division multiple access (WCDMA) and long term evolution system (LTE). This dissertation also proposes an adaptive Least Mean Square (LMS)-based predistorter that is optimized for error compensation introduced by LUT interpolation.

Text FK 2017 134 ir.pdf Download (1MB)

Actions (login required)

View Item