Advanced Signal Processing Techniques for Impairments Compensation and Linearization of SISO and MIMO Transmitters.

摘要

Modern wireless communication systems achieve higher data rates and better link reliability by using complex modulation and transmission techniques which add more constraints on the acceptable amount of distortions and nonlinearities in radio communication systems.;First, the focus was on the linearization and distortion compensation of a single-branch direct up-converter transmitter. Two sources of distortion -- the modulator's in-phase/quadrature (I/Q) imbalance and the power amplifier -- were investigated; and, two linearization approaches have been developed, in order to compensate for both sources of distortion and nonlinearity.;The work was then extended to the linearization of multi-branch transmitters. It was demonstrated that nonlinear crosstalk could degrade the overall performance of a transmitter. A new matrix memory polynomial technique was introduced to develop an accurate forward model and to employ in the development of a new crossover digital predistortion linearization algorithm in MIMO Transmitters.;The last topic of the dissertation concentrates on the linearization and distortion compensation of dual-band and multi-carrier transmitters. With conventional digital predistortion linearization techniques, the linearization of dual-band and multi-carrier transmitters must meet the sampling rate requirement of analog-to-digital converters (ADCs) and digital-to-analog converters (DACs), which should be at least equal to the Nyquist sampling rate of the signal multiplied by the order of the nonlinearity. In a multi-carrier and dual-band system with a broad frequency offset between the two carrier frequencies, this sampling rate requirement would be a big challenge. A novel multi-cell processing architecture is proposed that was developed to selectively compensate for the nonlinearities and distortions of the transmitter in different frequency bands, which significantly reduces the required sampling rate. Based on this new architecture, two linearization architectures were developed to compensate for the inter-band and intra-band distortions of dual-band and multi-carrier transmitters.;RF transmitters are composed of linear and nonlinear circuits, which make them composite and dynamic nonlinear systems. As a result, the quality and performance of a wireless communication system depends significantly on the performance of the RF transmitter. This dissertation proposes novel advanced signal processing techniques for the estimation and compensation of distortions and nonlinearities in modern single-input, single-output (SISO) and MIMO transmitters.