This article addresses the modeling and cancellation of self-interference infull-duplex direct-conversion radio transceivers, operating under practicalimperfect radio frequency (RF) components. Firstly, detailed self-interferencesignal modeling is carried out, taking into account the most important RFimperfections, namely transmitter power amplifier nonlinear distortion as wellas transmitter and receiver IQ mixer amplitude and phase imbalances. Theanalysis shows that after realistic antenna isolation and RF cancellation, thedominant self-interference waveform at receiver digital baseband can be modeledthrough a widely-linear transformation of the original transmit data, opposedto classical purely linear models. Such widely-linear self-interferencewaveform is physically stemming from the transmitter and receiver IQ imaging,and cannot be efficiently suppressed by classical linear digital cancellation.Motivated by this, novel widely-linear digital self-interference cancellationprocessing is then proposed and formulated, combined with efficient parameterestimation methods. Extensive simulation results demonstrate that the proposedwidely-linear cancellation processing clearly outperforms the existing linearsolutions, hence enabling the use of practical low-cost RF front-ends utilizingIQ mixing in full-duplex transceivers.
展开▼
