Optimal Training Sequences for Joint Channel and Frequency-Dependent IQ Imbalance Estimation in OFDM-based Receivers

Nowadays OFDM is combined with advanced multiple access and multiple antenna techniques in order to improve the link capacity. In this context, a lot of effort is spent on developing inexpensive wireless OFDM-based receivers. Direct-conversion radio frequency receivers are appealing because they avoid costly IF filters. This kind of receivers implies analog RF I/Q separation introducing an unwanted in-band image interfearence encedue to the mismatch between the in-phase and quadrature branches. Unfortunately, systems combining OFDM with multiple antennas and multiple access techniques are very sensitive to I/Q mismatch, mostly when high order modulation schemes are applied. A digital compensation of this un-wanted effect is required. In this paper, we propose a method for estimating jointly the frequency dependent I/Q imbalance and the propagation channel in the frequency domain. First the channel is estimated based on a pilot optimized to suppress the interference caused by I/Q imbalance. Second I/ C imbalance is estimated and compensated relying on the estimated channel. Both steps are low cost in terms of implementation complexity. Finally, it is shown that the performance of the channel estimate is not degraded by I/Q imbalance. The proposed algorithm enables the system to work at high SNRs.