Joint ML estimation of carrier frequency, channel, I/Q mismatch, and DC offset in communication receivers
Direct-conversion is a low-cost RF architecture that requires fewer external components in chip implementation. It introduces, however, extra RF impairments such as I-Q imbalance and DC-offset that degrade severely the receiver performance if left uncompensated. In this paper, a new method of joint estimation and compensation of frequency, DC-offset, I-Q imbalance and channel is developed for MIMO receivers. Numerical results show that the proposed method has a very good performance; less than 2 dB loss is observed at bit-error rate 10<sup>−4</sup>, as compared to the ideal receiver. In addition, a new training sequence along with a low-complexity architecture is proposed to reduce largely the receiver complexity with negligible performance loss.