Search

In this article, the development of a compact 10–14.5 GHz quadrature hybrid with digitally reconfigurable I/Q phase in 130 nm SiGe BiCMOS process is presented. Thanks to the switched capacitance loaded on I/Q path of the quadrature hybrid, the I/Q phase difference can be optimized and digitally reconfigured. The equivalent model is analyzed with even/odd mode theory, and the ABCD matrix is used for the circuit derivation. In order to obtain high coupling coefficient, the broadside coupled line sections are utilized, and compact hybrid size can be realized accordingly. Measured results show that the compact quadrature hybrid has optimized phase difference of 90 ± 1.0° and amplitude difference less than ±0.5 dB for 10–14.5 GHz, with an ultra-compact size of 460 µm × 151 µm, or 0.031λ0 × 0.011λ0. Meanwhile, with the seven reconfigurable phase states, the quadrature hybrid I/Q phase can be digitally reconfigured for a range of 3 degrees to compensate the I/Q phase imbalance in the quadrature system, without DC power consumption.

The in-phase/quadrature (I/Q) imbalance encountered in the zero-IF receiver leads to incomplete image frequency suppression, which severely deteriorates image rejection ratio (IRR) of the receiver system and must be improved using additional analog or digital signal processing. The I/Q linear phase imbalance (LPI) is the key of the I/Q imbalance, which consists of the time delay deviation (TDD) and the local oscillator (LO) phase offset. TDD is negligible in most literature, but it degrades system performance largely for wideband communication systems. This paper proposes a method based on the cross-power spectrum between the I/Q signal to address the estimation problem of LPI. Compared with other conventional methods, the proposed approach calculates LPI parameters simultaneously without any additional hardware. The MATLAB simulation is utilized to evaluate the effectiveness of the presented method. Moreover, the experimental platform of detailed design demonstrates the feasibility of the proposed estimation method, and IRR of the system before and after compensation shows that LPI has been accurately estimated and eliminated with the help of an appropriate compensation structure. Both reveal that the proposed method offers an effective solution to the LPI problem.