Emanuele Cardillo
ExistingIQimbalance correction methods are often impractical or challenging in real cases. Indeed, they usually require hardware modification or expensive precision actuators to create accurate and, above all, small movements. This last characteristic is strategic; indeed, if the range varies significantly during measurement, the signal-to-noise ratio (SNR) is not constant, and thus, theIQmodule varies, too. As explained in the text, a constant SNR during calibration is often required to ensure that theIQimbalance is solely due to hardware imperfections and not to changing environmental conditions. This poses several practical challenges in replicating the procedures described in the current literature. In this article, an automaticIQimbalance calibration method is proposed; it does not require any specific measurement settings. It finds theIQtrajectory with the maximum SNR, thus ensuring not only a constant SNR but also optimal performance, as the accuracy of ellipse-fitting methods is directly proportional to the SNR. The theoretical basis of the calibration method is carefully provided and validated with simulations and measurements. Its effectiveness is investigated both for displacement and micro-Doppler signature detection. This method can be used for many radar applications, from physiological and healthcare sensing to motion tracking and gesture detection, as shown experimentally; moreover, it can be implemented online, thus enabling the system’s automatic calibration.