| With the development of electronic information technology,miniaturization,low power consumption and large bandwidth become the development trend of future electronic reconnaissance equipment.As a key device in electronic reconnaissance,zero-IF receiver has been widely used with its simple architecture and high integration level.However,zero-IF receivers have the problem of I/Q imbalance error,which can introduce mirror image signals and affect the performance of electronic reconnaissance.Therefore,I/Q out-of-balance calibration is required.In this paper,we firstly propose a wideband I/Q imbalance calibration technique integrating channelization architecture and blind estimation compensation algorithm to solve the I/Q imbalance problem caused by zero-IF receivers.Secondly,this paper proposes a frequency domain blind calibration technique based on the least squares method under the framework of wideband I/Q imbalance calibration,which improves the real-time calibration.The main research contents of this paper are as follows.1.The sources of I/Q imbalance error in zero-IF receivers are analyzed,and the I/Q imbalance error model is established accordingly,and then the corresponding compensation scheme is given.For the problem of high computational complexity and large amount of operations of the narrowband signal I/Q imbalance calibration parameter estimation method,an optimization algorithm is given to reduce the system computational resources,and computer simulation is completed to verify.The simulation results show that the image rejection ratio is improved from 6.95 d B to 73.6711 d B after calibration.2.Based on the narrow-band signal I/Q imbalance compensation algorithm,we further propose a wideband I/Q imbalance calibration technique that integrates the channelized architecture and blind estimation compensation algorithm to achieve "offline estimation and online compensation" for the wideband signal I/Q imbalance problem.The technique divides the wideband signal into narrowband signal by channelization architecture,and calibrates the sub-channel with mirror signal online by using blind estimation algorithm.The measured results show that the average in-band image rejection ratio of the chirp signal reaches 55.6601 d B after calibration.3.To address the problem that the I/Q imbalance parameters of wideband signals cannot be estimated in real time,a blind calibration technique based on the statistical characteristics of the signal frequency domain is proposed to realize the "online estimation and online compensation" of the I/Q imbalance of wideband signals.The technique estimates the amplitude and phase error parameters separately by the least squares method,and uses the estimated error parameters to construct a calibration filter,and then completes the calibration of the broadband signal I/Q imbalance.The measured results show that the in-band average image rejection ratio of the chirp signal is 52.1351 d B.In comparison with the method in(2),this method achieves real-time calibration with less loss of image rejection ratio.4.To verify the actual compensation performance of the calibration technique,an ADRV9009-based radio testbed was built and test experiments were conducted.The experimental results show that the calibration technique proposed in this paper achieves effective calibration of the I/Q imbalance signal under the zero-IF receiver architecture while ensuring that the desired signal is not distorted. |
