Research And Implementation Of Ultra-Wideband Zero-IF Receiver And Its Error Compensation Technology

In the broadband transceiver architecture,the zero-IF architecture based on the principle of direct down conversion has the advantages of simple analog circuit structure,low power consumption,and low cost.However,due to the defects caused by its own structure and analog devices,the performance of the zero-IF structure is limited,and the I/Q(In-phase / quadrature)imbalance due to branch mismatch is one of the main reasons.For a wideband multi-channel transceiver,its I/Q imbalance will change with frequency within the receiving frequency band,which is called frequency-dependent I/Q imbalance.As the bandwidth increases,the I/Q imbalance will further deteriorate.At this time,maintaining I/Q balance and achieving a reasonable level of image suppression will become more complicated and difficult.As radar transceivers move toward broadband/ultra-wideband,it becomes more and more important to resolve frequency-dependent I/Q imbalances.This thesis introduces good performance compensation technology for I/Q imbalance and completes the design and implementation of the calibration scheme.After completing in-depth theoretical research,this thesis establishes a model for frequency-independent and frequency-dependent I/Q imbalance,designs an estimation compensation scheme for narrow-band frequency-independent I/Q imbalance,and applies this estimation method to a broadband I/Q imbalance calibration technology.This wideband I/Q imbalance compensation method uses the transmitted training signal to estimate the wideband I/Q unbalance error and complete the calibration filter design.In order to solve the problem that this calibration method can't eliminate the amplitude and phase fluctuations in the broadband signal band,this thesis studies a joint calibration method that uses the least squares algorithm to construct a cost function to establish a compensation matrix,complete I/Q imbalance compensation and channel equalization.After an in-depth study of the least squares algorithm,in order to improve the real-time nature of the compensation method,this thesis builds a "blind" adaptive I/Q imbalance compensation filter based on some second-order statistical characteristics of random signals.The adaptive filtering scheme can calibrate the I/Q imbalance of the signal in real time without using the training signal.The performance simulation of wide/narrowband I/Q imbalance compensation methods has been verified.In the simulation software MATLAB,simulations for various operating scenarios such as ultra-wideband linear frequency modulation(LFM)and multi-tone sinusoidal signals are completed.The simulation results show that the image suppression performance of the compensation method is good.Based on the simulation results,the advantages and limitations of each compensation method in practical applications are analyzed.This thesis designs and builds an ultra-wideband zero-IF receiving system and completes the design and implementation of two broadband I/Q imbalance compensation schemes.A high-speed data acquisition scheme is designed for ultra-wideband signals in the 2 ~ 3GHz frequency band,the compensation parameters are extracted,and finally the FIR calibration filter is designed and the RTL implementation of the calibration in the FPGA is completed.According to the final test results,the two wideband I/Q imbalance calibration techniques can achieve 35 dB and 50 dB full-band image suppression,respectively?...