Research On Implementation Method Of Wideband Array Radar Channel Equalization Technology

With the increasing application of broadband array radars,the inconsistency of amplitude-frequency and phase-frequency characteristics between various array channels must be considered when processing broadband array signals,that is,channel mismatch.The existence of channel mismatch will have a serious impact on the subsequent signal processing algorithm of the array radar,reducing the angle and ranging accuracy,resolution and anti-interference ability of the array radar,and then affecting the overall performance of the wideband array radar.Therefore,it is extremely important to correct the channel mismatch in the wideband array radar.The research content of this article is as follows:First,based on the analysis of the array model,the modeling process of several channel mismatch models is discussed.Through simulation verification,the advantages and disadvantages of several mismatch models are summarized,and the sinusoidal disturbance model and the FIR filter coefficient disturbance model are selected.As the main model used in this article,and based on the FIR filter coefficient perturbation model,the performance changes of the beamforming algorithm and array pulse compression results when there is a channel mismatch are simulated and analyzed.The simulation results show that the channel mismatch will have a serious impact on the array performance.Then,the wideband time domain and frequency domain algorithms are studied based on the sinusoidal disturbance model.In this paper,a comprehensive simulation analysis and comparison of time and frequency domain equalization algorithms are carried out,and several indicators for evaluating equalization performance are analyzed.Aiming at the problems that the equalization bandwidth of the frequency domain equalization algorithm is greater than the effective bandwidth will cause the equalization accuracy to decrease and the amount of calculation to increase,this article starts from the perspective of controlling the effective bandwidth of the equalization,and revises the traditional frequency domain equalization algorithm.The core idea of the correction is to select the effective equalization bandwidth by controlling the number of frequency points after the Fourier transform.The simulation results show that the residual amplitude mismatch and residual phase mismatch of the revised algorithm are closer to 0d B,the number of rows of the frequency factor matrix is changed from the original M to current M’,the equalization accuracy is obviously improved,and the computational complexity of the algorithm is also smaller.Finally,on a hardware platform of a wideband array radar,a wideband frequency domain equalization based on the FPGA+DSP combined architecture is designed and implemented.In the frequency domain equalization system designed in this paper: FPGA mainly completes the FFT transformation of input data and the FIR filter output of channel data.For the actual problem that the number of channels in the project is large,and the FPGA resources are limited,the channel data cannot be processed in parallel on the FPGA,This article designs a partially parallel FPGA data processing scheme,which enables the system to take into account the contradiction between time and resources;DSP mainly completes the functions of vector division,matrix inversion,and equalization coefficient calculation.In addition,this article discusses in detail the design and implementation process of SRIO-based real-time data communication between DSP and FPGA.At the end of the article,the overall functions and various modules of the system are verified,and the project implementation results are compared with the MATLAB simulation results,which proves the reliability and practicability of the scheme designed in this article.