Research On Wideband Acquisition And Processing For Digital Electronic Reconnaissance Receiver

Electronic reconnaissance satellite is an important means of acquiring electronic intelligence,especially in terms of rapid combat decision-making.With the development of modern radar and communication technology,the signal bandwidth is getting larger and larger,which makes the existing space-borne electronic reconnaissance system on satellite face challenges in signal reconnaissance tasks.Most traditional electronic reconnaissance satellites are in the frequency sweep mode,which makes it difficult to capture burst signals in wide band.And the receiver bandwidth is relatively limited,making it difficult to fully obtain the parameters of wideband signals such as wideband radar signal and spread spectrum communication signal.In addition,the ability of the original receiver to find weak and small signals in multi-signal environments is relatively limited.Therefore,in line with the development trend of IF or RF direct digitalization,large bandwidth and real-time processing,this paper makes a deep research on the wideband acquisition and processing technology of electronic reconnaissance receivers.In order to achieve wide-band sampling,this paper studies several high-speed sampling techniques based on parallel ADCs,including time-interleaved analog to digital convertor technology,hybrid filter bank technology and analog channelized synchronous frequency-conversion sampling technology.The characteristics and mismatch factors are analyzed,which is of reference value to the design and implementation of signal sampling link of wideband electronic reconnaissance receiver.In this paper,a dual-channel highspeed parallel sampling method based on spectrum folding is proposed,which solves the problem of insufficient analog bandwidth from the perspective of spectrum folding and recovery.The method has high channel symmetry and the mismatch factors are easy to correct.And it can be combined with other sampling methods such as time-interleaved method to further increase the sampling rate and achieve higher bandwidth and highprecision sampling.In order to solve the problem of real-time processing of high-speed parallel sampling data in wideband electronic reconnaissance digital receiver,considering the difficulties,extensibility,flexibility and system performance,a series of schemes for parallel signal real-time processing are designed,including mix-free wideband digital quadrature demodulation,parallel FIR filter based on data extraction and pipeline operation,narrowband digital receiver based on digital down-conversion,parallel wideband realtime spectrum analysis and so on.The difficulty of high-speed real-time processing of wideband electronic reconnaissance receiver with FPGA as the processing core is solved.The above real-time processing scheme is verified in the implementation of a digital wideband electronic reconnaissance receiver project.Traditional signal sampling methods generate a large amount of redundancy in the data.As one type of the analog-information converters based on the theory of compressed sensing,the Nyquist folding receiver can realize a high probability capture of signals in an ultra-wideband range of the entire radar band by a single-chip ADC,and it makes a good tradeoff between hardware resources and sampling performance.Its subsequent digital signal processing can be analyzed using classical algorithms.Based on the detailed analysis of Nyquist folding receiver,in order to solve the problem of its hybrid modulation signal processing,a dual-channel Nyquist folding receiver structure is proposed.This structure eliminates the modulation of the input signal itself by using the two channels' output,and has lower computational complexity,wider applicability,and higher performance in signal analysis.In order to process the commonly used linear frequency modulation wideband radar signals,several parameter estimation methods for Nyquist folding receiver are proposed in this paper.According to the composite modulation characteristics of the output signal,several parameter estimation methods are proposed,including two instantaneous autocorrelation method with specific intervals,a method based on time-frequency boundary fitting,and a method based on singular value decomposition,and these methods are compared and analyzed by simulation.These methods solve the problem that the existing parameter estimation scheme based on the spectral peak is not suitable for wideband signals,and provide theoretical support for the use of NYFR in the detection and analysis of wideband signals such as radar signal.At the end of the paper,the engineering implementation of wideband electronic reconnaissance receiver is discussed,including a small-scale electronic reconnaissance processing unit and a larger bandwidth terrestrial wideband electronic reconnaissance system,which validate and make full use of the research results of this paper,such as analog channelized synchronous sampling technology and wideband real-time spectrum analysis technology,are fully utilized and validated.In order to improve the wideband reconnaissance capability of spaceborne electronic reconnaissance receiver under limited hardware,a NYFR structure with composite local oscillator and real-number sampling is discussed.This is a highly achievable solution at the cost of adding a certain amount of hardware.