Research On Dynamic Digital Channelization Receiver And Its Subband Spectrum Detection Technology

With the increasing complexity of the modern battlefield electromagnetic environment,the radar signal system used on the battlefield has become more and more variable.As an important system for wireless signal reception in electronic warfare,the signals received by the receiver often have non-cooperative,a priori unknown information,and the number,bandwidth and position of the sub-band signals contained in the received signal are unknown.In order to be able to dynamically adapt to the large instantaneous bandwidth signals such as LPI in electronic warfare system,dynamic digital channelized receiver came into being.In the dynamic digital channelization receiver,the spectrum of the sub-band signal is correctly detected,the presence or absence of the signal in the sub-channel is judged,and then the sub-channels in which the signal is present are synthesized to reconstruct the corresponding original signal to achieve broadband The extraction and separation of multiple signals in a signal is a key problem to be solved by a dynamic digital channelization receiver,and plays a key role in subsequent signal processing such as signal sorting and direction finding.Therefore,it is of great significance to study dynamic digital channelized receivers and their spectrum detection techniques.This paper focuses on the design of dynamic digital channelized receiver and its subband spectrum detection technology.The main contents and innovative achievements of the research work are as follows:Firstly,the design method of dynamic digital channelized receiver based on whole band decomposition is studied based on the analysis of existing problems in the design method of existing channelized receiver.Its high-efficiency structure is deduced by using semi-band filter,complex exponential filter bank and multiphase decomposition of signal.And the simulation experiment is carried out to verify it.The structure is highly reconfigurable: When the signal changes dynamically,the sub-bands of the signal are judged according to the spectrum detection result,and they are combined to recover the original signal.This method has efficient computational efficiency and can be processed in real time in programmable devices.Secondly,aiming at the problems existing in traditional spectrum detection methods,the dynamic channelization subband spectrum detection method based on eig-envalues is studied.Through the research,it is found that most existing algorithms determine the threshold according to the distribution law of the maximum eigenvalue.The threshold accuracy obtained needs to be further improved,and the influence of the internal relationship between the eigenvalues of the sampling covariance matrix on the spectrum detection results is ignored.That is,only the maximum and minimum eigenvalues are used for spectrum detection,and the remaining eigenvalues are not used,resulting in resource waste.At the same time,the final threshold expression of the difference of eigenvalues methods are related to noise,and the detection results are affected by noise.For the above problems,in this paper,we use the latest research results of random matrix theory to adopt the distribution of the smallest eigenvalues of the more accurate signal sampling covariance matrix,and make full use of all the eigenvalues of the sampling covariance matrix.Using the ratio of two different forms of the average eigenvalue to the minimum eigenvalue as the test decision statistic,deriving and obtaining a better detection threshold expression.Two more accurate improved algorithms for dynamic channelized subband spectrum detection based on eigenvalues are proposed.The detection threshold of the algorithm is independent of noise.It does not require any prior information of known signals and noise to complete the detection,and overcomes the interference of noise variation on performance detection.It is a blind detection algorithm.The effectiveness of the proposed algorithm is verified by simulation experiments and comparative analysis with the existing methods in other literatures.Finally,the compressed sensing theory is introduced into the dynamic channelized subband spectrum detection algorithm based on eigenvalues.A eigenvalue spectrum detection algorithm based on non-reconstructed compressed sensing is proposed.The compressed sensing is used to replace the Nyquist sampling,and the observation matrix satisfying the condition is selected to execute the spectrum detection algorithm flow based on the eigenvalue without losing the main information of the signal.It is concluded that the statistical characteristics of the signal sampling covariance matrix are invariable before and after compression,and the signal can be detected without reconstruction.The algorithm maintains the advantages of the spectrum detection algorithm based on the eigenvalue,simulation experiments and comparative analysis of the computational complexity of the algorithm show that compared with the compressed sensing algorithm based on OMP reconstruction and the detection algorithm without compressed sensing,the performance loss is small,but the computational complexity is greatly reduced,the efficiency of spectrum detection is improved,and the algorithm is better used value.