Research On Algorithm Of Estimating Parameter Based On MWC Compressed Sampling Structure

In the electromagnetic environment under the modern battlefield,there are a large number of radar signals of different types.The electronic reconnaissance receiver can intercept the radar signal and perform other tasks such as type recognition,parameter estimation and so on.At present,in the field of electronic reconnaissance,the mature uniform channelized wideband digital receiver is used widely.However,there are still a series of problems that are difficult to solve in this type of wideband digital receiver.For example,signal cross-channel problems,complex system implementation and large amount of sampled data have different effects on subsequent signal processing.In recent years,compressed-sampling wideband digital receivers combining a modulated wideband converter(MWC)with a wideband digital receiver have received widespread attention.Such wideband digital receivers can realize compressed sampling to reduce the signal sampling rate and reduce the amount of data.In addition,such receivers have fewer processing branches,which facilitates hardware engineering implementation.At the same time,since input signals are all mixed to baseband in the frequency domain,so it can effectively avoid the appearance of signal cross-channel phenomenon.Since the MWC-based signal reconstruction algorithm is complex,the hardware implementation is difficult.And for electronic reconnaissance,its purpose is to obtain the parameters of the intercepted signal.Therefore,it is cumbersome and complicated to recover the original input signal and obtain the signal parameters by compressed sampled signal.So,this paper mainly studies the method of directly estimating the input signal by using compressed sampling data under non-reconstruction conditions.Aiming at the problem that the existing non-reconstructed carrier frequency(CF)estimation algorithm is not applicable to the real signal,a real signal CF estimation algorithm based on the mixing coefficient matrix is ? ? proposed.The algorithm takes advantage of the feature that the multiplication result of any two channels of compressed sampling signals and the inverse mixed weight matrix is the same,so the inverse mixed weight matrix which corresponding the input signal can be searched by traversing.And then,we can obtain the position of the sub-band where the input signal is located,so that the input signal CF estimation can be obtained.In addition,in the face of complex signals,it is proposed toimprove the MWC compression sampling structure,and use the Chinese remainder theorem to estimate the input signal carrier frequency.In the improved structure,there are three sub-prototype structures,and the periods of the mixing functions in different sub-structures are different,so that the respective baseband widths are different.Thus we can obtained the compressed sampling signals of different frequencies.The original frequency of the input signal can be directly calculated according to the different compressed sampled signal frequency values ? ? and the sub-band width values ? ? as inputs to the robust closed-time Chinese remainder theorem.Aiming to the problem that wideband cross-channel LFM signal is mixing to the baseband so that the compressed sampling signal’s spectrum is folded and its modulated type and parameter can not be estimated.A method based on spectral line stitching is used to complete the recognition and parameter estimation of wideband LFM signals.Firstly,the time-frequency curve of the compressed sampled signal is extracted,and the line stitching is performed to complete the recognition of the wideband LFM signal.Secondly,the effective part of the compressed sampled signal is intercepted,and the initial frequency and the frequency-modulated slope of the wideband LFM signal are estimated by the DPT algorithm.Finally,in order to realize the joint estimation of CF and direction-of-arrival(DOA)for the incident signal by using MWC compressed sampling structure,a delay-based MWC compressed sampling uniform linear array is studied.Each array element includes an original compressed sampling processing branch and a delay processing branch,and the delay module in the delay processing branch can perform delay operation on the incident signal.At this time,the array compressed sampling signal conclude information of CF and DOA.The MUSIC algorithm is then used to perform a two-dimensional spectral peak search on the array signals to complete the joint estimation of CF and DOA.In addition,the search range of the incident signal CF can be reduced according to the compressed sampling signal frequency value and the sub-band width,and then reducing the calculation time of the algorithm.