| Wideband digital reconnaissance system plays an important role in radar countermeasure, which has the ability of large instantaneous bandwidth and multi signal processing. It needs a high speed data acquisition technology to improve the instantaneous bandwidth, but AD chip's inherent analog bandwidth limits the development of high speed data acquisition technology. To solve these problems, this paper presents a new method of ultra wideband reconnaissance signal processing technology based on the theory of compressed sensing, which can break the Nyquist sampling theorem and reduce the sampling rate of AD conversion.According to the sparsity of radar signal in time, space and frequency domain, the compressed sensing theory is applied to broadband reconnaissance signal processing, in which the AIC(Analog-to-Information Conversion) sampling structure is used to get observation data. With great number of pseudo-random sequence, Gold code can be used to modulate the analog signal, and for simple radar signal the compression ratio can be up to 32. Then, this paper proposes a signal processing method based on the AIC structure, using observation data to estimate the parameters of the original signal instead of reconstructing it, which can improve the acquisition speed of the signal parameters in a large extent. As a result, The frequency range can be detected is:10MHz~16MHz,which covers Radar operating frequency range basically. On the basis, the detection method of the radar pulse signal is studied, and the effect of the interference frequency band is eliminated by introducing a new algorithm based on the discrete ellipsoid. In the end, the paper analyzes the influence of different factors on the extraction of signal parameters, which include observation matrix, signal density and signal dynamic range, signal to noise ratio and compression ratio. Thus the general application conditions of the ultra wideband signal processing method based on compressive sensing is obtained. |
PDF Full Text Request