Adaptive active sonar signal processing for optimal LFM echo detection of slowly moving targets

Abstract

To detect underwater moving targets effectively in the active sonar systems, the linear frequency modulation signal (LFM) is commonly used as a broadband signal for achieving more processing gain. However, the conventional method that uses the matched filter for detecting the LFM signal cannot remove these undesirable signals easily since its reverberation spectrum structure has a certain similarity to the LFM signal spectrum itself. To improve the ability of detecting the LFM signal in the shallow water region where the influence of noise and reverberation are large, this paper proposes a new signal detection method based on the spectrum characteristics of the LFM matched spectrum by using an adaptive technology that can effectively eliminate only the undesirable signals. Firstly, a detailed theoretical derivation is given in this paper. It shows that the spectrum of the LFM signal matched processing in frequency domain can be represented by cosine sequence. Simulation proof by means of spectrum analysis is used to verify the single frequency characteristics of the FM signal matched filter output, which shows that the spectrum peak has a corresponding relation to the target echo's position. Therefore the problem of detecting the LFM signal in reverberation can be converted into detecting single frequency signals in noise, while the adaptive technology is a powerful tool for single frequency signal detection in noise. Based on the considerations above, the adaptive technology is developed in this paper to detect the LFM echo of the underwater slowly moving target. Because the noise is not related to the measured signal, the adaptive technology can achieve inhibition of attenuation interference or noise as much as possible through adaptively adjusting the transmission characteristics of the filter, so as to improve signal-to-noise ratio(SNR). By applying the proposed method to some actual experimental data, the new method in this paper has high detection performance as compared to conventional methods. The effectiveness of the algorithm is verified.