Research On Filtering Algorithms Of LFM Signals Based On The Frwt

The research on the filtering of LFM in ocean environment has a practical significance and is one of the hot spots in underwater acoustic signal processing. This paper focuses on the filtering of LFM in ocean environment. Based on the present denoising methods like wavelet transform and fractional fourier transform, I use fractional wavelet transform to denoise the LFM signal. After the algorithm implementation, this paper uses the simulation signals and measured signals to dispose LFM signals in noise and reverberation environment. Through the result of disposal and compare of three methods, the paper validates the feasibility and validity of fractional wavelet transform which has factual value. This paper is deployed from the following parts:1?Introducing the physics meaning, basic character and algorithm implementation in filtering processing of wavelet transform and fractional fourier transform. Researching the definition, character and viable algorithm of fractional wavelet transform. Validating the feasibility and validity of fractional wavelet transform used in filtering LFM signals from theory analyse and algorithm implementation.2?Realizing the signal filtering algorithm of wavelet transform, fractional fourier transform and fractional wavelet transform by matlab which can be used to denoise in ocean environment.3?Dividing LFM signals into single component, multi component, multi geometric highlight and multi period and discussing the formation mechanism and characteristic.Realizing the simulation of LFM signals in ocean environment and reverberation background.To research filtering of LFM signal in reverberation background, Realizing the simulation of reverberation based on Middleton statistical model. The simulation of signals establishes the foundation of algorithm verification and efficiency analyse.4?In the situation of theory analysis and model simulation, the LFM signals of single,multi-component, multi- period and multi-time lapse are disposed based on WT, FRFT and FRWT, which shows the exactitude of algorithms and takes out a satisfying result.5?Kinds of LFM signals and real signals are analysed to take out the noise by all the three filtering algorithms. Result analysed shows that the performance of FRFT and FRWT methods are better than WT to take out the noise. However, compare FRFT with FRWT, it is possible for either of them to be the good one. The reason for this may be the limit of wavelet transform. With the deeper research of theory and practice and further improvement of FRWT algorithm,it will stand out the advantage of FRWT,which would gain a better result to filter the noise of LFM signals. However, in this paper, though the advantage of FRWT is not obvious,it has given a good result to filter LFM signals,which further means that FRWT is feasible and potential for filtering LFM signals and has a expansive potential to apply. All these could establish the foundation for works latter.