Investigation Of The Method Of Detection Underwater Target On The Bottom

This thesis revolves to detect the target echo technology to launch the research. Because the reverberation is the main background disturbance for detection the target on the bottom, multianalysis reverberation disturbance and its characteristic, and aimed at the reverberation's the non-stability and strongly colored, proposed one method which do not need the priori knowledge to detect the target echo under the reverberation background -a order best detection method, to partition pre-white copy-correlation detection method.Using GLR (Generalized Likelihood Ratio) to count the statistic distance, namely the generalized likelihood ratio statistics distance, is reasonable partition to the reverberation data, after partitioning, the neighboring two sections of reverberation data may be partial steady. Establishes the AR model to each section of reverberations data, calculates the AR model's the parameter using the Levinson algorithm, calculates the power spectrum of this section of reverberations data, thus structure pre- whited FIR numeral filter, to white the preceding section of reverberations data and transmitting signal white processing, namely estimated the kth section's power spectrum, then according to the kth section's power spectrum structure acts whited filter to whited processe to the k+1 section and the transmitting signal. After whited processing, whited transmitting signal and the whited reverberation data segment make copy-correlation processing, if this section of data has the target echo, then possibly appears the correlation peak. At last count this section of data average power, makes normalized processing to this section of data. Repeat above processing steps, until the last section of reverberations data.On the basis of this method and detection principle, and according to actual situation, joining bandpass filter, beam forming pretreatment method and so on, integrate detection system for detection the target echo on the bottom, and use the software method to realize it.For confirming this method, make lake experiment in October ,04. Then use this...