Research On USBL Positioning System Based On Neural Network Optimization

Underwater acoustic location technology is widely used in military and civil fields.Ultra-Short Baseline(USBL)positioning system is widely used in underwater positioning due to its small size and easy installation.However,the traditional USBL underwater positioning system still has some problems that have not been solved,such as poor anti-noise and anti-multipath performance of the positioning signal waveform,phase ambiguity of the positioning algorithm and other problems,which inevitably reduce the accuracy of positioning.Therefore,the USBL underwater positioning system needs further research.This paper studies the two problems existing in the USBL underwater positioning system,and the main related work is as follows:(1)The non-linear frequency modulation(NLFM)wave signal is introduced into the USBL underwater positioning system to solve the problems of the traditional USBL underwater positioning system positioning signal with high first side lobe level and poor anti-noise performance.Experimental simulation It is verified that the designed NLFM signal has lower sidelobes and better noise immunity,and makes the USBL system have higher range resolution.(2)An underwater Time Difference of Arrival(TDOA)estimation model based on deep learning is studied,which incorporates Time Convolutional Neural Network(TCNN)with the Long Short-term Memory(LSTM)module,the previous TDOA estimates can be used to predict the current and future TDOA values,which improves the system’s estimation accuracy and robustness of TDOA.The comparative simulation verifies the validity of the model.(3)A time-difference underwater localization algorithm based on a five-element uniform circular hydrophone array is used to solve the problem of low positioning accuracy due to phase ambiguity in the traditional USBL positioning system.This algorithm first obtains the TDOA when the acoustic signal arrives at different elements,uses TDOA to calculate the two-dimensional coordinates of the sound source,and then uses the Majorization-Minimization Optimization(MM)algorithm to calculate the results.Optimize to reduce the error.and finally use the angle-of-arrival(AOA)algorithm to calculate the third-dimensional coordinates of the sound source,that is,the depth of the sound source,which improves the positioning accuracy of the system.(4)The designed signal and positioning algorithm are applied to the USBL positioning system,and the effectiveness and robustness of the signal and the proposed algorithm are verified through actual underwater experiments.