Applying Improved BP Neural Network In Underwater Targets Recognition

Underwater targets recognition has always been a difficulty in anti-submarine warfare. The current recognition methods are either applying underwater signal processing technology or depending on the sonar operator's experience. But due to the complexity of the marine circumstances, it usually fails to meet the need of anti-submarine warfare when using single recognition method. Then we need to combine above two methods effectively to recognize the targets synthetically. The paper analyzes the advantages and disadvantages of two methods, combines underwater signal processing and human experiences to extract the distinguishable characteristics synthetically, then applies the BP neural network for auto-recognition. The paper also advances the improved arithmetic to the BP neural network. Through simulation test with the collected data and comparing with the result of other method of underwater targets recognition, the paper proves the effectiveness of applying improved BP neural network in underwater targets recognition.