Research On Location And Identification Algorithm Of Railway Perimeter Intrusion Based On Mach-zehnder Interferometer

The development of transportation is an important prerequisite for China's national economic and social development,and high-speed railway transportation has become the preferred mode of transportation for the majority of Chinese people due to its strong transportation capacity,safety and reliability.In order to further strengthen the transportation capacity of high-speed railway,high-speed railway transportation has been vigorously constructed and the “medium-term and long-term railway network plan” has been formulated in China.The first principle of high-speed railway operation is safety,not only to ensure the safety of passengers and goods at all times,but also to ensure the safety of infrastructure and train operation along the railway,so it is essential to study the perimeter security technology along the railway.At present,China's railway perimeter intrusion monitoring technologies mainly include electronic fence,microwave electromagnetic sensor,ultrasonic electromagnetic sensing and alarm,vibration cable,infrared electromagnetic countermeasures and railway video surveillance,etc.,but these technologies are vulnerable to the electromagnetic interference of weather,temperature,railway electrical equipment,and can not carry out long-distance positioning and distinguish multiple intrusion modes.Due to its own advantages,optical fiber sensing technology has a place in the field of perimeter security,especially the Mach-Zehnder optical fiber sensing technology is a distributed optical fiber,capable of long-distance transmission,anti-electromagnetic interference,small impact by the natural environment is suitable for the perimeter intrusion monitoring of railway tunnel and field section.In this thesis,the present situation of railway perimeter intrusion monitoring technology at home and abroad is studied.In view of the problems existing in the current monitoring technology,the asymmetric dual Mach-Zehnder interferometric optical fiber sensing technology combined with signal processing technology is used to realize the railway perimeter security protection.The main research contents are as follows:First of all,the fiber optic sensing technology applied to railway perimeter intrusion monitoring and their advantages and disadvantages are introduced.Traditional Mach-Zehnder interferometric optical fiber sensing technology is studied.On this basis,the asymmetric dual Mach-Zehnder interferometric optical fiber sensing technology is used as the detection technology of railway boundary intrusion monitoring system,and the sensing principle and mathematical model of the sensing technology are analyzed.Second,the positioning principle of asymmetric dual Mach-Zehnder interferometer was introduced.And in order to solve the problem that the remote location of perimeter security system based on asymmetric dual Mach-Zehnder interferometer optical fibe sensingtechnology is greatly affected by environmental noise due to the complex external environment along the railway,an improved locating method based on wavelet packet denoising and cross-correlation operation is proposed.On the one hand,the multi-threshold criterion and the new threshold function which can overcome the shortcomings of the traditional hard and soft threshold function are used to improve the wavelet packet denoising method and improve the signal to noise ratio.On the other hand,the time delay estimation based on cross-correlation algorithm is used to locate the intrusion event.Through this method,the high precision of railway perimeter intrusion location can be achieved.Thirdly,in order to solve the problem of high false alarm rate and false alarm rate at long distance in perimeter security system of dual Mach-Zehnder interferometric optical fiber sensing technology,a recognition method based on wavelet packet frequency band energy feature extraction and RBF neural network is proposed.On the one hand,wavelet packet feature extraction is used to extract the percentage of the coefficient energy of the reconstructed wavelet packet in the total energy as the feature vector,so as to increase the differences among various event features.On the other hand,RBF neural network is used as classifier to improve the recognition speed and accuracy of the system.It makes the external intrusion monitoring system along the high-speed railway meet the requirement of high reliability.Finally,the perimeter intrusion monitoring system based on asymmetric dual Mach-Zehnder optical fiber sensor built on the protection network of a section of railway line by Tianjin university is used to carry out experiments and collect information.The experimental results show that the asymmetric dual Mach-Zehnder interferometric optical fiber sensing technology,combined with the positioning algorithm and identification algorithm in this thesis,greatly improves the positioning accuracy and identification accuracy of the system,which fully meets the requirements of high positioning accuracy and low false alarm rate missing alarm rate of the perimeter security system along the railway and has a great development prospect.