An UGW Rail Break Detection System Based on Multi-Core DSP

Railway transportation is currently an important part of the national economic lifeline in China, and its safety is closely related to the lives, properties of everyone. How to improve the real-time performance, accuracy and stability of rail break detection is important. This work presents a real-time ultrasonic rail break detection system based on multi-core DSP. In response to the problems existing in the current signal processing methods, based on the characteristics of multi-core DSP, a multi-algorithm UGW signal processing mode based on multi-core parallel processing is proposed. To ensure the resolution of the feature signal in long-distance detection, the pulse compression method is selected to process the received signal. Then, considering the influence on rail break detection when a train passes, to further enhance the real-time performance and reliability of the entire system, an energy spectrum analysis algorithm is proposed to process the signal and assist in the rail break detection judgment. Finally, a signal processing mode for parallel processing and analysis of the received signal by the two algorithms is designed for rail break detection. The experimental results show that the three parallel algorithms can effectively and in real time determine which state the track is in: free, occupied or broken.