A Vehicle Weigh-in-Motion System Based on Hopfield Neural Network Adaptive Filter

A processing method of vehicle WIM signals is researched in this paper. According to the characteristics of vehicle weigh-in-motion signals, a self-adaptive filtering variable step size LMS algorithm based on neural network is proposed to replace the traditional self-adaptive filtering method. This method can filter out the noise in each band of WIM signal. In different circumstances, for different vehicle types, it has good adaptability, high accuracy and high speed. Based on this method of signal processing, a vehicle WIM system based on Hopfield Neural Network adaptive filter is developed, a high-performance chip of TMS32C2812 is selected to design a high-performance system. This system can measure the weight of vehicle on highway accurately and timely.