A novel algorithm for frequency extraction of ABS signals by using DTDNNs

Abstract

Intelligent transportations system (ITSs) have emerged to increase safety and convenience of people invehicles. In an ITS, communication devices in the vehicle or along the streets send the information gathered from thevehicle to information management centers as well as sending processed information to the vehicle. Furthermore, it isnecessary to locate the exact location of the vehicle on a digital map in order to navigate the vehicle precisely in controland navigation systems. One of the technologies for this purpose is the antilock brake system (ABS), which can avoidaccidents effectively and can also be utilized to determine vehicle speed and location by using its pulses. To do so, thefrequency of ABS pulses should be extracted. In this paper, a novel method for frequency extraction is introduced inwhich one type of neural network, the distributed time-delay neural network (DTDNN), is used. Simulation results showthat the output of the neural network can acceptably follow frequency variations of ABS signals after convergence.