A real-time demodulation algorithm for FSK signals in railway tracks and its implementation

A new demodulation method is developed for frequency-shift key (FSK) signals in railway tracks. Because of the harsh electrical environment, strong background noises are usually found in the sampled FSK signals. Based upon a zoom fast Fourier transform (ZFFT) algorithm, the segmented data blocks are used in an overlapping way for further processing. Theoretical analysis shows that, by using the algorithm proposed in this paper, the useful information can be successfully snatched in a fast way from the low SN ratio FSK signals. Moreover, a low-cost demodulation system is implement with a DSP chip TMS320F206. Experimental results prove that it can not only meet the requirements of real-time demodulation, but also greatly reduce the probability of incorrect decisions.