Downloadable ContentDownload PDF
A nonconventional digital communication system
A nonconventional digital communication system, called the Bayesian Communication System, is presented. In the Bayesian system, the baseband signal is shifted to a higher frequency for transmission by modulating the frequency of the sine-wave carrier. This modulation is in accordance with the current binary symbol and the two successive binary symbols. Thus, each transmitted pulse represents a sequence of binary symbols, rather than one of two binary symbols. The Bayesian signal requires eight coherent detectors at the receiver. The performance of the Bayesian system is analyzed using the narrowband representation of noise and assuming zero-mean gaussian noise is added to the signal during transmission. Expressions are derived for the probability of error at the output of the coherent detectors. This error rate is present at the input of the Bayesian detector, the main unit in the receiver which determines the output binary sequence. Bayes' Theorem is implemented in the detection process. To this end, expressions are derived for the probability of detection and the probability of false alarm. A computer program is designed which simulates the Bayesian system. The output error rate is determined, and comparisons are drawn between the Bayesian system and some of the conventional binary communication systems. The performance of the Bayesian system surpasses the performance of the noncoherent FSK system. Also, for SNR levels above 2 dB, the Bayesian system performs better than the coherent FSK system.