Hardware/Software Design for the Simulation of AWGN Communication Channels

Abstract

This paper presents a hardware/software design to simulate additive white Gaussian noise (AWGN) channels. Such a tool is employed here to evaluate the bit error rate (BER) of a communication system. The implementation of the noise generator is based on the central limit theorem by means of accumulating samples of a uniform random variable obtained using a linear congruential generator. The results show the importance of developing an accurate noise generator in terms of its statistics. In this regard, there is a trade-off between the accuracy of the noise samples and the complexity of the generator. We show how the accuracy of the samples influences on the BER results producing a meaningful error with respect to theoretical computations when the precision of the generator is low.