A representation for the symbol error rate of arbitrary constellations under AWGN

The symbol error rate of the maximum likelihood detector for an arbitrary multi-dimensional constellation impaired by additive white Gaussian noise is characterized as the product of a completely monotone function with a non-negative power of the signal to noise ratio. Using this general result, it is proved that the symbol error rate is completely monotone if the rank of its constellation matrix is either one or two. Further, it is shown that, if the rank of the constellation matrix is greater than or equal to four, then the symbol error rate cannot be a completely monotone function of the SNR. Applications to stochastic ordering of wireless system performance is also discussed.