Bit error probability of M-QAM under η-μ or κ-μ fading and impulsive noise gated by a signal characterized by a Markov process or Poisson process

Abstract

This article presents exact expressions for the bit error probability (BEP) of the M-ary quadrature amplitude modulation (M-QAM) scheme subject to η-μ or κ-μ fading and gated additive white Gaussian noise (GAWGN), which is a noise formed by the sum of a white Gaussian component of variance σg2 and a white Gaussian noise component of variance σi2 gated by a signal C(t). In this work, the signal C(t) is characterized by a Markov process or a Poisson process (telegraphic process) and is used to model the instants of appearance and disappearance of the impulsive noise in the communication system. The expressions presented in this paper are new, simple, written in terms of elementary functions and encompass, as special cases, the BEP expressions of a variety of fading models, such as Nakagami-m, Nakagami-q, Rayleigh, Rice and one side Gaussian. BEP curves are also shown in this work, corroborated by simulations performed with Monte Carlo method, for different values of the signal to impulsive noise ratio, modulation order and parameters that characterize the number of transitions of the signal C(t) from state zero to state one in a symbol interval.